logs/exp_dc_f32c32_FM.log

nohup: ignoring input
The following values were not passed to `accelerate launch` and had defaults used instead:
	`--num_processes` was set to a value of `8`
		More than one GPU was found, enabling multi-GPU training.
		If this was unintended please pass in `--num_processes=1`.
	`--num_machines` was set to a value of `1`
	`--mixed_precision` was set to a value of `'no'`
	`--dynamo_backend` was set to a value of `'no'`
To avoid this warning pass in values for each of the problematic parameters or run `accelerate config`.
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/__init__.py:1617: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.)
  _C._set_float32_matmul_precision(precision)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/__init__.py:1617: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.)
  _C._set_float32_matmul_precision(precision)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/__init__.py:1617: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.)
  _C._set_float32_matmul_precision(precision)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/__init__.py:1617: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.)
  _C._set_float32_matmul_precision(precision)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/__init__.py:1617: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.)
  _C._set_float32_matmul_precision(precision)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/__init__.py:1617: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.)
  _C._set_float32_matmul_precision(precision)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/__init__.py:1617: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.)
  _C._set_float32_matmul_precision(precision)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/__init__.py:1617: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.)
  _C._set_float32_matmul_precision(precision)
Setting up [LPIPS] perceptual loss: trunk [alex], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [alex], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [alex], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [alex], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [alex], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [alex], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [alex], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [alex], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/alex.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/alex.pth
Using cached /workspace/DC_SSDAE/runs/cache
Using cached /workspace/DC_SSDAE/runs/cache
Using cache found in /workspace/DC_SSDAE/runs/cache/facebookresearch_dino_main
Using cache found in /workspace/DC_SSDAE/runs/cache/facebookresearch_dino_main
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/alex.pth
Using cached /workspace/DC_SSDAE/runs/cache
Using cache found in /workspace/DC_SSDAE/runs/cache/facebookresearch_dino_main
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/alex.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/alex.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/alex.pth
Using cached /workspace/DC_SSDAE/runs/cache
Using cached /workspace/DC_SSDAE/runs/cache
Using cache found in /workspace/DC_SSDAE/runs/cache/facebookresearch_dino_main
Using cache found in /workspace/DC_SSDAE/runs/cache/facebookresearch_dino_main
Using cached /workspace/DC_SSDAE/runs/cache
Using cache found in /workspace/DC_SSDAE/runs/cache/facebookresearch_dino_main
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/alex.pth
Using cached /workspace/DC_SSDAE/runs/cache
Using cache found in /workspace/DC_SSDAE/runs/cache/facebookresearch_dino_main
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/alex.pth
Using cached /workspace/DC_SSDAE/runs/cache
Using cache found in /workspace/DC_SSDAE/runs/cache/facebookresearch_dino_main
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/nn/utils/weight_norm.py:144: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
  WeightNorm.apply(module, name, dim)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/nn/utils/weight_norm.py:144: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
  WeightNorm.apply(module, name, dim)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/nn/utils/weight_norm.py:144: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
  WeightNorm.apply(module, name, dim)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/nn/utils/weight_norm.py:144: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
  WeightNorm.apply(module, name, dim)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/nn/utils/weight_norm.py:144: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
  WeightNorm.apply(module, name, dim)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/nn/utils/weight_norm.py:144: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
  WeightNorm.apply(module, name, dim)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/nn/utils/weight_norm.py:144: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
  WeightNorm.apply(module, name, dim)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/nn/utils/weight_norm.py:144: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
  WeightNorm.apply(module, name, dim)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/peft/tuners/tuners_utils.py:196: UserWarning: Already found a `peft_config` attribute in the model. This will lead to having multiple adapters in the model. Make sure to know what you are doing!
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/peft/tuners/tuners_utils.py:196: UserWarning: Already found a `peft_config` attribute in the model. This will lead to having multiple adapters in the model. Make sure to know what you are doing!
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/peft/tuners/tuners_utils.py:196: UserWarning: Already found a `peft_config` attribute in the model. This will lead to having multiple adapters in the model. Make sure to know what you are doing!
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/peft/tuners/tuners_utils.py:196: UserWarning: Already found a `peft_config` attribute in the model. This will lead to having multiple adapters in the model. Make sure to know what you are doing!
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/peft/tuners/tuners_utils.py:196: UserWarning: Already found a `peft_config` attribute in the model. This will lead to having multiple adapters in the model. Make sure to know what you are doing!
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/peft/tuners/tuners_utils.py:196: UserWarning: Already found a `peft_config` attribute in the model. This will lead to having multiple adapters in the model. Make sure to know what you are doing!
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/peft/tuners/tuners_utils.py:196: UserWarning: Already found a `peft_config` attribute in the model. This will lead to having multiple adapters in the model. Make sure to know what you are doing!
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/peft/tuners/tuners_utils.py:196: UserWarning: Already found a `peft_config` attribute in the model. This will lead to having multiple adapters in the model. Make sure to know what you are doing!
  warnings.warn(
File already exists: /workspace/DC_SSDAE/runs/cache/weights-inception-2015-12-05-6726825d.pth
File already exists: /workspace/DC_SSDAE/runs/cache/weights-inception-2015-12-05-6726825d.pth
File already exists: /workspace/DC_SSDAE/runs/cache/weights-inception-2015-12-05-6726825d.pth
File already exists: /workspace/DC_SSDAE/runs/cache/weights-inception-2015-12-05-6726825d.pth
File already exists: /workspace/DC_SSDAE/runs/cache/weights-inception-2015-12-05-6726825d.pth
File already exists: /workspace/DC_SSDAE/runs/cache/weights-inception-2015-12-05-6726825d.pth
File already exists: /workspace/DC_SSDAE/runs/cache/weights-inception-2015-12-05-6726825d.pth
File already exists: /workspace/DC_SSDAE/runs/cache/weights-inception-2015-12-05-6726825d.pth
[2025-10-25 04:11:21,158][main][INFO] - Will write tensorboard logs inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/tensorboard_logs[2025-10-25 04:11:21,179][main][INFO] - Runtime at /workspace/DC_SSDAE[2025-10-25 04:11:21,180][main][INFO] - Running inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM[2025-10-25 04:11:21,181][main][INFO] - Running args: ['main.py', 'run_name=train_enc_dc_f32c32_FM', 'dataset.im_size=128', 'dataset.aug_scale=2', 'training.epochs=20', 'dc_ssdae.encoder_train=true'][2025-10-25 04:11:21,182][main][INFO] - Command: 'main.py' 'run_name=train_enc_dc_f32c32_FM' 'dataset.im_size=128' 'dataset.aug_scale=2' 'training.epochs=20' 'dc_ssdae.encoder_train=true'[2025-10-25 04:11:21,182][main][INFO] - Accelerator with 8 processes, running on cuda:0[2025-10-25 04:11:21,186][main][INFO] - Hydra configuration:
seed: 0
task: train
runtime_path: ${hydra:runtime.cwd}
ckpt_dir: ${runtime_path}/runs
run_name: train_enc_dc_f32c32_FM
cache_dir: ${ckpt_dir}/cache
run_dir: ${ckpt_dir}/jobs/${run_name}
checkpoint_path: ${run_dir}/checkpoints
dataset:
  imagenet_root: imagenet_data
  im_size: 128
  batch_size: 192
  aug_scale: 2
  limit: null
distill_teacher: false
dc_ssdae:
  compile: false
  checkpoint: null
  encoder: f32c32
  encoder_checkpoint: null
  encoder_train: true
  decoder: S
  trainer_type: FM
  encoder_type: dc
  sampler:
    steps: 10
  ema:
    decay: 0.999
    start_iter: 50000
aux_losses:
  compile: ${dc_ssdae.compile}
  repa:
    i_extract: 4
    n_layers: 2
  lpips: true
training:
  sdpa_kernel: 2
  mixed_precision: bf16
  grad_accumulate: 1
  grad_clip: 0.1
  epochs: 20
  eval_freq: 1
  save_on_best: FID
  log_freq: 100
  lr: 0.0003
  weight_decay: 0.001
losses:
  diffusion: 1
  repa: 0.25
  lpips: 0.5
  kl: 1.0e-06
show_samples: 8


[2025-10-25 04:11:35,084][main][INFO] - Loaded ImageNet dataset: {'train': Dataset ImageNet
    Number of datapoints: 1279867
    Root location: ../../../imagenet_data
    Split: train
    StandardTransform
Transform: Compose(
                 RandomResize(min_size=128, max_size=256, interpolation=InterpolationMode.LANCZOS, antialias=True)
                 RandomCrop(size=(128, 128), pad_if_needed=False, fill=0, padding_mode=constant)
                 RandomHorizontalFlip(p=0.5)
                 ToImage()
                 ToDtype(scale=True)
                 Normalize(mean=[0.5], std=[0.5], inplace=False)
           ), 'test': Dataset ImageNet
    Number of datapoints: 49950
    Root location: ../../../imagenet_data
    Split: validation
    StandardTransform
Transform: Compose(
                 Resize(size=[128], interpolation=InterpolationMode.BILINEAR, antialias=True)
                 CenterCrop(size=(128, 128))
                 ToImage()
                 ToDtype(scale=True)
                 Normalize(mean=[0.5], std=[0.5], inplace=False)
           )}[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_mean. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_var. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.num_batches_tracked. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_mean. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_var. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.num_batches_tracked. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_mean. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_var. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.num_batches_tracked. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_mean. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_var. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.num_batches_tracked. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_mean. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_var. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.num_batches_tracked. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_mean. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_var. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.num_batches_tracked. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_mean. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_var. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.num_batches_tracked. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_mean. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.running_var. Will default to 'ingore'.
[WARNING] Model buffers behavior should be defined using the '_ema' parameter. No _ema key for the buffer decoder.batch_norm_z.num_batches_tracked. Will default to 'ingore'.
Setting up [LPIPS] perceptual loss: trunk [vgg], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [vgg], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [vgg], v[0.1], spatial [off]
Setting up [LPIPS] perceptual loss: trunk [vgg], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [vgg], v[0.1], spatial [off]
Setting up [LPIPS] perceptual loss: trunk [vgg], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [vgg], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Setting up [LPIPS] perceptual loss: trunk [vgg], v[0.1], spatial [off]
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/vgg.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/vgg.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/vgg.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/vgg.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/vgg.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/vgg.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/vgg.pth
Loading model from: /workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/lpips/weights/v0.1/vgg.pth
[2025-10-25 04:11:47,948][main][INFO] - ae parameters count:[2025-10-25 04:11:47,953][main][INFO] - Total: #230.9M   (trainable: #230.9M)[2025-10-25 04:11:47,954][main][INFO] - - encoder: #217.4M   (trainable: #217.4M)[2025-10-25 04:11:47,955][main][INFO] -      - project_in: #1.8K   (trainable: #1.8K)[2025-10-25 04:11:47,956][main][INFO] -      - stages: #216.9M   (trainable: #216.9M)[2025-10-25 04:11:47,956][main][INFO] -      - project_out: #576.1K   (trainable: #576.1K)[2025-10-25 04:11:47,958][main][INFO] - - decoder: #13.5M   (trainable: #13.5M)[2025-10-25 04:11:47,958][main][INFO] -      - conv_in_img: #896   (trainable: #896)[2025-10-25 04:11:47,959][main][INFO] -      - conv_in_z: #9.0K   (trainable: #9.0K)[2025-10-25 04:11:47,959][main][INFO] -      - conv_in: #36.1K   (trainable: #36.1K)[2025-10-25 04:11:47,959][main][INFO] -      - batch_norm_z: #64   (trainable: #64)[2025-10-25 04:11:47,960][main][INFO] -      - time_proj: #0   (trainable: #0)[2025-10-25 04:11:47,960][main][INFO] -      - time_embedding: #80.5K   (trainable: #80.5K)[2025-10-25 04:11:47,960][main][INFO] -      - ada_ctx_proj: #54.1K   (trainable: #54.1K)[2025-10-25 04:11:47,961][main][INFO] -      - down_blocks: #3.0M   (trainable: #3.0M)[2025-10-25 04:11:47,962][main][INFO] -      - mid_block: #3.4M   (trainable: #3.4M)[2025-10-25 04:11:47,963][main][INFO] -      - up_blocks: #6.9M   (trainable: #6.9M)[2025-10-25 04:11:47,963][main][INFO] -      - conv_norm_out: #128   (trainable: #128)[2025-10-25 04:11:47,964][main][INFO] -      - conv_out_act: #0   (trainable: #0)[2025-10-25 04:11:47,964][main][INFO] -      - conv_out: #1.7K   (trainable: #1.7K)[2025-10-25 04:11:47,969][main][INFO] - ae: EMAWrapper(
  (model): DistributedDataParallel(
    (module): DC_SSDAE(
      (encoder): DCEncoder(
        (project_in): ConvPixelUnshuffleDownSampleLayer(
          (conv): ConvLayer(
            (conv): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          )
        )
        (stages): ModuleList(
          (0): OpSequential(
            (op_list): ModuleList()
          )
          (1): OpSequential(
            (op_list): ModuleList(
              (0-4): 5 x ResidualBlock(
                (main): ResBlock(
                  (conv1): ConvLayer(
                    (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                    (act): SiLU()
                  )
                  (conv2): ConvLayer(
                    (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
                  )
                )
                (shortcut): IdentityLayer()
              )
              (5): ResidualBlock(
                (main): ConvPixelUnshuffleDownSampleLayer(
                  (conv): ConvLayer(
                    (conv): Conv2d(256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                  )
                )
                (shortcut): PixelUnshuffleChannelAveragingDownSampleLayer()
              )
            )
          )
          (2): OpSequential(
            (op_list): ModuleList(
              (0-9): 10 x ResidualBlock(
                (main): ResBlock(
                  (conv1): ConvLayer(
                    (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                    (act): SiLU()
                  )
                  (conv2): ConvLayer(
                    (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
                  )
                )
                (shortcut): IdentityLayer()
              )
              (10): ResidualBlock(
                (main): ConvPixelUnshuffleDownSampleLayer(
                  (conv): ConvLayer(
                    (conv): Conv2d(512, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                  )
                )
                (shortcut): PixelUnshuffleChannelAveragingDownSampleLayer()
              )
            )
          )
          (3): OpSequential(
            (op_list): ModuleList(
              (0-3): 4 x ResidualBlock(
                (main): ResBlock(
                  (conv1): ConvLayer(
                    (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                    (act): SiLU()
                  )
                  (conv2): ConvLayer(
                    (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
                  )
                )
                (shortcut): IdentityLayer()
              )
              (4): ResidualBlock(
                (main): ConvPixelUnshuffleDownSampleLayer(
                  (conv): ConvLayer(
                    (conv): Conv2d(512, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                  )
                )
                (shortcut): PixelUnshuffleChannelAveragingDownSampleLayer()
              )
            )
          )
          (4): OpSequential(
            (op_list): ModuleList(
              (0-3): 4 x ResidualBlock(
                (main): ResBlock(
                  (conv1): ConvLayer(
                    (conv): Conv2d(1024, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                    (act): SiLU()
                  )
                  (conv2): ConvLayer(
                    (conv): Conv2d(1024, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
                  )
                )
                (shortcut): IdentityLayer()
              )
              (4): ResidualBlock(
                (main): ConvPixelUnshuffleDownSampleLayer(
                  (conv): ConvLayer(
                    (conv): Conv2d(1024, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                  )
                )
                (shortcut): PixelUnshuffleChannelAveragingDownSampleLayer()
              )
            )
          )
          (5): OpSequential(
            (op_list): ModuleList(
              (0-3): 4 x ResidualBlock(
                (main): ResBlock(
                  (conv1): ConvLayer(
                    (conv): Conv2d(1024, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                    (act): SiLU()
                  )
                  (conv2): ConvLayer(
                    (conv): Conv2d(1024, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
                  )
                )
                (shortcut): IdentityLayer()
              )
            )
          )
        )
        (project_out): OpSequential(
          (op_list): ModuleList(
            (0): ConvLayer(
              (conv): Conv2d(1024, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
            )
          )
        )
      )
      (decoder): UViTDecoder(
        (conv_in_img): Conv2d(3, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        (conv_in_z): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        (conv_in): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        (batch_norm_z): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (time_proj): Timesteps()
        (time_embedding): TimestepEmbedding(
          (linear_1): Linear(in_features=64, out_features=256, bias=True)
          (act): SiLU()
          (linear_2): Linear(in_features=256, out_features=256, bias=True)
        )
        (ada_ctx_proj): Sequential(
          (0): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): SiLU()
          (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
        (down_blocks): ModuleList(
          (0): DownBlock2D(
            (resnets): ModuleList(
              (0-1): 2 x ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 128, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=128, bias=True)
                (norm2): GroupNorm(32, 64, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
              )
            )
            (downsamplers): ModuleList(
              (0): Downsample2D(
                (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
              )
            )
          )
          (1): DownBlock2D(
            (resnets): ModuleList(
              (0): ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 128, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(64, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=192, bias=True)
                (norm2): GroupNorm(32, 96, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(64, 96, kernel_size=(1, 1), stride=(1, 1))
              )
              (1): ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 192, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=192, bias=True)
                (norm2): GroupNorm(32, 96, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
              )
            )
            (downsamplers): ModuleList(
              (0): Downsample2D(
                (conv): Conv2d(96, 96, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
              )
            )
          )
          (2): DownBlock2D(
            (resnets): ModuleList(
              (0): ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 192, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(96, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=320, bias=True)
                (norm2): GroupNorm(32, 160, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(96, 160, kernel_size=(1, 1), stride=(1, 1))
              )
              (1): ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 320, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=320, bias=True)
                (norm2): GroupNorm(32, 160, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
              )
            )
            (downsamplers): ModuleList(
              (0): Downsample2D(
                (conv): Conv2d(160, 160, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
              )
            )
          )
          (3): DownBlock2D(
            (resnets): ModuleList(
              (0-1): 2 x ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 320, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=320, bias=True)
                (norm2): GroupNorm(32, 160, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
              )
            )
          )
        )
        (mid_block): UViTMiddleTransformer(
          (proj_in): Linear(in_features=160, out_features=160, bias=True)
          (transformer_blocks): ModuleList(
            (0-7): 8 x TransformerBlock(
              (norm1): AdaLayerNorm(
                (silu): SiLU()
                (linear): Linear(in_features=64, out_features=320, bias=True)
                (norm): LayerNorm((160,), eps=1e-05, elementwise_affine=False)
              )
              (attn1): Attention(
                (to_q): Linear(in_features=160, out_features=160, bias=False)
                (to_k): Linear(in_features=160, out_features=160, bias=False)
                (to_v): Linear(in_features=160, out_features=160, bias=False)
                (out_proj): Linear(in_features=160, out_features=160, bias=True)
                (out_drop): Dropout(p=0.0, inplace=False)
              )
              (norm2): LayerNorm((160,), eps=1e-05, elementwise_affine=True)
              (ff): FeedForward(
                (proj_in_act): GEGLU(
                  (proj): Linear(in_features=160, out_features=1280, bias=True)
                )
                (drop): Dropout(p=0.0, inplace=False)
                (proj_out): Linear(in_features=640, out_features=160, bias=True)
              )
              (relative_position_bias): RelativePositionBias()
            )
          )
          (proj_out): Linear(in_features=160, out_features=160, bias=True)
          (norm): GroupNorm(32, 160, eps=1e-06, affine=True)
        )
        (up_blocks): ModuleList(
          (0): UpBlock2D(
            (resnets): ModuleList(
              (0-2): 3 x ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 640, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(320, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=320, bias=True)
                (norm2): GroupNorm(32, 160, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(320, 160, kernel_size=(1, 1), stride=(1, 1))
              )
            )
            (upsamplers): ModuleList(
              (0): Upsample2D(
                (conv): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
              )
            )
          )
          (1): UpBlock2D(
            (resnets): ModuleList(
              (0-1): 2 x ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 640, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(320, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=320, bias=True)
                (norm2): GroupNorm(32, 160, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(320, 160, kernel_size=(1, 1), stride=(1, 1))
              )
              (2): ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 512, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(256, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=320, bias=True)
                (norm2): GroupNorm(32, 160, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(256, 160, kernel_size=(1, 1), stride=(1, 1))
              )
            )
            (upsamplers): ModuleList(
              (0): Upsample2D(
                (conv): Conv2d(160, 160, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
              )
            )
          )
          (2): UpBlock2D(
            (resnets): ModuleList(
              (0): ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 512, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(256, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=192, bias=True)
                (norm2): GroupNorm(32, 96, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(256, 96, kernel_size=(1, 1), stride=(1, 1))
              )
              (1): ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 384, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(192, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=192, bias=True)
                (norm2): GroupNorm(32, 96, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1))
              )
              (2): ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 320, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(160, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=192, bias=True)
                (norm2): GroupNorm(32, 96, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(160, 96, kernel_size=(1, 1), stride=(1, 1))
              )
            )
            (upsamplers): ModuleList(
              (0): Upsample2D(
                (conv): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
              )
            )
          )
          (3): UpBlock2D(
            (resnets): ModuleList(
              (0): ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 320, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(160, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=128, bias=True)
                (norm2): GroupNorm(32, 64, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(160, 64, kernel_size=(1, 1), stride=(1, 1))
              )
              (1-2): 2 x ResnetBlock2D(
                (norm1): AdaGroupNorm2D(
                  (ctx_proj): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1))
                )
                (conv1): Conv2d(128, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (time_emb_proj): Linear(in_features=256, out_features=128, bias=True)
                (norm2): GroupNorm(32, 64, eps=1e-05, affine=True)
                (dropout): Dropout(p=0.0, inplace=False)
                (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
                (nonlinearity): SiLU()
                (conv_shortcut): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1))
              )
            )
          )
        )
        (conv_norm_out): GroupNorm(32, 64, eps=1e-05, affine=True)
        (conv_out_act): SiLU()
        (conv_out): Conv2d(64, 3, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      )
    )
  )
  (ema): EMA(ema_model=DC_SSDAE, decay=0.999, start_iter=50000)
)[2025-10-25 04:11:47,970][main][INFO] - aux_losses parameters count:[2025-10-25 04:11:47,971][main][INFO] - Total: #96.7M   (trainable: #145.9K)[2025-10-25 04:11:47,972][main][INFO] - - repa_loss: #82.7M   (trainable: #145.9K)[2025-10-25 04:11:47,972][main][INFO] - - lpips_loss: #14.0M   (trainable: #0)[2025-10-25 04:11:47,973][main][INFO] - aux_losses: DistributedDataParallel(
  (module): SSDDLosses(
    (repa_loss): REPALoss(
      (features_extractor): Frozen(DinoEncoder/Dinov2Model)
      (repa_mlp): Sequential(
        (0): Linear(in_features=160, out_features=160, bias=True)
        (1): SiLU()
        (2): Linear(in_features=160, out_features=768, bias=True)
      )
      (repa_loss): CosineSimilarity()
    )
    (lpips_loss): Frozen(LPIPS)
  )
)[2025-10-25 04:11:47,978][main][INFO] - Optimizer for autoencoder: RAdamScheduleFree (
Parameter Group 0
    betas: (0.9, 0.999)
    eps: 1e-08
    foreach: True
    k: 0
    lr: 0.0003
    lr_max: -1.0
    r: 0.0
    scheduled_lr: 0.0
    silent_sgd_phase: True
    train_mode: False
    weight_decay: 0.001
    weight_lr_power: 2.0
    weight_sum: 0.0

Parameter Group 1
    betas: (0.9, 0.999)
    eps: 1e-08
    foreach: True
    k: 0
    lr: 0.0003
    lr_max: -1.0
    r: 0.0
    scheduled_lr: 0.0
    silent_sgd_phase: True
    train_mode: False
    weight_decay: 0.0
    weight_lr_power: 2.0
    weight_sum: 0.0
)[2025-10-25 04:11:47,983][main][INFO] - No training state found to resume from None[2025-10-25 04:11:47,984][main][INFO] - ====================== RUNNING TASK train[2025-10-25 04:11:47,984][main][INFO] - Starting training[2025-10-25 04:11:47,984][main][INFO] - Batch size of 192 (24 per GPU, 1 acumulation step(s) 8 process(es))[2025-10-25 04:11:47,993][main][INFO] - ---

[2025-10-25 04:11:47,993][main][INFO] - [T_total=00:00:26 | T_train=00:00:00] Start epoch 0[T_total=00:00:36 | T_train=00:00:09 | T_epoch=00:00:09] Epoch 0, batch 1 / 6666 (step 0) loss=7.98227e+07 (avg=7.982e+07) [[all losses: diffusion=1.35239 ; kl=7.98227e+13 ; lpips=0.750356 ; repa=0.995288]]
[T_total=00:03:20 | T_train=00:02:53 | T_epoch=00:02:53] Epoch 0, batch 101 / 6666 (step 100) loss=4.42995e+06 (avg=4.43e+06) [[all losses: diffusion=1.1898 ; kl=4.42995e+12 ; lpips=0.700778 ; repa=0.953524 ; sum_loss=4.42995e+06]]
[T_total=00:06:04 | T_train=00:05:37 | T_epoch=00:05:37] Epoch 0, batch 201 / 6666 (step 200) loss=2.226e+06 (avg=2.226e+06) [[all losses: diffusion=0.747228 ; kl=2.22599e+12 ; lpips=0.637647 ; repa=0.927318 ; sum_loss=2.226e+06]]
[T_total=00:08:48 | T_train=00:08:21 | T_epoch=00:08:21] Epoch 0, batch 301 / 6666 (step 300) loss=1.48646e+06 (avg=1.486e+06) [[all losses: diffusion=0.553895 ; kl=1.48646e+12 ; lpips=0.587502 ; repa=0.901102 ; sum_loss=1.48646e+06]]
[T_total=00:11:32 | T_train=00:11:05 | T_epoch=00:11:05] Epoch 0, batch 401 / 6666 (step 400) loss=1.11577e+06 (avg=1.116e+06) [[all losses: diffusion=0.451203 ; kl=1.11577e+12 ; lpips=0.55372 ; repa=0.877731 ; sum_loss=1.11577e+06]]
[T_total=00:14:16 | T_train=00:13:49 | T_epoch=00:13:49] Epoch 0, batch 501 / 6666 (step 500) loss=893065 (avg=8.931e+05) [[all losses: diffusion=0.387243 ; kl=8.93064e+11 ; lpips=0.530042 ; repa=0.858555 ; sum_loss=893065]]
[T_total=00:17:00 | T_train=00:16:33 | T_epoch=00:16:33] Epoch 0, batch 601 / 6666 (step 600) loss=744468 (avg=7.445e+05) [[all losses: diffusion=0.343565 ; kl=7.44468e+11 ; lpips=0.512361 ; repa=0.84252 ; sum_loss=744468]]
[T_total=00:19:44 | T_train=00:19:17 | T_epoch=00:19:17] Epoch 0, batch 701 / 6666 (step 700) loss=638267 (avg=6.383e+05) [[all losses: diffusion=0.311698 ; kl=6.38267e+11 ; lpips=0.498459 ; repa=0.828846 ; sum_loss=638267]]
[T_total=00:22:28 | T_train=00:22:01 | T_epoch=00:22:01] Epoch 0, batch 801 / 6666 (step 800) loss=558584 (avg=5.586e+05) [[all losses: diffusion=0.287489 ; kl=5.58583e+11 ; lpips=0.48754 ; repa=0.817161 ; sum_loss=558584]]
[T_total=00:25:12 | T_train=00:24:45 | T_epoch=00:24:45] Epoch 0, batch 901 / 6666 (step 900) loss=496588 (avg=4.966e+05) [[all losses: diffusion=0.268364 ; kl=4.96587e+11 ; lpips=0.478222 ; repa=0.807135 ; sum_loss=496588]]
[T_total=00:27:56 | T_train=00:27:29 | T_epoch=00:27:29] Epoch 0, batch 1001 / 6666 (step 1000) loss=446979 (avg=4.47e+05) [[all losses: diffusion=0.252811 ; kl=4.46978e+11 ; lpips=0.470516 ; repa=0.798425 ; sum_loss=446979]]
[T_total=00:30:40 | T_train=00:30:13 | T_epoch=00:30:13] Epoch 0, batch 1101 / 6666 (step 1100) loss=406381 (avg=4.064e+05) [[all losses: diffusion=0.240129 ; kl=4.06381e+11 ; lpips=0.463378 ; repa=0.790714 ; sum_loss=406381]]
[T_total=00:33:24 | T_train=00:32:57 | T_epoch=00:32:57] Epoch 0, batch 1201 / 6666 (step 1200) loss=372627 (avg=3.726e+05) [[all losses: diffusion=0.229782 ; kl=3.72626e+11 ; lpips=0.45847 ; repa=0.783973 ; sum_loss=372627]]
[T_total=00:36:08 | T_train=00:35:41 | T_epoch=00:35:41] Epoch 0, batch 1301 / 6666 (step 1300) loss=343985 (avg=3.44e+05) [[all losses: diffusion=0.220689 ; kl=3.43984e+11 ; lpips=0.453284 ; repa=0.777787 ; sum_loss=343985]]
[T_total=00:38:52 | T_train=00:38:25 | T_epoch=00:38:25] Epoch 0, batch 1401 / 6666 (step 1400) loss=319432 (avg=3.194e+05) [[all losses: diffusion=0.212837 ; kl=3.19432e+11 ; lpips=0.448663 ; repa=0.772056 ; sum_loss=319432]]
[T_total=00:41:36 | T_train=00:41:09 | T_epoch=00:41:09] Epoch 0, batch 1501 / 6666 (step 1500) loss=298161 (avg=2.982e+05) [[all losses: diffusion=0.206076 ; kl=2.9816e+11 ; lpips=0.444743 ; repa=0.766849 ; sum_loss=298161]]
[T_total=00:44:19 | T_train=00:43:53 | T_epoch=00:43:53] Epoch 0, batch 1601 / 6666 (step 1600) loss=279538 (avg=2.795e+05) [[all losses: diffusion=0.200098 ; kl=2.79537e+11 ; lpips=0.44049 ; repa=0.761926 ; sum_loss=279538]]
[T_total=00:47:03 | T_train=00:46:37 | T_epoch=00:46:37] Epoch 0, batch 1701 / 6666 (step 1700) loss=263104 (avg=2.631e+05) [[all losses: diffusion=0.194673 ; kl=2.63103e+11 ; lpips=0.437218 ; repa=0.757459 ; sum_loss=263104]]
[T_total=00:49:47 | T_train=00:49:20 | T_epoch=00:49:20] Epoch 0, batch 1801 / 6666 (step 1800) loss=248495 (avg=2.485e+05) [[all losses: diffusion=0.189843 ; kl=2.48495e+11 ; lpips=0.433805 ; repa=0.753188 ; sum_loss=248495]]
[T_total=00:52:31 | T_train=00:52:04 | T_epoch=00:52:04] Epoch 0, batch 1901 / 6666 (step 1900) loss=235423 (avg=2.354e+05) [[all losses: diffusion=0.185441 ; kl=2.35423e+11 ; lpips=0.430557 ; repa=0.74918 ; sum_loss=235423]]
[T_total=00:55:15 | T_train=00:54:48 | T_epoch=00:54:48] Epoch 0, batch 2001 / 6666 (step 2000) loss=223658 (avg=2.237e+05) [[all losses: diffusion=0.18155 ; kl=2.23658e+11 ; lpips=0.42789 ; repa=0.74553 ; sum_loss=223658]]
[T_total=00:57:59 | T_train=00:57:32 | T_epoch=00:57:32] Epoch 0, batch 2101 / 6666 (step 2100) loss=213013 (avg=2.13e+05) [[all losses: diffusion=0.177931 ; kl=2.13013e+11 ; lpips=0.425059 ; repa=0.742016 ; sum_loss=213013]]
[T_total=01:00:43 | T_train=01:00:16 | T_epoch=01:00:16] Epoch 0, batch 2201 / 6666 (step 2200) loss=203335 (avg=2.033e+05) [[all losses: diffusion=0.174669 ; kl=2.03335e+11 ; lpips=0.422229 ; repa=0.738699 ; sum_loss=203335]]
[T_total=01:03:27 | T_train=01:03:01 | T_epoch=01:03:01] Epoch 0, batch 2301 / 6666 (step 2300) loss=194498 (avg=1.945e+05) [[all losses: diffusion=0.171763 ; kl=1.94498e+11 ; lpips=0.420135 ; repa=0.735667 ; sum_loss=194498]]
[T_total=01:06:11 | T_train=01:05:44 | T_epoch=01:05:44] Epoch 0, batch 2401 / 6666 (step 2400) loss=186398 (avg=1.864e+05) [[all losses: diffusion=0.16893 ; kl=1.86397e+11 ; lpips=0.417684 ; repa=0.732678 ; sum_loss=186398]]
[T_total=01:08:55 | T_train=01:08:28 | T_epoch=01:08:28] Epoch 0, batch 2501 / 6666 (step 2500) loss=178945 (avg=1.789e+05) [[all losses: diffusion=0.166345 ; kl=1.78944e+11 ; lpips=0.415514 ; repa=0.729864 ; sum_loss=178945]]
[T_total=01:11:39 | T_train=01:11:12 | T_epoch=01:11:12] Epoch 0, batch 2601 / 6666 (step 2600) loss=172065 (avg=1.721e+05) [[all losses: diffusion=0.16394 ; kl=1.72064e+11 ; lpips=0.41318 ; repa=0.727205 ; sum_loss=172065]]
[T_total=01:14:22 | T_train=01:13:55 | T_epoch=01:13:55] Epoch 0, batch 2701 / 6666 (step 2700) loss=165695 (avg=1.657e+05) [[all losses: diffusion=0.161669 ; kl=1.65694e+11 ; lpips=0.411074 ; repa=0.724636 ; sum_loss=165695]]
[T_total=01:17:05 | T_train=01:16:39 | T_epoch=01:16:39] Epoch 0, batch 2801 / 6666 (step 2800) loss=159779 (avg=1.598e+05) [[all losses: diffusion=0.159516 ; kl=1.59779e+11 ; lpips=0.409232 ; repa=0.722217 ; sum_loss=159779]]
[T_total=01:19:49 | T_train=01:19:23 | T_epoch=01:19:23] Epoch 0, batch 2901 / 6666 (step 2900) loss=154271 (avg=1.543e+05) [[all losses: diffusion=0.157551 ; kl=1.54271e+11 ; lpips=0.407319 ; repa=0.719872 ; sum_loss=154271]]
[T_total=01:22:33 | T_train=01:22:06 | T_epoch=01:22:06] Epoch 0, batch 3001 / 6666 (step 3000) loss=149147 (avg=1.491e+05) [[all losses: diffusion=0.155757 ; kl=1.49147e+11 ; lpips=0.405743 ; repa=0.717694 ; sum_loss=149147]]
[T_total=01:25:17 | T_train=01:24:50 | T_epoch=01:24:50] Epoch 0, batch 3101 / 6666 (step 3100) loss=144338 (avg=1.443e+05) [[all losses: diffusion=0.154043 ; kl=1.44337e+11 ; lpips=0.403968 ; repa=0.715546 ; sum_loss=144338]]
[T_total=01:28:01 | T_train=01:27:34 | T_epoch=01:27:34] Epoch 0, batch 3201 / 6666 (step 3200) loss=139828 (avg=1.398e+05) [[all losses: diffusion=0.15237 ; kl=1.39828e+11 ; lpips=0.402247 ; repa=0.713448 ; sum_loss=139828]]
[T_total=01:30:44 | T_train=01:30:18 | T_epoch=01:30:18] Epoch 0, batch 3301 / 6666 (step 3300) loss=135592 (avg=1.356e+05) [[all losses: diffusion=0.150841 ; kl=1.35592e+11 ; lpips=0.400374 ; repa=0.711401 ; sum_loss=135592]]
[T_total=01:33:28 | T_train=01:33:01 | T_epoch=01:33:01] Epoch 0, batch 3401 / 6666 (step 3400) loss=131606 (avg=1.316e+05) [[all losses: diffusion=0.149364 ; kl=1.31605e+11 ; lpips=0.398906 ; repa=0.709518 ; sum_loss=131606]]
[T_total=01:36:12 | T_train=01:35:45 | T_epoch=01:35:45] Epoch 0, batch 3501 / 6666 (step 3500) loss=127848 (avg=1.278e+05) [[all losses: diffusion=0.147982 ; kl=1.27847e+11 ; lpips=0.397529 ; repa=0.707721 ; sum_loss=127848]]
[T_total=01:38:56 | T_train=01:38:29 | T_epoch=01:38:29] Epoch 0, batch 3601 / 6666 (step 3600) loss=124298 (avg=1.243e+05) [[all losses: diffusion=0.146635 ; kl=1.24297e+11 ; lpips=0.395941 ; repa=0.705905 ; sum_loss=124298]]
[T_total=01:41:39 | T_train=01:41:13 | T_epoch=01:41:13] Epoch 0, batch 3701 / 6666 (step 3700) loss=120939 (avg=1.209e+05) [[all losses: diffusion=0.14539 ; kl=1.20939e+11 ; lpips=0.394409 ; repa=0.704188 ; sum_loss=120939]]
[T_total=01:44:23 | T_train=01:43:56 | T_epoch=01:43:56] Epoch 0, batch 3801 / 6666 (step 3800) loss=117757 (avg=1.178e+05) [[all losses: diffusion=0.144199 ; kl=1.17757e+11 ; lpips=0.39279 ; repa=0.702464 ; sum_loss=117757]]
[T_total=01:47:07 | T_train=01:46:40 | T_epoch=01:46:40] Epoch 0, batch 3901 / 6666 (step 3900) loss=114739 (avg=1.147e+05) [[all losses: diffusion=0.143076 ; kl=1.14738e+11 ; lpips=0.391549 ; repa=0.700919 ; sum_loss=114739]]
[T_total=01:49:51 | T_train=01:49:24 | T_epoch=01:49:24] Epoch 0, batch 4001 / 6666 (step 4000) loss=111871 (avg=1.119e+05) [[all losses: diffusion=0.142003 ; kl=1.11871e+11 ; lpips=0.390032 ; repa=0.699311 ; sum_loss=111871]]
[T_total=01:52:35 | T_train=01:52:08 | T_epoch=01:52:08] Epoch 0, batch 4101 / 6666 (step 4100) loss=109143 (avg=1.091e+05) [[all losses: diffusion=0.140951 ; kl=1.09143e+11 ; lpips=0.38867 ; repa=0.69779 ; sum_loss=109143]]
[T_total=01:55:19 | T_train=01:54:52 | T_epoch=01:54:52] Epoch 0, batch 4201 / 6666 (step 4200) loss=106545 (avg=1.065e+05) [[all losses: diffusion=0.139927 ; kl=1.06545e+11 ; lpips=0.387279 ; repa=0.696287 ; sum_loss=106545]]
[T_total=01:58:02 | T_train=01:57:36 | T_epoch=01:57:36] Epoch 0, batch 4301 / 6666 (step 4300) loss=104068 (avg=1.041e+05) [[all losses: diffusion=0.138941 ; kl=1.04067e+11 ; lpips=0.385905 ; repa=0.694821 ; sum_loss=104068]]
[T_total=02:00:46 | T_train=02:00:19 | T_epoch=02:00:19] Epoch 0, batch 4401 / 6666 (step 4400) loss=101703 (avg=1.017e+05) [[all losses: diffusion=0.138088 ; kl=1.01703e+11 ; lpips=0.384818 ; repa=0.693478 ; sum_loss=101703]]
[T_total=02:03:30 | T_train=02:03:03 | T_epoch=02:03:03] Epoch 0, batch 4501 / 6666 (step 4500) loss=99443.8 (avg=9.944e+04) [[all losses: diffusion=0.137204 ; kl=9.94433e+10 ; lpips=0.383534 ; repa=0.692084 ; sum_loss=99443.8]]
[T_total=02:06:13 | T_train=02:05:46 | T_epoch=02:05:46] Epoch 0, batch 4601 / 6666 (step 4600) loss=97285.3 (avg=9.729e+04) [[all losses: diffusion=0.136382 ; kl=9.72848e+10 ; lpips=0.382465 ; repa=0.690806 ; sum_loss=97285.3]]
[T_total=02:08:57 | T_train=02:08:30 | T_epoch=02:08:30] Epoch 0, batch 4701 / 6666 (step 4700) loss=95215.9 (avg=9.522e+04) [[all losses: diffusion=0.135576 ; kl=9.52154e+10 ; lpips=0.381187 ; repa=0.689488 ; sum_loss=95215.9]]
[T_total=02:11:40 | T_train=02:11:14 | T_epoch=02:11:14] Epoch 0, batch 4801 / 6666 (step 4800) loss=93232.6 (avg=9.323e+04) [[all losses: diffusion=0.134785 ; kl=9.32321e+10 ; lpips=0.380088 ; repa=0.688212 ; sum_loss=93232.6]]
[T_total=02:14:24 | T_train=02:13:57 | T_epoch=02:13:57] Epoch 0, batch 4901 / 6666 (step 4900) loss=91330.3 (avg=9.133e+04) [[all losses: diffusion=0.134018 ; kl=9.13298e+10 ; lpips=0.378915 ; repa=0.686957 ; sum_loss=91330.3]]
[T_total=02:17:08 | T_train=02:16:41 | T_epoch=02:16:41] Epoch 0, batch 5001 / 6666 (step 5000) loss=89504.1 (avg=8.95e+04) [[all losses: diffusion=0.133269 ; kl=8.95036e+10 ; lpips=0.377776 ; repa=0.685715 ; sum_loss=89504.1]]
[T_total=02:19:52 | T_train=02:19:25 | T_epoch=02:19:25] Epoch 0, batch 5101 / 6666 (step 5100) loss=87749.5 (avg=8.775e+04) [[all losses: diffusion=0.132582 ; kl=8.7749e+10 ; lpips=0.37657 ; repa=0.684502 ; sum_loss=87749.5]]
[T_total=02:22:35 | T_train=02:22:09 | T_epoch=02:22:09] Epoch 0, batch 5201 / 6666 (step 5200) loss=86062.4 (avg=8.606e+04) [[all losses: diffusion=0.131981 ; kl=8.6062e+10 ; lpips=0.375763 ; repa=0.683465 ; sum_loss=86062.4]]
[T_total=02:25:19 | T_train=02:24:52 | T_epoch=02:24:52] Epoch 0, batch 5301 / 6666 (step 5300) loss=84438.9 (avg=8.444e+04) [[all losses: diffusion=0.131319 ; kl=8.44385e+10 ; lpips=0.374613 ; repa=0.682315 ; sum_loss=84438.9]]
[T_total=02:28:03 | T_train=02:27:36 | T_epoch=02:27:36] Epoch 0, batch 5401 / 6666 (step 5400) loss=82875.6 (avg=8.288e+04) [[all losses: diffusion=0.13068 ; kl=8.28751e+10 ; lpips=0.373568 ; repa=0.681193 ; sum_loss=82875.6]]
[T_total=02:30:47 | T_train=02:30:20 | T_epoch=02:30:20] Epoch 0, batch 5501 / 6666 (step 5500) loss=81369 (avg=8.137e+04) [[all losses: diffusion=0.130048 ; kl=8.13685e+10 ; lpips=0.372563 ; repa=0.680121 ; sum_loss=81369]]
[T_total=02:33:30 | T_train=02:33:03 | T_epoch=02:33:03] Epoch 0, batch 5601 / 6666 (step 5600) loss=79918.6 (avg=7.992e+04) [[all losses: diffusion=0.129543 ; kl=7.99181e+10 ; lpips=0.371987 ; repa=0.679231 ; sum_loss=79918.6]]
[T_total=02:36:14 | T_train=02:35:47 | T_epoch=02:35:47] Epoch 0, batch 5701 / 6666 (step 5700) loss=78516.8 (avg=7.852e+04) [[all losses: diffusion=0.128967 ; kl=7.85163e+10 ; lpips=0.370938 ; repa=0.678176 ; sum_loss=78516.8]]
[T_total=02:38:58 | T_train=02:38:31 | T_epoch=02:38:31] Epoch 0, batch 5801 / 6666 (step 5800) loss=77163.3 (avg=7.716e+04) [[all losses: diffusion=0.128416 ; kl=7.71628e+10 ; lpips=0.369938 ; repa=0.677144 ; sum_loss=77163.3]]
[T_total=02:41:41 | T_train=02:41:15 | T_epoch=02:41:15] Epoch 0, batch 5901 / 6666 (step 5900) loss=75855.7 (avg=7.586e+04) [[all losses: diffusion=0.127848 ; kl=7.58552e+10 ; lpips=0.369027 ; repa=0.676136 ; sum_loss=75855.7]]
[T_total=02:44:25 | T_train=02:43:58 | T_epoch=02:43:58] Epoch 0, batch 6001 / 6666 (step 6000) loss=74591.6 (avg=7.459e+04) [[all losses: diffusion=0.127307 ; kl=7.45912e+10 ; lpips=0.368068 ; repa=0.675133 ; sum_loss=74591.6]]
[T_total=02:47:09 | T_train=02:46:42 | T_epoch=02:46:42] Epoch 0, batch 6101 / 6666 (step 6100) loss=73369 (avg=7.337e+04) [[all losses: diffusion=0.126812 ; kl=7.33685e+10 ; lpips=0.36707 ; repa=0.674153 ; sum_loss=73369]]
[T_total=02:49:53 | T_train=02:49:26 | T_epoch=02:49:26] Epoch 0, batch 6201 / 6666 (step 6200) loss=72187.3 (avg=7.219e+04) [[all losses: diffusion=0.126414 ; kl=7.21869e+10 ; lpips=0.3666 ; repa=0.673374 ; sum_loss=72187.3]]
[T_total=02:52:36 | T_train=02:52:09 | T_epoch=02:52:09] Epoch 0, batch 6301 / 6666 (step 6300) loss=71041.7 (avg=7.104e+04) [[all losses: diffusion=0.12591 ; kl=7.10412e+10 ; lpips=0.365679 ; repa=0.672437 ; sum_loss=71041.7]]
[T_total=02:55:20 | T_train=02:54:53 | T_epoch=02:54:53] Epoch 0, batch 6401 / 6666 (step 6400) loss=69931.8 (avg=6.993e+04) [[all losses: diffusion=0.125412 ; kl=6.99314e+10 ; lpips=0.36484 ; repa=0.67151 ; sum_loss=69931.8]]
[T_total=02:58:04 | T_train=02:57:37 | T_epoch=02:57:37] Epoch 0, batch 6501 / 6666 (step 6500) loss=68856.1 (avg=6.886e+04) [[all losses: diffusion=0.12495 ; kl=6.88557e+10 ; lpips=0.36393 ; repa=0.670594 ; sum_loss=68856.1]]
[T_total=03:00:48 | T_train=03:00:21 | T_epoch=03:00:21] Epoch 0, batch 6601 / 6666 (step 6600) loss=67813 (avg=6.781e+04) [[all losses: diffusion=0.124489 ; kl=6.78126e+10 ; lpips=0.363029 ; repa=0.669699 ; sum_loss=67813]]
[2025-10-25 07:13:55,982][main][INFO] - [T_total=03:02:34 | T_train=03:02:07 | T_epoch=03:02:07] End of epoch 0 (6666 steps) train loss 67151.8[2025-10-25 07:13:55,984][main][INFO] - [Epoch 0] All losses: [[diffusion=0.124198 ; kl=6.71513e+10 ; lpips=0.362462 ; repa=0.669115]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:02<12:46,  2.95s/it]
Reconstructing from test set:   1%|          | 2/261 [00:03<07:13,  1.67s/it]
Reconstructing from test set:   1%|          | 3/261 [00:04<05:25,  1.26s/it]
Reconstructing from test set:   2%|▏         | 4/261 [00:05<04:36,  1.08s/it]
Reconstructing from test set:   2%|▏         | 5/261 [00:06<04:07,  1.03it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:06<03:50,  1.10it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:07<03:39,  1.16it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:08<03:32,  1.19it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:09<03:26,  1.22it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:09<03:22,  1.24it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:10<03:20,  1.25it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:11<03:17,  1.26it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:12<03:15,  1.27it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:13<03:15,  1.27it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:13<03:14,  1.26it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:14<03:14,  1.26it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:15<03:13,  1.26it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:16<03:12,  1.26it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:17<03:11,  1.26it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:17<03:09,  1.27it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:18<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:19<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:20<03:05,  1.28it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:20<03:04,  1.29it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:22<03:03,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:23<03:03,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:24<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:24<03:00,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:25<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:26<02:59,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:27<02:58,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:27<02:57,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:28<02:56,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:29<02:56,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:30<02:55,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:31<02:54,  1.29it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:31<02:53,  1.29it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:32<02:52,  1.29it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:33<02:51,  1.29it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:34<02:50,  1.29it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:34<02:49,  1.29it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:35<02:49,  1.29it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:36<02:48,  1.29it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:37<02:47,  1.29it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:38<02:47,  1.29it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:38<02:46,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:39<02:46,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:40<02:45,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:41<02:45,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:42<02:45,  1.27it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:42<02:43,  1.28it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:43<02:43,  1.28it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:44<02:42,  1.27it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:45<02:41,  1.28it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:45<02:40,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:46<02:40,  1.27it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:47<02:39,  1.27it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:48<02:38,  1.27it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:49<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:49<02:36,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:50<02:35,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:51<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:52<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:52<02:32,  1.28it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:53<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:54<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:55<02:30,  1.28it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:56<02:30,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:56<02:29,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:57<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:58<02:27,  1.29it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:59<02:26,  1.29it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:59<02:25,  1.29it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [01:00<02:24,  1.29it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [01:01<02:23,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:02<02:23,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:03<02:22,  1.28it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:03<02:21,  1.28it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:05<02:20,  1.29it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:06<02:19,  1.29it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:06<02:18,  1.29it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:07<02:17,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:08<02:16,  1.29it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:09<02:16,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:10<02:15,  1.28it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:10<02:15,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:11<02:14,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:12<02:13,  1.28it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:13<02:12,  1.29it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:13<02:11,  1.29it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:14<02:10,  1.29it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:15<02:09,  1.29it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:16<02:08,  1.29it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:17<02:07,  1.29it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:17<02:06,  1.29it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:18<02:06,  1.29it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:19<02:05,  1.29it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:20<02:04,  1.29it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:20<02:04,  1.29it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:21<02:03,  1.29it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:22<02:02,  1.29it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:23<02:01,  1.29it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:24<02:01,  1.29it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:24<02:00,  1.28it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:25<02:00,  1.28it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:26<01:59,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:27<01:58,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:27<01:57,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:28<01:56,  1.29it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:29<01:55,  1.29it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:30<01:55,  1.29it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:31<01:54,  1.29it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:31<01:53,  1.29it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:32<01:52,  1.29it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:33<01:52,  1.28it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:34<01:51,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:34<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:35<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:36<01:49,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:37<01:48,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:38<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:38<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:39<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:40<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:41<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:42<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:42<01:43,  1.28it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:43<01:42,  1.28it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:44<01:41,  1.28it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:45<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:45<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:46<01:39,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:47<01:38,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:48<01:37,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:49<01:37,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:49<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:50<01:35,  1.28it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:51<01:34,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:52<01:33,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:52<01:33,  1.28it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:53<01:32,  1.27it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:54<01:31,  1.27it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:55<01:31,  1.27it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:56<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:56<01:29,  1.27it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:57<01:28,  1.27it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:58<01:27,  1.27it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:59<01:27,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [02:00<01:26,  1.27it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [02:00<01:25,  1.28it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:01<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:02<01:23,  1.27it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:03<01:23,  1.28it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:03<01:22,  1.28it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:04<01:21,  1.28it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:05<01:20,  1.28it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:06<01:19,  1.28it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:07<01:19,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:07<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:08<01:17,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:09<01:16,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:10<01:15,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:10<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:11<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:12<01:13,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:13<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:14<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:14<01:11,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:15<01:10,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:16<01:09,  1.28it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:17<01:08,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:18<01:08,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:18<01:07,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:19<01:06,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:20<01:05,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:21<01:05,  1.28it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:21<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:22<01:03,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:23<01:02,  1.28it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:24<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:25<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:25<01:00,  1.28it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:26<00:59,  1.28it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:27<00:58,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:28<00:57,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:28<00:57,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:29<00:56,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:30<00:55,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:31<00:54,  1.28it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:32<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:32<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:33<00:52,  1.28it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:34<00:51,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:35<00:50,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:36<00:50,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:36<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:37<00:48,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:38<00:47,  1.27it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:39<00:46,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:39<00:46,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:40<00:45,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:41<00:44,  1.28it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:42<00:43,  1.27it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:43<00:43,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:43<00:42,  1.28it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:44<00:41,  1.28it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:45<00:40,  1.28it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:46<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:46<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:47<00:38,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:48<00:37,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:49<00:36,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:50<00:36,  1.27it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:50<00:35,  1.27it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:51<00:34,  1.27it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:52<00:33,  1.27it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:53<00:33,  1.27it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:54<00:32,  1.27it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:54<00:31,  1.27it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:55<00:30,  1.27it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:56<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:57<00:28,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:57<00:28,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:58<00:27,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:59<00:26,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [03:00<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [03:01<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:01<00:24,  1.28it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:02<00:23,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:03<00:22,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:04<00:21,  1.28it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:05<00:21,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:05<00:20,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:06<00:19,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:07<00:18,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:08<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:08<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:09<00:16,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:10<00:15,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:11<00:14,  1.29it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:12<00:13,  1.29it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:12<00:13,  1.29it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:13<00:12,  1.29it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:14<00:11,  1.29it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:15<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:15<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:16<00:09,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:17<00:08,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:18<00:07,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:19<00:06,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:19<00:06,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:20<00:05,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:21<00:04,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:22<00:03,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:22<00:03,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:23<00:02,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:24<00:01,  1.29it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:25<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:26<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:26<00:00,  1.27it/s]
[2025-10-25 07:17:26,094][main][INFO] - [Epoch 1] Test metrics: [[MSE=55.56 | MAE=0.1763 | LPIPS=0.4494 | PSNR=12.55 | SSIM=0.2492 | dreamsim=0.6301 | FID=116.4]][2025-10-25 07:17:26,096][main][INFO] - [Epoch 1] Best metrics: [[min_MSE=55.56 | min_MAE=0.1763 | min_LPIPS=0.4494 | max_PSNR=12.55 | max_SSIM=0.2492 | min_dreamsim=0.6301 | min_FID=116.4]][2025-10-25 07:17:26,097][main][DEBUG] - Writing images to disk...[2025-10-25 07:17:27,322][main][DEBUG] - Image(s) saved on disk[2025-10-25 07:17:27,651][main][INFO] - End of epoch timers: [T_train=03:02:07 | T_epoch=03:02:07 | T_eval=00:03:31 | T_total=03:06:06][2025-10-25 07:17:27,652][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-25 07:17:38,934][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-25 07:17:49,112][main][INFO] - ---

[2025-10-25 07:17:49,113][main][INFO] - [T_total=03:06:27 | T_train=03:02:07] Start epoch 1[T_total=03:06:30 | T_train=03:02:10 | T_epoch=00:00:02] Epoch 1, batch 1 / 6666 (step 6666) loss=67141.7 (avg=0.3966) [[all losses: diffusion=0.124193 ; kl=6.71413e+10 ; lpips=0.362454 ; repa=0.669106 ; sum_loss=67141.7]]
[T_total=03:09:14 | T_train=03:04:54 | T_epoch=00:02:46] Epoch 1, batch 101 / 6666 (step 6766) loss=66149.5 (avg=0.4015) [[all losses: diffusion=0.123744 ; kl=6.61491e+10 ; lpips=0.361625 ; repa=0.668228 ; sum_loss=66149.5]]
[T_total=03:11:57 | T_train=03:07:38 | T_epoch=00:05:30] Epoch 1, batch 201 / 6666 (step 6866) loss=65192.8 (avg=225.7) [[all losses: diffusion=0.123389 ; kl=6.51924e+10 ; lpips=0.361139 ; repa=0.667521 ; sum_loss=65192.8]]
[T_total=03:14:41 | T_train=03:10:21 | T_epoch=00:08:13] Epoch 1, batch 301 / 6666 (step 6966) loss=64257.1 (avg=150.9) [[all losses: diffusion=0.122959 ; kl=6.42566e+10 ; lpips=0.360301 ; repa=0.666668 ; sum_loss=64257.1]]
[T_total=03:17:25 | T_train=03:13:05 | T_epoch=00:10:57] Epoch 1, batch 401 / 6666 (step 7066) loss=63347.9 (avg=113.3) [[all losses: diffusion=0.122555 ; kl=6.33474e+10 ; lpips=0.359462 ; repa=0.665835 ; sum_loss=63347.9]]
[T_total=03:20:09 | T_train=03:15:49 | T_epoch=00:13:41] Epoch 1, batch 501 / 6666 (step 7166) loss=62464 (avg=90.8) [[all losses: diffusion=0.12216 ; kl=6.24635e+10 ; lpips=0.358621 ; repa=0.665007 ; sum_loss=62464]]
[T_total=03:22:52 | T_train=03:18:32 | T_epoch=00:16:24] Epoch 1, batch 601 / 6666 (step 7266) loss=61604.4 (avg=75.76) [[all losses: diffusion=0.121757 ; kl=6.1604e+10 ; lpips=0.357764 ; repa=0.664191 ; sum_loss=61604.4]]
[T_total=03:25:36 | T_train=03:21:16 | T_epoch=00:19:08] Epoch 1, batch 701 / 6666 (step 7366) loss=60768.2 (avg=65.01) [[all losses: diffusion=0.121387 ; kl=6.07677e+10 ; lpips=0.356922 ; repa=0.663392 ; sum_loss=60768.2]]
[T_total=03:28:20 | T_train=03:24:00 | T_epoch=00:21:52] Epoch 1, batch 801 / 6666 (step 7466) loss=59954.4 (avg=56.94) [[all losses: diffusion=0.121016 ; kl=5.99539e+10 ; lpips=0.356072 ; repa=0.662587 ; sum_loss=59954.4]]
[T_total=03:31:04 | T_train=03:26:44 | T_epoch=00:24:36] Epoch 1, batch 901 / 6666 (step 7566) loss=59162.1 (avg=50.67) [[all losses: diffusion=0.120664 ; kl=5.91616e+10 ; lpips=0.355421 ; repa=0.661859 ; sum_loss=59162.1]]
[T_total=03:33:48 | T_train=03:29:28 | T_epoch=00:27:20] Epoch 1, batch 1001 / 6666 (step 7666) loss=58390.4 (avg=45.65) [[all losses: diffusion=0.120312 ; kl=5.839e+10 ; lpips=0.354618 ; repa=0.661084 ; sum_loss=58390.4]]
[T_total=03:36:32 | T_train=03:32:12 | T_epoch=00:30:04] Epoch 1, batch 1101 / 6666 (step 7766) loss=57638.7 (avg=41.54) [[all losses: diffusion=0.119958 ; kl=5.76382e+10 ; lpips=0.353855 ; repa=0.660331 ; sum_loss=57638.7]]
[T_total=03:39:15 | T_train=03:34:55 | T_epoch=00:32:47] Epoch 1, batch 1201 / 6666 (step 7866) loss=56906.1 (avg=38.5) [[all losses: diffusion=0.119707 ; kl=5.69056e+10 ; lpips=0.353329 ; repa=0.659705 ; sum_loss=56906.1]]
[T_total=03:41:59 | T_train=03:37:39 | T_epoch=00:35:31] Epoch 1, batch 1301 / 6666 (step 7966) loss=56191.8 (avg=35.57) [[all losses: diffusion=0.119382 ; kl=5.61913e+10 ; lpips=0.352532 ; repa=0.658959 ; sum_loss=56191.8]]
[T_total=03:44:43 | T_train=03:40:23 | T_epoch=00:38:15] Epoch 1, batch 1401 / 6666 (step 8066) loss=55495.2 (avg=33.06) [[all losses: diffusion=0.119078 ; kl=5.54948e+10 ; lpips=0.351739 ; repa=0.658234 ; sum_loss=55495.2]]
[T_total=03:47:27 | T_train=03:43:07 | T_epoch=00:40:59] Epoch 1, batch 1501 / 6666 (step 8166) loss=54815.7 (avg=30.9) [[all losses: diffusion=0.118789 ; kl=5.48153e+10 ; lpips=0.351212 ; repa=0.657598 ; sum_loss=54815.7]]
[T_total=03:50:10 | T_train=03:45:50 | T_epoch=00:43:42] Epoch 1, batch 1601 / 6666 (step 8266) loss=54152.7 (avg=29.01) [[all losses: diffusion=0.118474 ; kl=5.41522e+10 ; lpips=0.350614 ; repa=0.656938 ; sum_loss=54152.7]]
[T_total=03:52:54 | T_train=03:48:34 | T_epoch=00:46:26] Epoch 1, batch 1701 / 6666 (step 8366) loss=53505.5 (avg=27.33) [[all losses: diffusion=0.118189 ; kl=5.3505e+10 ; lpips=0.349841 ; repa=0.656239 ; sum_loss=53505.5]]
[T_total=03:55:38 | T_train=03:51:18 | T_epoch=00:49:10] Epoch 1, batch 1801 / 6666 (step 8466) loss=52884.7 (avg=78.11) [[all losses: diffusion=0.117991 ; kl=5.28842e+10 ; lpips=0.349474 ; repa=0.655706 ; sum_loss=52884.7]]
[T_total=03:58:22 | T_train=03:54:02 | T_epoch=00:51:54] Epoch 1, batch 1901 / 6666 (step 8566) loss=52267.4 (avg=74.02) [[all losses: diffusion=0.1177 ; kl=5.22669e+10 ; lpips=0.348757 ; repa=0.655024 ; sum_loss=52267.4]]
[T_total=04:01:06 | T_train=03:56:46 | T_epoch=00:54:38] Epoch 1, batch 2001 / 6666 (step 8666) loss=51664.3 (avg=70.34) [[all losses: diffusion=0.117417 ; kl=5.16639e+10 ; lpips=0.348065 ; repa=0.654374 ; sum_loss=51664.3]]
[T_total=04:03:49 | T_train=03:59:29 | T_epoch=00:57:21] Epoch 1, batch 2101 / 6666 (step 8766) loss=51075 (avg=67.01) [[all losses: diffusion=0.117153 ; kl=5.10746e+10 ; lpips=0.347434 ; repa=0.653746 ; sum_loss=51075]]
[T_total=04:06:33 | T_train=04:02:13 | T_epoch=01:00:05] Epoch 1, batch 2201 / 6666 (step 8866) loss=50499 (avg=63.99) [[all losses: diffusion=0.116876 ; kl=5.04986e+10 ; lpips=0.346704 ; repa=0.653082 ; sum_loss=50499]]
[T_total=04:09:17 | T_train=04:04:57 | T_epoch=01:02:49] Epoch 1, batch 2301 / 6666 (step 8966) loss=49935.8 (avg=61.22) [[all losses: diffusion=0.116586 ; kl=4.99354e+10 ; lpips=0.34602 ; repa=0.652425 ; sum_loss=49935.8]]
[T_total=04:12:00 | T_train=04:07:40 | T_epoch=01:05:32] Epoch 1, batch 2401 / 6666 (step 9066) loss=49385.1 (avg=58.69) [[all losses: diffusion=0.11631 ; kl=4.93847e+10 ; lpips=0.345348 ; repa=0.651786 ; sum_loss=49385.1]]
[T_total=04:14:44 | T_train=04:10:24 | T_epoch=01:08:16] Epoch 1, batch 2501 / 6666 (step 9166) loss=48846.4 (avg=56.36) [[all losses: diffusion=0.116048 ; kl=4.88459e+10 ; lpips=0.34465 ; repa=0.651151 ; sum_loss=48846.4]]
[T_total=04:17:28 | T_train=04:13:08 | T_epoch=01:11:00] Epoch 1, batch 2601 / 6666 (step 9266) loss=48319.3 (avg=54.2) [[all losses: diffusion=0.115771 ; kl=4.83188e+10 ; lpips=0.343944 ; repa=0.650517 ; sum_loss=48319.3]]
[T_total=04:20:12 | T_train=04:15:52 | T_epoch=01:13:44] Epoch 1, batch 2701 / 6666 (step 9366) loss=47803.4 (avg=52.21) [[all losses: diffusion=0.115509 ; kl=4.7803e+10 ; lpips=0.343254 ; repa=0.649885 ; sum_loss=47803.4]]
[T_total=04:22:55 | T_train=04:18:35 | T_epoch=01:16:27] Epoch 1, batch 2801 / 6666 (step 9466) loss=47298.5 (avg=50.36) [[all losses: diffusion=0.115261 ; kl=4.72981e+10 ; lpips=0.342565 ; repa=0.649272 ; sum_loss=47298.5]]
[T_total=04:25:39 | T_train=04:21:19 | T_epoch=01:19:11] Epoch 1, batch 2901 / 6666 (step 9566) loss=46804.2 (avg=48.97) [[all losses: diffusion=0.115038 ; kl=4.68038e+10 ; lpips=0.341964 ; repa=0.648699 ; sum_loss=46804.2]]
[T_total=04:28:23 | T_train=04:24:03 | T_epoch=01:21:55] Epoch 1, batch 3001 / 6666 (step 9666) loss=46320.1 (avg=47.35) [[all losses: diffusion=0.114792 ; kl=4.63196e+10 ; lpips=0.341295 ; repa=0.648096 ; sum_loss=46320.1]]
[T_total=04:31:07 | T_train=04:26:47 | T_epoch=01:24:39] Epoch 1, batch 3101 / 6666 (step 9766) loss=45845.8 (avg=45.83) [[all losses: diffusion=0.114555 ; kl=4.58454e+10 ; lpips=0.340638 ; repa=0.647491 ; sum_loss=45845.8]]
[T_total=04:33:50 | T_train=04:29:30 | T_epoch=01:27:22] Epoch 1, batch 3201 / 6666 (step 9866) loss=45381.2 (avg=44.41) [[all losses: diffusion=0.114331 ; kl=4.53807e+10 ; lpips=0.339964 ; repa=0.646903 ; sum_loss=45381.2]]
[T_total=04:36:34 | T_train=04:32:14 | T_epoch=01:30:06] Epoch 1, batch 3301 / 6666 (step 9966) loss=44925.9 (avg=43.08) [[all losses: diffusion=0.114102 ; kl=4.49254e+10 ; lpips=0.33931 ; repa=0.646316 ; sum_loss=44925.9]]
[T_total=04:39:18 | T_train=04:34:58 | T_epoch=01:32:50] Epoch 1, batch 3401 / 6666 (step 10066) loss=44479.6 (avg=41.82) [[all losses: diffusion=0.11388 ; kl=4.44792e+10 ; lpips=0.338648 ; repa=0.645732 ; sum_loss=44479.6]]
[T_total=04:42:02 | T_train=04:37:42 | T_epoch=01:35:34] Epoch 1, batch 3501 / 6666 (step 10166) loss=44042.1 (avg=40.64) [[all losses: diffusion=0.113657 ; kl=4.40417e+10 ; lpips=0.337992 ; repa=0.645156 ; sum_loss=44042.1]]
[T_total=04:44:45 | T_train=04:40:25 | T_epoch=01:38:17] Epoch 1, batch 3601 / 6666 (step 10266) loss=43613.1 (avg=39.52) [[all losses: diffusion=0.113419 ; kl=4.36127e+10 ; lpips=0.337363 ; repa=0.644574 ; sum_loss=43613.1]]
[T_total=04:47:29 | T_train=04:43:09 | T_epoch=01:41:01] Epoch 1, batch 3701 / 6666 (step 10366) loss=43192.5 (avg=38.46) [[all losses: diffusion=0.1132 ; kl=4.3192e+10 ; lpips=0.336711 ; repa=0.644 ; sum_loss=43192.5]]
[T_total=04:50:13 | T_train=04:45:53 | T_epoch=01:43:45] Epoch 1, batch 3801 / 6666 (step 10466) loss=42779.8 (avg=37.53) [[all losses: diffusion=0.113063 ; kl=4.27794e+10 ; lpips=0.336388 ; repa=0.643572 ; sum_loss=42779.8]]
[T_total=04:52:56 | T_train=04:48:36 | T_epoch=01:46:28] Epoch 1, batch 3901 / 6666 (step 10566) loss=42375 (avg=36.58) [[all losses: diffusion=0.11286 ; kl=4.23746e+10 ; lpips=0.335765 ; repa=0.643012 ; sum_loss=42375]]
[T_total=04:55:40 | T_train=04:51:20 | T_epoch=01:49:12] Epoch 1, batch 4001 / 6666 (step 10666) loss=41977.7 (avg=35.68) [[all losses: diffusion=0.112648 ; kl=4.19773e+10 ; lpips=0.335169 ; repa=0.64246 ; sum_loss=41977.7]]
[T_total=04:58:24 | T_train=04:54:04 | T_epoch=01:51:56] Epoch 1, batch 4101 / 6666 (step 10766) loss=41587.9 (avg=34.82) [[all losses: diffusion=0.112447 ; kl=4.15874e+10 ; lpips=0.334556 ; repa=0.641918 ; sum_loss=41587.9]]
[T_total=05:01:07 | T_train=04:56:47 | T_epoch=01:54:39] Epoch 1, batch 4201 / 6666 (step 10866) loss=41205.2 (avg=34) [[all losses: diffusion=0.112243 ; kl=4.12047e+10 ; lpips=0.333954 ; repa=0.641376 ; sum_loss=41205.2]]
[T_total=05:03:51 | T_train=04:59:31 | T_epoch=01:57:23] Epoch 1, batch 4301 / 6666 (step 10966) loss=40829.5 (avg=33.21) [[all losses: diffusion=0.112033 ; kl=4.0829e+10 ; lpips=0.333365 ; repa=0.640841 ; sum_loss=40829.5]]
[T_total=05:06:35 | T_train=05:02:15 | T_epoch=02:00:07] Epoch 1, batch 4401 / 6666 (step 11066) loss=40461 (avg=33.74) [[all losses: diffusion=0.11191 ; kl=4.04606e+10 ; lpips=0.333076 ; repa=0.640439 ; sum_loss=40461]]
[T_total=05:09:19 | T_train=05:04:59 | T_epoch=02:02:51] Epoch 1, batch 4501 / 6666 (step 11166) loss=40098.7 (avg=32.99) [[all losses: diffusion=0.111713 ; kl=4.00983e+10 ; lpips=0.33249 ; repa=0.639912 ; sum_loss=40098.7]]
[T_total=05:12:02 | T_train=05:07:42 | T_epoch=02:05:34] Epoch 1, batch 4601 / 6666 (step 11266) loss=39742.8 (avg=32.29) [[all losses: diffusion=0.111519 ; kl=3.97424e+10 ; lpips=0.33191 ; repa=0.639385 ; sum_loss=39742.8]]
[T_total=05:14:46 | T_train=05:10:26 | T_epoch=02:08:18] Epoch 1, batch 4701 / 6666 (step 11366) loss=39393.2 (avg=31.61) [[all losses: diffusion=0.111328 ; kl=3.93928e+10 ; lpips=0.331316 ; repa=0.638872 ; sum_loss=39393.2]]
[T_total=05:17:30 | T_train=05:13:10 | T_epoch=02:11:02] Epoch 1, batch 4801 / 6666 (step 11466) loss=39049.9 (avg=31.52) [[all losses: diffusion=0.111173 ; kl=3.90495e+10 ; lpips=0.330897 ; repa=0.638424 ; sum_loss=39049.9]]
[T_total=05:20:14 | T_train=05:15:54 | T_epoch=02:13:46] Epoch 1, batch 4901 / 6666 (step 11566) loss=38712.3 (avg=30.88) [[all losses: diffusion=0.111005 ; kl=3.87119e+10 ; lpips=0.330305 ; repa=0.637915 ; sum_loss=38712.3]]
[T_total=05:22:57 | T_train=05:18:37 | T_epoch=02:16:29] Epoch 1, batch 5001 / 6666 (step 11666) loss=38380.5 (avg=30.27) [[all losses: diffusion=0.110831 ; kl=3.83801e+10 ; lpips=0.329706 ; repa=0.637415 ; sum_loss=38380.5]]
[T_total=05:25:41 | T_train=05:21:21 | T_epoch=02:19:13] Epoch 1, batch 5101 / 6666 (step 11766) loss=38054.3 (avg=29.69) [[all losses: diffusion=0.110647 ; kl=3.80539e+10 ; lpips=0.329155 ; repa=0.636917 ; sum_loss=38054.3]]
[T_total=05:28:25 | T_train=05:24:05 | T_epoch=02:21:57] Epoch 1, batch 5201 / 6666 (step 11866) loss=37733.7 (avg=29.12) [[all losses: diffusion=0.110466 ; kl=3.77332e+10 ; lpips=0.328573 ; repa=0.636421 ; sum_loss=37733.7]]
[T_total=05:31:08 | T_train=05:26:48 | T_epoch=02:24:40] Epoch 1, batch 5301 / 6666 (step 11966) loss=37418.3 (avg=28.58) [[all losses: diffusion=0.110292 ; kl=3.74179e+10 ; lpips=0.328021 ; repa=0.635922 ; sum_loss=37418.3]]
[T_total=05:33:52 | T_train=05:29:32 | T_epoch=02:27:24] Epoch 1, batch 5401 / 6666 (step 12066) loss=37108.3 (avg=28.06) [[all losses: diffusion=0.110121 ; kl=3.71078e+10 ; lpips=0.327457 ; repa=0.635432 ; sum_loss=37108.3]]
[T_total=05:36:36 | T_train=05:32:16 | T_epoch=02:30:08] Epoch 1, batch 5501 / 6666 (step 12166) loss=36803.3 (avg=27.55) [[all losses: diffusion=0.109946 ; kl=3.68028e+10 ; lpips=0.326951 ; repa=0.634958 ; sum_loss=36803.3]]
[T_total=05:39:20 | T_train=05:35:00 | T_epoch=02:32:52] Epoch 1, batch 5601 / 6666 (step 12266) loss=36503.3 (avg=27.07) [[all losses: diffusion=0.10978 ; kl=3.65028e+10 ; lpips=0.3264 ; repa=0.634481 ; sum_loss=36503.3]]
[T_total=05:42:04 | T_train=05:37:44 | T_epoch=02:35:36] Epoch 1, batch 5701 / 6666 (step 12366) loss=36208.1 (avg=26.6) [[all losses: diffusion=0.1096 ; kl=3.62077e+10 ; lpips=0.325871 ; repa=0.634002 ; sum_loss=36208.1]]
[T_total=05:44:48 | T_train=05:40:28 | T_epoch=02:38:20] Epoch 1, batch 5801 / 6666 (step 12466) loss=35917.7 (avg=26.15) [[all losses: diffusion=0.109445 ; kl=3.59172e+10 ; lpips=0.325313 ; repa=0.63353 ; sum_loss=35917.7]]
[T_total=05:47:32 | T_train=05:43:12 | T_epoch=02:41:04] Epoch 1, batch 5901 / 6666 (step 12566) loss=35631.9 (avg=25.71) [[all losses: diffusion=0.109288 ; kl=3.56314e+10 ; lpips=0.324779 ; repa=0.633072 ; sum_loss=35631.9]]
[T_total=05:50:15 | T_train=05:45:55 | T_epoch=02:43:47] Epoch 1, batch 6001 / 6666 (step 12666) loss=35350.6 (avg=25.29) [[all losses: diffusion=0.109135 ; kl=3.53501e+10 ; lpips=0.324246 ; repa=0.632612 ; sum_loss=35350.6]]
[T_total=05:52:59 | T_train=05:48:39 | T_epoch=02:46:31] Epoch 1, batch 6101 / 6666 (step 12766) loss=35073.7 (avg=24.88) [[all losses: diffusion=0.108987 ; kl=3.50733e+10 ; lpips=0.323699 ; repa=0.632149 ; sum_loss=35073.7]]
[T_total=05:55:43 | T_train=05:51:23 | T_epoch=02:49:15] Epoch 1, batch 6201 / 6666 (step 12866) loss=34801.1 (avg=24.48) [[all losses: diffusion=0.108836 ; kl=3.48007e+10 ; lpips=0.323157 ; repa=0.631694 ; sum_loss=34801.1]]
[T_total=05:58:27 | T_train=05:54:07 | T_epoch=02:51:59] Epoch 1, batch 6301 / 6666 (step 12966) loss=34532.7 (avg=24.1) [[all losses: diffusion=0.108685 ; kl=3.45323e+10 ; lpips=0.322645 ; repa=0.631245 ; sum_loss=34532.7]]
[T_total=06:01:10 | T_train=05:56:50 | T_epoch=02:54:42] Epoch 1, batch 6401 / 6666 (step 13066) loss=34269.8 (avg=26.56) [[all losses: diffusion=0.108629 ; kl=3.42694e+10 ; lpips=0.32254 ; repa=0.630979 ; sum_loss=34269.8]]
[T_total=06:03:54 | T_train=05:59:34 | T_epoch=02:57:26] Epoch 1, batch 6501 / 6666 (step 13166) loss=34009.6 (avg=26.16) [[all losses: diffusion=0.108475 ; kl=3.40091e+10 ; lpips=0.322036 ; repa=0.630531 ; sum_loss=34009.6]]
[T_total=06:06:38 | T_train=06:02:18 | T_epoch=03:00:10] Epoch 1, batch 6601 / 6666 (step 13266) loss=33753.2 (avg=25.77) [[all losses: diffusion=0.108357 ; kl=3.37528e+10 ; lpips=0.321644 ; repa=0.630148 ; sum_loss=33753.2]]
[2025-10-25 10:19:46,050][main][INFO] - [T_total=06:08:24 | T_train=06:04:04 | T_epoch=03:01:56] End of epoch 1 (13332 steps) train loss 25.5196[2025-10-25 10:19:46,051][main][INFO] - [Epoch 1] All losses: [[diffusion=0.0923335 ; kl=2.51396e+07 ; lpips=0.280141 ; repa=0.590596]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:01<06:46,  1.56s/it]
Reconstructing from test set:   1%|          | 2/261 [00:02<04:45,  1.10s/it]
Reconstructing from test set:   1%|          | 3/261 [00:03<04:06,  1.05it/s]
Reconstructing from test set:   2%|▏         | 4/261 [00:03<03:48,  1.12it/s]
Reconstructing from test set:   2%|▏         | 5/261 [00:04<03:37,  1.18it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:05<03:30,  1.21it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:06<03:26,  1.23it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:07<03:23,  1.24it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:07<03:20,  1.26it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:08<03:18,  1.26it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:09<03:16,  1.27it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:10<03:15,  1.27it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:10<03:14,  1.27it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:11<03:14,  1.27it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:12<03:12,  1.28it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:13<03:12,  1.28it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:14<03:11,  1.28it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:14<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:15<03:09,  1.28it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:16<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:17<03:07,  1.28it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:17<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:18<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:19<03:05,  1.28it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:20<03:04,  1.28it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:21<03:02,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:21<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:22<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:23<03:01,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:24<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:25<02:59,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:25<02:59,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:26<02:58,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:27<02:57,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:28<02:56,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:28<02:55,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:29<02:55,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:30<02:54,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:31<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:32<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:32<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:33<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:34<02:50,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:35<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:36<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:37<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:38<02:46,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:39<02:45,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:39<02:44,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:40<02:44,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:41<02:43,  1.28it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:42<02:42,  1.28it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:43<02:42,  1.28it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:43<02:41,  1.28it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:44<02:40,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:45<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:46<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:47<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:48<02:36,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:49<02:35,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:50<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:50<02:33,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:51<02:32,  1.28it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:52<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:53<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:53<02:30,  1.28it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:54<02:29,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:55<02:29,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:56<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:57<02:27,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:57<02:26,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:58<02:26,  1.28it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [00:59<02:25,  1.28it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [01:00<02:24,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:00<02:23,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:01<02:22,  1.28it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:02<02:22,  1.28it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:03<02:21,  1.28it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:05<02:19,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:06<02:18,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:07<02:18,  1.27it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:08<02:17,  1.27it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:08<02:16,  1.27it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:09<02:15,  1.27it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:10<02:14,  1.27it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:11<02:14,  1.28it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:11<02:13,  1.27it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:12<02:12,  1.27it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:13<02:11,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:14<02:10,  1.28it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:15<02:09,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:15<02:08,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:16<02:08,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:17<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:18<02:06,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:18<02:05,  1.28it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:19<02:05,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:20<02:04,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:21<02:03,  1.28it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:22<02:02,  1.28it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:22<02:02,  1.28it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:23<02:01,  1.28it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:24<02:00,  1.28it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:25<01:59,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:26<01:59,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:26<01:58,  1.27it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:27<01:57,  1.27it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:28<01:56,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:29<01:55,  1.28it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:29<01:54,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:30<01:53,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:31<01:53,  1.28it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:32<01:52,  1.28it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:33<01:51,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:33<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:34<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:35<01:49,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:36<01:48,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:36<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:37<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:38<01:45,  1.29it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:39<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:40<01:45,  1.27it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:40<01:44,  1.27it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:41<01:43,  1.27it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:42<01:42,  1.28it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:43<01:42,  1.27it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:44<01:41,  1.27it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:44<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:45<01:39,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:46<01:38,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:47<01:37,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:47<01:37,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:48<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:49<01:35,  1.28it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:50<01:34,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:51<01:34,  1.27it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:51<01:33,  1.27it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:52<01:32,  1.27it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:53<01:31,  1.27it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:54<01:31,  1.27it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:55<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:55<01:29,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:56<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:57<01:27,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:58<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [01:58<01:25,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [01:59<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:00<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:01<01:23,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:03<01:21,  1.28it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:04<01:20,  1.28it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:05<01:19,  1.28it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:05<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:06<01:17,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:07<01:17,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:08<01:16,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:09<01:15,  1.29it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:09<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:10<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:11<01:13,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:12<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:12<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:13<01:11,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:14<01:10,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:15<01:09,  1.28it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:16<01:08,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:16<01:08,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:17<01:07,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:18<01:06,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:19<01:05,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:20<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:20<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:21<01:03,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:22<01:02,  1.28it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:23<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:23<01:00,  1.28it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:24<01:00,  1.28it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:25<00:59,  1.28it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:26<00:58,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:27<00:57,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:27<00:57,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:28<00:56,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:29<00:55,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:30<00:54,  1.28it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:30<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:31<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:32<00:52,  1.28it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:33<00:51,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:34<00:50,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:34<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:35<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:36<00:48,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:37<00:47,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:37<00:46,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:38<00:45,  1.29it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:39<00:45,  1.29it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:40<00:44,  1.28it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:41<00:43,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:41<00:42,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:42<00:42,  1.28it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:43<00:41,  1.28it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:44<00:40,  1.28it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:45<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:45<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:46<00:38,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:47<00:37,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:49<00:35,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:50<00:34,  1.28it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:51<00:33,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:53<00:31,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:54<00:30,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:55<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:55<00:28,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:56<00:27,  1.29it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:57<00:27,  1.29it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:58<00:26,  1.29it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [02:59<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [02:59<00:24,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:00<00:24,  1.28it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:01<00:23,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:02<00:22,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:02<00:21,  1.28it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:03<00:21,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:04<00:20,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:05<00:19,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:06<00:18,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:06<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:08<00:16,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:09<00:15,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:09<00:14,  1.29it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:10<00:13,  1.29it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:11<00:13,  1.29it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:12<00:12,  1.29it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:13<00:11,  1.29it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:13<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:14<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:15<00:09,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:16<00:08,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:16<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:17<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:18<00:06,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:19<00:05,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:20<00:04,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:20<00:03,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:21<00:03,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:22<00:02,  1.28it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:23<00:01,  1.28it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:23<00:00,  1.28it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.27it/s]
[2025-10-25 10:23:14,102][main][INFO] - [Epoch 2] Test metrics: [[MSE=47.13 | MAE=0.1616 | LPIPS=0.3504 | PSNR=13.27 | SSIM=0.2853 | dreamsim=0.5293 | FID=90.87]][2025-10-25 10:23:14,104][main][INFO] - [Epoch 2] Best metrics: [[min_MSE=47.13 | min_MAE=0.1616 | min_LPIPS=0.3504 | max_PSNR=13.27 | max_SSIM=0.2853 | min_dreamsim=0.5293 | min_FID=90.87]][2025-10-25 10:23:14,106][main][DEBUG] - Writing images to disk...[2025-10-25 10:23:15,187][main][DEBUG] - Image(s) saved on disk[2025-10-25 10:23:15,438][main][INFO] - End of epoch timers: [T_train=06:04:04 | T_epoch=03:01:56 | T_eval=00:07:00 | T_total=06:11:54][2025-10-25 10:23:15,439][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-25 10:23:26,115][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-25 10:23:36,944][main][INFO] - ---

[2025-10-25 10:23:36,945][main][INFO] - [T_total=06:12:15 | T_train=06:04:04] Start epoch 2[T_total=06:12:18 | T_train=06:04:07 | T_epoch=00:00:02] Epoch 2, batch 1 / 6666 (step 13332) loss=33586.1 (avg=0.3623) [[all losses: diffusion=0.108265 ; kl=3.35857e+10 ; lpips=0.321295 ; repa=0.629851 ; sum_loss=33586.1]]
[T_total=06:15:02 | T_train=06:06:51 | T_epoch=00:02:46] Epoch 2, batch 101 / 6666 (step 13432) loss=33336.1 (avg=0.36) [[all losses: diffusion=0.108113 ; kl=3.33357e+10 ; lpips=0.320796 ; repa=0.62941 ; sum_loss=33336.1]]
[T_total=06:17:46 | T_train=06:09:35 | T_epoch=00:05:30] Epoch 2, batch 201 / 6666 (step 13532) loss=33089.8 (avg=0.3596) [[all losses: diffusion=0.107976 ; kl=3.30894e+10 ; lpips=0.320269 ; repa=0.628971 ; sum_loss=33089.8]]
[T_total=06:20:29 | T_train=06:12:19 | T_epoch=00:08:14] Epoch 2, batch 301 / 6666 (step 13632) loss=32847.1 (avg=0.3596) [[all losses: diffusion=0.107833 ; kl=3.28466e+10 ; lpips=0.319774 ; repa=0.628538 ; sum_loss=32847.1]]
[T_total=06:23:13 | T_train=06:15:02 | T_epoch=00:10:57] Epoch 2, batch 401 / 6666 (step 13732) loss=32607.9 (avg=0.3594) [[all losses: diffusion=0.107686 ; kl=3.26075e+10 ; lpips=0.319282 ; repa=0.628112 ; sum_loss=32607.9]]
[T_total=06:25:57 | T_train=06:17:46 | T_epoch=00:13:41] Epoch 2, batch 501 / 6666 (step 13832) loss=32372.2 (avg=0.3593) [[all losses: diffusion=0.107545 ; kl=3.23717e+10 ; lpips=0.318797 ; repa=0.627689 ; sum_loss=32372.2]]
[T_total=06:28:41 | T_train=06:20:30 | T_epoch=00:16:25] Epoch 2, batch 601 / 6666 (step 13932) loss=32139.8 (avg=0.4724) [[all losses: diffusion=0.107449 ; kl=3.21394e+10 ; lpips=0.318487 ; repa=0.627348 ; sum_loss=32139.8]]
[T_total=06:31:25 | T_train=06:23:14 | T_epoch=00:19:09] Epoch 2, batch 701 / 6666 (step 14032) loss=31912.5 (avg=33.52) [[all losses: diffusion=0.107407 ; kl=3.1912e+10 ; lpips=0.318374 ; repa=0.627103 ; sum_loss=31912.5]]
[T_total=06:34:08 | T_train=06:25:58 | T_epoch=00:21:53] Epoch 2, batch 801 / 6666 (step 14132) loss=31686.7 (avg=29.38) [[all losses: diffusion=0.107277 ; kl=3.16862e+10 ; lpips=0.317928 ; repa=0.6267 ; sum_loss=31686.7]]
[T_total=06:36:52 | T_train=06:28:41 | T_epoch=00:24:36] Epoch 2, batch 901 / 6666 (step 14232) loss=31464 (avg=26.16) [[all losses: diffusion=0.107144 ; kl=3.14636e+10 ; lpips=0.317432 ; repa=0.626282 ; sum_loss=31464]]
[T_total=06:39:36 | T_train=06:31:25 | T_epoch=00:27:20] Epoch 2, batch 1001 / 6666 (step 14332) loss=31244.5 (avg=23.7) [[all losses: diffusion=0.107029 ; kl=3.12441e+10 ; lpips=0.317032 ; repa=0.62591 ; sum_loss=31244.5]]
[T_total=06:42:20 | T_train=06:34:09 | T_epoch=00:30:04] Epoch 2, batch 1101 / 6666 (step 14432) loss=31028 (avg=21.58) [[all losses: diffusion=0.106902 ; kl=3.10276e+10 ; lpips=0.316558 ; repa=0.625508 ; sum_loss=31028]]
[T_total=06:45:03 | T_train=06:36:52 | T_epoch=00:32:48] Epoch 2, batch 1201 / 6666 (step 14532) loss=30814.5 (avg=19.82) [[all losses: diffusion=0.106774 ; kl=3.08141e+10 ; lpips=0.316077 ; repa=0.625102 ; sum_loss=30814.5]]
[T_total=06:47:47 | T_train=06:39:36 | T_epoch=00:35:31] Epoch 2, batch 1301 / 6666 (step 14632) loss=30604 (avg=18.32) [[all losses: diffusion=0.106636 ; kl=3.06035e+10 ; lpips=0.315619 ; repa=0.624702 ; sum_loss=30604]]
[T_total=06:50:31 | T_train=06:42:20 | T_epoch=00:38:15] Epoch 2, batch 1401 / 6666 (step 14732) loss=30396.2 (avg=17.04) [[all losses: diffusion=0.106504 ; kl=3.03958e+10 ; lpips=0.315157 ; repa=0.624296 ; sum_loss=30396.2]]
[T_total=06:53:15 | T_train=06:45:04 | T_epoch=00:40:59] Epoch 2, batch 1501 / 6666 (step 14832) loss=30191.3 (avg=15.93) [[all losses: diffusion=0.106385 ; kl=3.01909e+10 ; lpips=0.314667 ; repa=0.623894 ; sum_loss=30191.3]]
[T_total=06:55:59 | T_train=06:47:48 | T_epoch=00:43:43] Epoch 2, batch 1601 / 6666 (step 14932) loss=29989.1 (avg=14.95) [[all losses: diffusion=0.10626 ; kl=2.99887e+10 ; lpips=0.314194 ; repa=0.623503 ; sum_loss=29989.1]]
[T_total=06:58:43 | T_train=06:50:32 | T_epoch=00:46:27] Epoch 2, batch 1701 / 6666 (step 15032) loss=29789.7 (avg=14.1) [[all losses: diffusion=0.106131 ; kl=2.97892e+10 ; lpips=0.313738 ; repa=0.623112 ; sum_loss=29789.7]]
[T_total=07:01:26 | T_train=06:53:15 | T_epoch=00:49:11] Epoch 2, batch 1801 / 6666 (step 15132) loss=29592.8 (avg=13.33) [[all losses: diffusion=0.10601 ; kl=2.95924e+10 ; lpips=0.313275 ; repa=0.622718 ; sum_loss=29592.8]]
[T_total=07:04:10 | T_train=06:55:59 | T_epoch=00:51:54] Epoch 2, batch 1901 / 6666 (step 15232) loss=29398.5 (avg=12.65) [[all losses: diffusion=0.105895 ; kl=2.93981e+10 ; lpips=0.312812 ; repa=0.622333 ; sum_loss=29398.5]]
[T_total=07:06:54 | T_train=06:58:43 | T_epoch=00:54:38] Epoch 2, batch 2001 / 6666 (step 15332) loss=29206.8 (avg=12.03) [[all losses: diffusion=0.105774 ; kl=2.92064e+10 ; lpips=0.312355 ; repa=0.621941 ; sum_loss=29206.8]]
[T_total=07:09:37 | T_train=07:01:26 | T_epoch=00:57:21] Epoch 2, batch 2101 / 6666 (step 15432) loss=29017.8 (avg=13) [[all losses: diffusion=0.105766 ; kl=2.90174e+10 ; lpips=0.312327 ; repa=0.621759 ; sum_loss=29017.8]]
[T_total=07:12:21 | T_train=07:04:10 | T_epoch=01:00:05] Epoch 2, batch 2201 / 6666 (step 15532) loss=28831 (avg=12.42) [[all losses: diffusion=0.105646 ; kl=2.88305e+10 ; lpips=0.311894 ; repa=0.621379 ; sum_loss=28831]]
[T_total=07:15:05 | T_train=07:06:54 | T_epoch=01:02:49] Epoch 2, batch 2301 / 6666 (step 15632) loss=28646.5 (avg=11.9) [[all losses: diffusion=0.105535 ; kl=2.86461e+10 ; lpips=0.311431 ; repa=0.621001 ; sum_loss=28646.5]]
[T_total=07:17:48 | T_train=07:09:37 | T_epoch=01:05:32] Epoch 2, batch 2401 / 6666 (step 15732) loss=28464.5 (avg=11.42) [[all losses: diffusion=0.105418 ; kl=2.8464e+10 ; lpips=0.311038 ; repa=0.620637 ; sum_loss=28464.5]]
[T_total=07:20:32 | T_train=07:12:21 | T_epoch=01:08:16] Epoch 2, batch 2501 / 6666 (step 15832) loss=28284.7 (avg=10.98) [[all losses: diffusion=0.105304 ; kl=2.82843e+10 ; lpips=0.310592 ; repa=0.620262 ; sum_loss=28284.7]]
[T_total=07:23:15 | T_train=07:15:04 | T_epoch=01:11:00] Epoch 2, batch 2601 / 6666 (step 15932) loss=28107.2 (avg=10.57) [[all losses: diffusion=0.105198 ; kl=2.81068e+10 ; lpips=0.310141 ; repa=0.619892 ; sum_loss=28107.2]]
[T_total=07:25:59 | T_train=07:17:48 | T_epoch=01:13:43] Epoch 2, batch 2701 / 6666 (step 16032) loss=27931.9 (avg=10.19) [[all losses: diffusion=0.105089 ; kl=2.79314e+10 ; lpips=0.309707 ; repa=0.619523 ; sum_loss=27931.9]]
[T_total=07:28:43 | T_train=07:20:32 | T_epoch=01:16:27] Epoch 2, batch 2801 / 6666 (step 16132) loss=27758.7 (avg=9.837) [[all losses: diffusion=0.10498 ; kl=2.77583e+10 ; lpips=0.309278 ; repa=0.619161 ; sum_loss=27758.7]]
[T_total=07:31:26 | T_train=07:23:16 | T_epoch=01:19:11] Epoch 2, batch 2901 / 6666 (step 16232) loss=27587.7 (avg=9.51) [[all losses: diffusion=0.104865 ; kl=2.75873e+10 ; lpips=0.308846 ; repa=0.618791 ; sum_loss=27587.7]]
[T_total=07:34:10 | T_train=07:25:59 | T_epoch=01:21:54] Epoch 2, batch 3001 / 6666 (step 16332) loss=27418.8 (avg=9.205) [[all losses: diffusion=0.104759 ; kl=2.74184e+10 ; lpips=0.30843 ; repa=0.618432 ; sum_loss=27418.8]]
[T_total=07:36:54 | T_train=07:28:43 | T_epoch=01:24:38] Epoch 2, batch 3101 / 6666 (step 16432) loss=27252 (avg=8.919) [[all losses: diffusion=0.104655 ; kl=2.72516e+10 ; lpips=0.308013 ; repa=0.618076 ; sum_loss=27252]]
[T_total=07:39:37 | T_train=07:31:27 | T_epoch=01:27:22] Epoch 2, batch 3201 / 6666 (step 16532) loss=27087.1 (avg=8.652) [[all losses: diffusion=0.104547 ; kl=2.70867e+10 ; lpips=0.307594 ; repa=0.61772 ; sum_loss=27087.1]]
[T_total=07:42:21 | T_train=07:34:10 | T_epoch=01:30:06] Epoch 2, batch 3301 / 6666 (step 16632) loss=26924.3 (avg=8.4) [[all losses: diffusion=0.104438 ; kl=2.69239e+10 ; lpips=0.307188 ; repa=0.617364 ; sum_loss=26924.3]]
[T_total=07:45:05 | T_train=07:36:54 | T_epoch=01:32:49] Epoch 2, batch 3401 / 6666 (step 16732) loss=26763.4 (avg=8.163) [[all losses: diffusion=0.104326 ; kl=2.6763e+10 ; lpips=0.3068 ; repa=0.617011 ; sum_loss=26763.4]]
[T_total=07:47:49 | T_train=07:39:38 | T_epoch=01:35:33] Epoch 2, batch 3501 / 6666 (step 16832) loss=26604.4 (avg=7.94) [[all losses: diffusion=0.104218 ; kl=2.6604e+10 ; lpips=0.306403 ; repa=0.616658 ; sum_loss=26604.4]]
[T_total=07:50:33 | T_train=07:42:22 | T_epoch=01:38:17] Epoch 2, batch 3601 / 6666 (step 16932) loss=26447.3 (avg=7.729) [[all losses: diffusion=0.104122 ; kl=2.64469e+10 ; lpips=0.305991 ; repa=0.616314 ; sum_loss=26447.3]]
[T_total=07:53:16 | T_train=07:45:06 | T_epoch=01:41:01] Epoch 2, batch 3701 / 6666 (step 17032) loss=26292 (avg=7.53) [[all losses: diffusion=0.104025 ; kl=2.62916e+10 ; lpips=0.305587 ; repa=0.615969 ; sum_loss=26292]]
[T_total=07:56:00 | T_train=07:47:49 | T_epoch=01:43:44] Epoch 2, batch 3801 / 6666 (step 17132) loss=26138.6 (avg=7.341) [[all losses: diffusion=0.103937 ; kl=2.61381e+10 ; lpips=0.305215 ; repa=0.615643 ; sum_loss=26138.6]]
[T_total=07:58:44 | T_train=07:50:33 | T_epoch=01:46:28] Epoch 2, batch 3901 / 6666 (step 17232) loss=25986.9 (avg=7.162) [[all losses: diffusion=0.103826 ; kl=2.59865e+10 ; lpips=0.304827 ; repa=0.615298 ; sum_loss=25986.9]]
[T_total=08:01:28 | T_train=07:53:17 | T_epoch=01:49:12] Epoch 2, batch 4001 / 6666 (step 17332) loss=25837 (avg=6.991) [[all losses: diffusion=0.103729 ; kl=2.58365e+10 ; lpips=0.304433 ; repa=0.61496 ; sum_loss=25837]]
[T_total=08:04:12 | T_train=07:56:01 | T_epoch=01:51:56] Epoch 2, batch 4101 / 6666 (step 17432) loss=25688.7 (avg=6.829) [[all losses: diffusion=0.103627 ; kl=2.56883e+10 ; lpips=0.304041 ; repa=0.614624 ; sum_loss=25688.7]]
[T_total=08:06:55 | T_train=07:58:45 | T_epoch=01:54:40] Epoch 2, batch 4201 / 6666 (step 17532) loss=25542.2 (avg=6.675) [[all losses: diffusion=0.10353 ; kl=2.55418e+10 ; lpips=0.303646 ; repa=0.614292 ; sum_loss=25542.2]]
[T_total=08:09:39 | T_train=08:01:28 | T_epoch=01:57:23] Epoch 2, batch 4301 / 6666 (step 17632) loss=25397.4 (avg=6.528) [[all losses: diffusion=0.103431 ; kl=2.5397e+10 ; lpips=0.303254 ; repa=0.613954 ; sum_loss=25397.4]]
[T_total=08:12:23 | T_train=08:04:12 | T_epoch=02:00:07] Epoch 2, batch 4401 / 6666 (step 17732) loss=25254.2 (avg=6.387) [[all losses: diffusion=0.103338 ; kl=2.52538e+10 ; lpips=0.30286 ; repa=0.613624 ; sum_loss=25254.2]]
[T_total=08:15:06 | T_train=08:06:56 | T_epoch=02:02:51] Epoch 2, batch 4501 / 6666 (step 17832) loss=25112.6 (avg=6.253) [[all losses: diffusion=0.103243 ; kl=2.51121e+10 ; lpips=0.302487 ; repa=0.613294 ; sum_loss=25112.6]]
[T_total=08:17:50 | T_train=08:09:39 | T_epoch=02:05:34] Epoch 2, batch 4601 / 6666 (step 17932) loss=24972.5 (avg=6.125) [[all losses: diffusion=0.103148 ; kl=2.49721e+10 ; lpips=0.302112 ; repa=0.612965 ; sum_loss=24972.5]]
[T_total=08:20:34 | T_train=08:12:23 | T_epoch=02:08:18] Epoch 2, batch 4701 / 6666 (step 18032) loss=26009 (avg=4513) [[all losses: diffusion=0.103162 ; kl=2.60086e+10 ; lpips=0.302155 ; repa=0.612841 ; sum_loss=26009]]
[T_total=08:23:18 | T_train=08:15:07 | T_epoch=02:11:02] Epoch 2, batch 4801 / 6666 (step 18132) loss=25865.6 (avg=4419) [[all losses: diffusion=0.103067 ; kl=2.58651e+10 ; lpips=0.301774 ; repa=0.612516 ; sum_loss=25865.6]]
[T_total=08:26:01 | T_train=08:17:51 | T_epoch=02:13:46] Epoch 2, batch 4901 / 6666 (step 18232) loss=25723.7 (avg=4329) [[all losses: diffusion=0.102968 ; kl=2.57233e+10 ; lpips=0.301415 ; repa=0.612195 ; sum_loss=25723.7]]
[T_total=08:28:45 | T_train=08:20:34 | T_epoch=02:16:29] Epoch 2, batch 5001 / 6666 (step 18332) loss=25583.4 (avg=4242) [[all losses: diffusion=0.102878 ; kl=2.5583e+10 ; lpips=0.301042 ; repa=0.611873 ; sum_loss=25583.4]]
[T_total=08:31:29 | T_train=08:23:18 | T_epoch=02:19:13] Epoch 2, batch 5101 / 6666 (step 18432) loss=25444.7 (avg=4160) [[all losses: diffusion=0.102881 ; kl=2.54443e+10 ; lpips=0.300997 ; repa=0.611715 ; sum_loss=25444.7]]
[T_total=08:34:13 | T_train=08:26:02 | T_epoch=02:21:57] Epoch 2, batch 5201 / 6666 (step 18532) loss=25307.4 (avg=4080) [[all losses: diffusion=0.102792 ; kl=2.5307e+10 ; lpips=0.300668 ; repa=0.611408 ; sum_loss=25307.4]]
[T_total=08:36:56 | T_train=08:28:45 | T_epoch=02:24:40] Epoch 2, batch 5301 / 6666 (step 18632) loss=25171.6 (avg=4003) [[all losses: diffusion=0.102695 ; kl=2.51712e+10 ; lpips=0.30032 ; repa=0.611091 ; sum_loss=25171.6]]
[T_total=08:39:40 | T_train=08:31:29 | T_epoch=02:27:24] Epoch 2, batch 5401 / 6666 (step 18732) loss=25037.2 (avg=3929) [[all losses: diffusion=0.102605 ; kl=2.50368e+10 ; lpips=0.299957 ; repa=0.610776 ; sum_loss=25037.2]]
[T_total=08:42:24 | T_train=08:34:13 | T_epoch=02:30:08] Epoch 2, batch 5501 / 6666 (step 18832) loss=24904.3 (avg=3857) [[all losses: diffusion=0.102514 ; kl=2.49039e+10 ; lpips=0.299592 ; repa=0.610462 ; sum_loss=24904.3]]
[T_total=08:45:08 | T_train=08:36:57 | T_epoch=02:32:52] Epoch 2, batch 5601 / 6666 (step 18932) loss=24772.7 (avg=3788) [[all losses: diffusion=0.102431 ; kl=2.47723e+10 ; lpips=0.29923 ; repa=0.610148 ; sum_loss=24772.7]]
[T_total=08:47:51 | T_train=08:39:41 | T_epoch=02:35:36] Epoch 2, batch 5701 / 6666 (step 19032) loss=24642.6 (avg=3722) [[all losses: diffusion=0.102345 ; kl=2.46422e+10 ; lpips=0.298877 ; repa=0.609844 ; sum_loss=24642.6]]
[T_total=08:50:35 | T_train=08:42:24 | T_epoch=02:38:19] Epoch 2, batch 5801 / 6666 (step 19132) loss=24513.8 (avg=3658) [[all losses: diffusion=0.10225 ; kl=2.45134e+10 ; lpips=0.298527 ; repa=0.609535 ; sum_loss=24513.8]]
[T_total=08:53:19 | T_train=08:45:08 | T_epoch=02:41:03] Epoch 2, batch 5901 / 6666 (step 19232) loss=24386.3 (avg=3596) [[all losses: diffusion=0.102167 ; kl=2.43859e+10 ; lpips=0.298166 ; repa=0.609232 ; sum_loss=24386.3]]
[T_total=08:56:03 | T_train=08:47:52 | T_epoch=02:43:47] Epoch 2, batch 6001 / 6666 (step 19332) loss=24260.2 (avg=3536) [[all losses: diffusion=0.102077 ; kl=2.42598e+10 ; lpips=0.297838 ; repa=0.608934 ; sum_loss=24260.2]]
[T_total=08:58:46 | T_train=08:50:35 | T_epoch=02:46:31] Epoch 2, batch 6101 / 6666 (step 19432) loss=24135.4 (avg=3478) [[all losses: diffusion=0.101994 ; kl=2.4135e+10 ; lpips=0.297481 ; repa=0.608631 ; sum_loss=24135.4]]
[T_total=09:01:30 | T_train=08:53:19 | T_epoch=02:49:14] Epoch 2, batch 6201 / 6666 (step 19532) loss=24011.8 (avg=3422) [[all losses: diffusion=0.101909 ; kl=2.40114e+10 ; lpips=0.29713 ; repa=0.608328 ; sum_loss=24011.8]]
[T_total=09:04:14 | T_train=08:56:03 | T_epoch=02:51:58] Epoch 2, batch 6301 / 6666 (step 19632) loss=23889.5 (avg=3368) [[all losses: diffusion=0.101817 ; kl=2.38891e+10 ; lpips=0.296787 ; repa=0.608034 ; sum_loss=23889.5]]
[T_total=09:06:58 | T_train=08:58:47 | T_epoch=02:54:42] Epoch 2, batch 6401 / 6666 (step 19732) loss=23768.4 (avg=3315) [[all losses: diffusion=0.101735 ; kl=2.3768e+10 ; lpips=0.296439 ; repa=0.607739 ; sum_loss=23768.4]]
[T_total=09:09:41 | T_train=09:01:31 | T_epoch=02:57:26] Epoch 2, batch 6501 / 6666 (step 19832) loss=23648.6 (avg=3264) [[all losses: diffusion=0.101654 ; kl=2.36482e+10 ; lpips=0.296104 ; repa=0.607446 ; sum_loss=23648.6]]
[T_total=09:12:25 | T_train=09:04:14 | T_epoch=03:00:09] Epoch 2, batch 6601 / 6666 (step 19932) loss=23530 (avg=3214) [[all losses: diffusion=0.101578 ; kl=2.35296e+10 ; lpips=0.29576 ; repa=0.60715 ; sum_loss=23530]]
[2025-10-25 13:25:33,577][main][INFO] - [T_total=09:14:12 | T_train=09:06:01 | T_epoch=03:01:56] End of epoch 2 (19998 steps) train loss 3183.13[2025-10-25 13:25:33,578][main][INFO] - [Epoch 2] All losses: [[diffusion=0.0880481 ; kl=3.18278e+09 ; lpips=0.243993 ; repa=0.561169]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:01<06:41,  1.54s/it]
Reconstructing from test set:   1%|          | 2/261 [00:02<04:43,  1.10s/it]
Reconstructing from test set:   1%|          | 3/261 [00:03<04:05,  1.05it/s]
Reconstructing from test set:   2%|▏         | 4/261 [00:03<03:47,  1.13it/s]
Reconstructing from test set:   2%|▏         | 5/261 [00:04<03:37,  1.18it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:05<03:31,  1.21it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:06<03:25,  1.23it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:07<03:22,  1.25it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:07<03:20,  1.26it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:08<03:17,  1.27it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:09<03:16,  1.27it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:10<03:15,  1.27it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:10<03:14,  1.27it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:11<03:13,  1.27it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:12<03:13,  1.27it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:13<03:12,  1.27it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:14<03:11,  1.28it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:14<03:10,  1.27it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:15<03:09,  1.28it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:16<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:17<03:07,  1.28it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:17<03:07,  1.28it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:18<03:05,  1.28it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:19<03:05,  1.28it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:20<03:04,  1.28it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:21<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:22<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:23<03:01,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:24<03:01,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:25<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:25<02:59,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:26<02:58,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:27<02:57,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:28<02:56,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:28<02:55,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:29<02:55,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:30<02:54,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:31<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:32<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:32<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:33<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:34<02:50,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:35<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:36<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:37<02:48,  1.27it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:38<02:47,  1.27it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:39<02:46,  1.27it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:39<02:45,  1.27it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:40<02:44,  1.27it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:41<02:44,  1.27it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:42<02:43,  1.27it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:43<02:42,  1.27it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:43<02:42,  1.27it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:44<02:41,  1.27it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:45<02:40,  1.27it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:46<02:39,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:47<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:48<02:36,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:49<02:35,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:50<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:50<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:51<02:33,  1.28it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:52<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:53<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:54<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:54<02:30,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:55<02:29,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:56<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:57<02:28,  1.27it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:57<02:27,  1.27it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:58<02:26,  1.27it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [00:59<02:26,  1.27it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [01:00<02:25,  1.27it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:01<02:24,  1.27it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:01<02:23,  1.27it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:02<02:22,  1.28it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:03<02:22,  1.27it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:04<02:21,  1.27it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:05<02:20,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:05<02:19,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:06<02:18,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:07<02:17,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:08<02:16,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:08<02:16,  1.28it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:09<02:15,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:10<02:14,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:11<02:14,  1.27it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:12<02:13,  1.27it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:12<02:12,  1.27it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:13<02:11,  1.27it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:14<02:11,  1.27it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:15<02:10,  1.27it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:15<02:09,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:16<02:08,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:17<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:18<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:19<02:06,  1.27it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:19<02:05,  1.27it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:20<02:04,  1.27it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:21<02:03,  1.28it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:22<02:03,  1.28it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:23<02:02,  1.28it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:23<02:01,  1.27it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:24<02:00,  1.27it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:25<01:59,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:26<01:58,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:26<01:57,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:27<01:57,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:28<01:57,  1.27it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:29<01:56,  1.27it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:30<01:55,  1.27it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:30<01:54,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:31<01:53,  1.28it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:32<01:52,  1.28it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:33<01:51,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:34<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:34<01:49,  1.28it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:35<01:49,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:36<01:48,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:37<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:37<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:38<01:45,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:39<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:41<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:41<01:43,  1.28it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:42<01:42,  1.28it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:43<01:41,  1.28it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:44<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:44<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:45<01:39,  1.27it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:46<01:38,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:47<01:38,  1.27it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:48<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:48<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:49<01:35,  1.28it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:50<01:34,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:51<01:33,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:51<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:52<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:53<01:31,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:54<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:55<01:29,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:55<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:56<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:57<01:27,  1.27it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:58<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [01:59<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [01:59<01:25,  1.28it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:00<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:01<01:23,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:03<01:21,  1.28it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:04<01:20,  1.28it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:05<01:19,  1.28it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:06<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:06<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:07<01:17,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:08<01:16,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:09<01:16,  1.27it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:09<01:15,  1.27it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:10<01:14,  1.27it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:11<01:13,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:12<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:13<01:12,  1.27it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:13<01:11,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:14<01:10,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:15<01:09,  1.28it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:16<01:08,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:17<01:08,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:17<01:07,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:18<01:06,  1.27it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:19<01:05,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:20<01:05,  1.28it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:20<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:21<01:03,  1.27it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:22<01:02,  1.28it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:23<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:24<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:24<01:00,  1.28it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:25<00:59,  1.27it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:26<00:58,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:27<00:58,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:28<00:57,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:28<00:56,  1.27it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:29<00:55,  1.27it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:30<00:55,  1.27it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:31<00:54,  1.27it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:31<00:53,  1.27it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:32<00:52,  1.27it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:33<00:51,  1.27it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:34<00:50,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:35<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:35<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:36<00:48,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:37<00:47,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:38<00:46,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:38<00:45,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:39<00:45,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:40<00:44,  1.28it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:41<00:43,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:42<00:43,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:42<00:42,  1.28it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:43<00:41,  1.28it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:44<00:40,  1.28it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:45<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:46<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:46<00:38,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:47<00:37,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:49<00:36,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:49<00:35,  1.27it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:50<00:34,  1.28it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:51<00:33,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:53<00:32,  1.28it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:53<00:31,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:54<00:30,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:55<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:56<00:29,  1.27it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:57<00:27,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:58<00:26,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [02:59<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [03:00<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:00<00:24,  1.28it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:01<00:23,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:02<00:22,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:03<00:21,  1.28it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:04<00:21,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:04<00:20,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:05<00:19,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:06<00:18,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:08<00:16,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:09<00:15,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:10<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:11<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:11<00:13,  1.28it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:12<00:12,  1.28it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:13<00:11,  1.28it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:14<00:10,  1.28it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:14<00:10,  1.28it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:15<00:09,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:16<00:08,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:17<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:18<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:18<00:06,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:19<00:05,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:20<00:04,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:21<00:03,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:21<00:03,  1.28it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:22<00:02,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:23<00:01,  1.29it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:25<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:25<00:00,  1.27it/s]
[2025-10-25 13:29:01,907][main][INFO] - [Epoch 3] Test metrics: [[MSE=39.14 | MAE=0.1464 | LPIPS=0.2767 | PSNR=14.07 | SSIM=0.3174 | dreamsim=0.4307 | FID=65.23]][2025-10-25 13:29:01,909][main][INFO] - [Epoch 3] Best metrics: [[min_MSE=39.14 | min_MAE=0.1464 | min_LPIPS=0.2767 | max_PSNR=14.07 | max_SSIM=0.3174 | min_dreamsim=0.4307 | min_FID=65.23]][2025-10-25 13:29:01,910][main][DEBUG] - Writing images to disk...[2025-10-25 13:29:02,747][main][DEBUG] - Image(s) saved on disk[2025-10-25 13:29:02,993][main][INFO] - End of epoch timers: [T_train=09:06:01 | T_epoch=03:01:56 | T_eval=00:10:30 | T_total=09:17:41][2025-10-25 13:29:02,995][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-25 13:29:15,297][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-25 13:29:27,004][main][INFO] - ---

[2025-10-25 13:29:27,025][main][INFO] - [T_total=09:18:05 | T_train=09:06:01] Start epoch 3[T_total=09:18:08 | T_train=09:06:04 | T_epoch=00:00:02] Epoch 3, batch 1 / 6666 (step 19998) loss=23452.3 (avg=0.3464) [[all losses: diffusion=0.101526 ; kl=2.34519e+10 ; lpips=0.295529 ; repa=0.606957 ; sum_loss=23452.3]]
[T_total=09:20:52 | T_train=09:08:47 | T_epoch=00:02:46] Epoch 3, batch 101 / 6666 (step 20098) loss=23335.6 (avg=0.3383) [[all losses: diffusion=0.101441 ; kl=2.33352e+10 ; lpips=0.295199 ; repa=0.606664 ; sum_loss=23335.6]]
[T_total=09:23:36 | T_train=09:11:31 | T_epoch=00:05:30] Epoch 3, batch 201 / 6666 (step 20198) loss=23220.1 (avg=0.3393) [[all losses: diffusion=0.101365 ; kl=2.32197e+10 ; lpips=0.294872 ; repa=0.606379 ; sum_loss=23220.1]]
[T_total=09:26:20 | T_train=09:14:15 | T_epoch=00:08:14] Epoch 3, batch 301 / 6666 (step 20298) loss=23105.7 (avg=0.3384) [[all losses: diffusion=0.101283 ; kl=2.31053e+10 ; lpips=0.294532 ; repa=0.606087 ; sum_loss=23105.7]]
[T_total=09:29:04 | T_train=09:16:59 | T_epoch=00:10:58] Epoch 3, batch 401 / 6666 (step 20398) loss=22992.4 (avg=0.3384) [[all losses: diffusion=0.101208 ; kl=2.2992e+10 ; lpips=0.294198 ; repa=0.605797 ; sum_loss=22992.4]]
[T_total=09:31:47 | T_train=09:19:43 | T_epoch=00:13:41] Epoch 3, batch 501 / 6666 (step 20498) loss=22880.3 (avg=0.3384) [[all losses: diffusion=0.10113 ; kl=2.28799e+10 ; lpips=0.293878 ; repa=0.60551 ; sum_loss=22880.3]]
[T_total=09:34:31 | T_train=09:22:26 | T_epoch=00:16:25] Epoch 3, batch 601 / 6666 (step 20598) loss=22769.2 (avg=0.3381) [[all losses: diffusion=0.10105 ; kl=2.27688e+10 ; lpips=0.293551 ; repa=0.605225 ; sum_loss=22769.2]]
[T_total=09:37:14 | T_train=09:25:10 | T_epoch=00:19:09] Epoch 3, batch 701 / 6666 (step 20698) loss=22659.2 (avg=0.3381) [[all losses: diffusion=0.100979 ; kl=2.26588e+10 ; lpips=0.29322 ; repa=0.604943 ; sum_loss=22659.2]]
[T_total=09:39:58 | T_train=09:27:54 | T_epoch=00:21:52] Epoch 3, batch 801 / 6666 (step 20798) loss=22550.3 (avg=0.3379) [[all losses: diffusion=0.100901 ; kl=2.25499e+10 ; lpips=0.292896 ; repa=0.604662 ; sum_loss=22550.3]]
[T_total=09:42:42 | T_train=09:30:38 | T_epoch=00:24:36] Epoch 3, batch 901 / 6666 (step 20898) loss=22442.4 (avg=0.3378) [[all losses: diffusion=0.100826 ; kl=2.2442e+10 ; lpips=0.292576 ; repa=0.604384 ; sum_loss=22442.4]]
[T_total=09:45:26 | T_train=09:33:22 | T_epoch=00:27:20] Epoch 3, batch 1001 / 6666 (step 20998) loss=22335.5 (avg=0.351) [[all losses: diffusion=0.100769 ; kl=2.23351e+10 ; lpips=0.292328 ; repa=0.604137 ; sum_loss=22335.5]]
[T_total=09:48:10 | T_train=09:36:06 | T_epoch=00:30:04] Epoch 3, batch 1101 / 6666 (step 21098) loss=22229.6 (avg=0.3496) [[all losses: diffusion=0.100693 ; kl=2.22292e+10 ; lpips=0.29201 ; repa=0.603862 ; sum_loss=22229.6]]
[T_total=09:50:54 | T_train=09:38:49 | T_epoch=00:32:48] Epoch 3, batch 1201 / 6666 (step 21198) loss=22124.8 (avg=0.3484) [[all losses: diffusion=0.100618 ; kl=2.21244e+10 ; lpips=0.29169 ; repa=0.603581 ; sum_loss=22124.8]]
[T_total=09:53:38 | T_train=09:41:34 | T_epoch=00:35:32] Epoch 3, batch 1301 / 6666 (step 21298) loss=22020.9 (avg=0.3476) [[all losses: diffusion=0.100549 ; kl=2.20205e+10 ; lpips=0.291382 ; repa=0.603305 ; sum_loss=22020.9]]
[T_total=09:56:22 | T_train=09:44:17 | T_epoch=00:38:16] Epoch 3, batch 1401 / 6666 (step 21398) loss=21918 (avg=0.3467) [[all losses: diffusion=0.100475 ; kl=2.19176e+10 ; lpips=0.291074 ; repa=0.603029 ; sum_loss=21918]]
[T_total=09:59:05 | T_train=09:47:01 | T_epoch=00:40:59] Epoch 3, batch 1501 / 6666 (step 21498) loss=21816 (avg=0.346) [[all losses: diffusion=0.100404 ; kl=2.18157e+10 ; lpips=0.290762 ; repa=0.602754 ; sum_loss=21816]]
[T_total=10:01:49 | T_train=09:49:45 | T_epoch=00:43:43] Epoch 3, batch 1601 / 6666 (step 21598) loss=21715 (avg=0.3453) [[all losses: diffusion=0.100328 ; kl=2.17147e+10 ; lpips=0.290458 ; repa=0.602478 ; sum_loss=21715]]
[T_total=10:04:33 | T_train=09:52:29 | T_epoch=00:46:27] Epoch 3, batch 1701 / 6666 (step 21698) loss=21615 (avg=0.3447) [[all losses: diffusion=0.100253 ; kl=2.16146e+10 ; lpips=0.290159 ; repa=0.602208 ; sum_loss=21615]]
[T_total=10:07:17 | T_train=09:55:12 | T_epoch=00:49:11] Epoch 3, batch 1801 / 6666 (step 21798) loss=21515.8 (avg=0.3441) [[all losses: diffusion=0.100184 ; kl=2.15154e+10 ; lpips=0.289853 ; repa=0.601937 ; sum_loss=21515.8]]
[T_total=10:10:00 | T_train=09:57:56 | T_epoch=00:51:55] Epoch 3, batch 1901 / 6666 (step 21898) loss=21417.6 (avg=0.3435) [[all losses: diffusion=0.100108 ; kl=2.14172e+10 ; lpips=0.289548 ; repa=0.601665 ; sum_loss=21417.6]]
[T_total=10:12:44 | T_train=10:00:40 | T_epoch=00:54:38] Epoch 3, batch 2001 / 6666 (step 21998) loss=21320.2 (avg=0.3431) [[all losses: diffusion=0.100033 ; kl=2.13198e+10 ; lpips=0.289254 ; repa=0.601399 ; sum_loss=21320.2]]
[T_total=10:15:28 | T_train=10:03:24 | T_epoch=00:57:22] Epoch 3, batch 2101 / 6666 (step 22098) loss=21223.7 (avg=0.3427) [[all losses: diffusion=0.0999654 ; kl=2.12233e+10 ; lpips=0.288957 ; repa=0.601136 ; sum_loss=21223.7]]
[T_total=10:18:12 | T_train=10:06:08 | T_epoch=01:00:06] Epoch 3, batch 2201 / 6666 (step 22198) loss=21128.2 (avg=0.547) [[all losses: diffusion=0.0999226 ; kl=2.11278e+10 ; lpips=0.288759 ; repa=0.600925 ; sum_loss=21128.2]]
[T_total=10:20:55 | T_train=10:08:51 | T_epoch=01:02:50] Epoch 3, batch 2301 / 6666 (step 22298) loss=21033.4 (avg=0.5783) [[all losses: diffusion=0.0999335 ; kl=2.1033e+10 ; lpips=0.288791 ; repa=0.60082 ; sum_loss=21033.4]]
[T_total=10:23:39 | T_train=10:11:35 | T_epoch=01:05:33] Epoch 3, batch 2401 / 6666 (step 22398) loss=20939.5 (avg=0.5682) [[all losses: diffusion=0.099863 ; kl=2.09391e+10 ; lpips=0.288522 ; repa=0.600563 ; sum_loss=20939.5]]
[T_total=10:26:22 | T_train=10:14:18 | T_epoch=01:08:17] Epoch 3, batch 2501 / 6666 (step 22498) loss=20846.4 (avg=0.5588) [[all losses: diffusion=0.0997923 ; kl=2.08461e+10 ; lpips=0.288231 ; repa=0.600304 ; sum_loss=20846.4]]
[T_total=10:29:06 | T_train=10:17:02 | T_epoch=01:11:00] Epoch 3, batch 2601 / 6666 (step 22598) loss=20754.2 (avg=0.5501) [[all losses: diffusion=0.0997197 ; kl=2.07538e+10 ; lpips=0.287944 ; repa=0.600046 ; sum_loss=20754.2]]
[T_total=10:31:50 | T_train=10:19:45 | T_epoch=01:13:44] Epoch 3, batch 2701 / 6666 (step 22698) loss=20662.8 (avg=0.5421) [[all losses: diffusion=0.0996489 ; kl=2.06624e+10 ; lpips=0.287652 ; repa=0.599787 ; sum_loss=20662.8]]
[T_total=10:34:33 | T_train=10:22:29 | T_epoch=01:16:28] Epoch 3, batch 2801 / 6666 (step 22798) loss=20572.1 (avg=0.5346) [[all losses: diffusion=0.0995776 ; kl=2.05717e+10 ; lpips=0.287371 ; repa=0.599529 ; sum_loss=20572.1]]
[T_total=10:37:17 | T_train=10:25:13 | T_epoch=01:19:11] Epoch 3, batch 2901 / 6666 (step 22898) loss=20482.3 (avg=0.5276) [[all losses: diffusion=0.0995115 ; kl=2.04819e+10 ; lpips=0.287086 ; repa=0.59927 ; sum_loss=20482.3]]
[T_total=10:40:01 | T_train=10:27:57 | T_epoch=01:21:55] Epoch 3, batch 3001 / 6666 (step 22998) loss=20393.2 (avg=0.5211) [[all losses: diffusion=0.0994451 ; kl=2.03929e+10 ; lpips=0.2868 ; repa=0.599016 ; sum_loss=20393.2]]
[T_total=10:42:45 | T_train=10:30:41 | T_epoch=01:24:39] Epoch 3, batch 3101 / 6666 (step 23098) loss=20305 (avg=0.515) [[all losses: diffusion=0.0993834 ; kl=2.03046e+10 ; lpips=0.286509 ; repa=0.598762 ; sum_loss=20305]]
[T_total=10:45:29 | T_train=10:33:24 | T_epoch=01:27:23] Epoch 3, batch 3201 / 6666 (step 23198) loss=20217.4 (avg=0.5092) [[all losses: diffusion=0.0993114 ; kl=2.0217e+10 ; lpips=0.286221 ; repa=0.59851 ; sum_loss=20217.4]]
[T_total=10:48:13 | T_train=10:36:08 | T_epoch=01:30:07] Epoch 3, batch 3301 / 6666 (step 23298) loss=20130.7 (avg=0.5038) [[all losses: diffusion=0.0992415 ; kl=2.01303e+10 ; lpips=0.285937 ; repa=0.598257 ; sum_loss=20130.7]]
[T_total=10:50:56 | T_train=10:38:52 | T_epoch=01:32:50] Epoch 3, batch 3401 / 6666 (step 23398) loss=20044.6 (avg=0.4987) [[all losses: diffusion=0.0991755 ; kl=2.00442e+10 ; lpips=0.285652 ; repa=0.598007 ; sum_loss=20044.6]]
[T_total=10:53:40 | T_train=10:41:36 | T_epoch=01:35:34] Epoch 3, batch 3501 / 6666 (step 23498) loss=19959.3 (avg=0.4939) [[all losses: diffusion=0.0991079 ; kl=1.99589e+10 ; lpips=0.285383 ; repa=0.59776 ; sum_loss=19959.3]]
[T_total=10:56:24 | T_train=10:44:19 | T_epoch=01:38:18] Epoch 3, batch 3601 / 6666 (step 23598) loss=19874.8 (avg=0.4894) [[all losses: diffusion=0.0990401 ; kl=1.98744e+10 ; lpips=0.285112 ; repa=0.597507 ; sum_loss=19874.8]]
[T_total=10:59:08 | T_train=10:47:03 | T_epoch=01:41:02] Epoch 3, batch 3701 / 6666 (step 23698) loss=19790.9 (avg=0.4851) [[all losses: diffusion=0.0989743 ; kl=1.97905e+10 ; lpips=0.284841 ; repa=0.59726 ; sum_loss=19790.9]]
[T_total=11:01:51 | T_train=10:49:47 | T_epoch=01:43:45] Epoch 3, batch 3801 / 6666 (step 23798) loss=19707.7 (avg=0.4956) [[all losses: diffusion=0.0989439 ; kl=1.97074e+10 ; lpips=0.284694 ; repa=0.597079 ; sum_loss=19707.7]]
[T_total=11:04:35 | T_train=10:52:30 | T_epoch=01:46:29] Epoch 3, batch 3901 / 6666 (step 23898) loss=19625.3 (avg=0.4913) [[all losses: diffusion=0.0988834 ; kl=1.96249e+10 ; lpips=0.284416 ; repa=0.596836 ; sum_loss=19625.3]]
[T_total=11:07:18 | T_train=10:55:14 | T_epoch=01:49:12] Epoch 3, batch 4001 / 6666 (step 23998) loss=19543.5 (avg=0.4873) [[all losses: diffusion=0.0988156 ; kl=1.95431e+10 ; lpips=0.284141 ; repa=0.596591 ; sum_loss=19543.5]]
[T_total=11:10:02 | T_train=10:57:58 | T_epoch=01:51:57] Epoch 3, batch 4101 / 6666 (step 24098) loss=19462.4 (avg=0.4835) [[all losses: diffusion=0.0987598 ; kl=1.9462e+10 ; lpips=0.283874 ; repa=0.596352 ; sum_loss=19462.4]]
[T_total=11:12:46 | T_train=11:00:42 | T_epoch=01:54:40] Epoch 3, batch 4201 / 6666 (step 24198) loss=19382 (avg=0.7854) [[all losses: diffusion=0.0987129 ; kl=1.93817e+10 ; lpips=0.283633 ; repa=0.596125 ; sum_loss=19382]]
[T_total=11:15:30 | T_train=11:03:26 | T_epoch=01:57:24] Epoch 3, batch 4301 / 6666 (step 24298) loss=19302.3 (avg=0.7748) [[all losses: diffusion=0.0986482 ; kl=1.93019e+10 ; lpips=0.283365 ; repa=0.595885 ; sum_loss=19302.3]]
[T_total=11:18:14 | T_train=11:06:09 | T_epoch=02:00:08] Epoch 3, batch 4401 / 6666 (step 24398) loss=19223.2 (avg=0.7647) [[all losses: diffusion=0.0985893 ; kl=1.92228e+10 ; lpips=0.283094 ; repa=0.595646 ; sum_loss=19223.2]]
[T_total=11:20:57 | T_train=11:08:53 | T_epoch=02:02:52] Epoch 3, batch 4501 / 6666 (step 24498) loss=19144.7 (avg=0.755) [[all losses: diffusion=0.0985266 ; kl=1.91443e+10 ; lpips=0.282828 ; repa=0.595402 ; sum_loss=19144.7]]
[T_total=11:23:41 | T_train=11:11:37 | T_epoch=02:05:35] Epoch 3, batch 4601 / 6666 (step 24598) loss=19066.9 (avg=0.7458) [[all losses: diffusion=0.0984721 ; kl=1.90665e+10 ; lpips=0.282563 ; repa=0.595167 ; sum_loss=19066.9]]
[T_total=11:26:25 | T_train=11:14:20 | T_epoch=02:08:19] Epoch 3, batch 4701 / 6666 (step 24698) loss=18989.7 (avg=0.7369) [[all losses: diffusion=0.0984089 ; kl=1.89893e+10 ; lpips=0.282303 ; repa=0.59493 ; sum_loss=18989.7]]
[T_total=11:29:08 | T_train=11:17:04 | T_epoch=02:11:03] Epoch 3, batch 4801 / 6666 (step 24798) loss=18913.1 (avg=0.7284) [[all losses: diffusion=0.0983518 ; kl=1.89127e+10 ; lpips=0.282035 ; repa=0.594695 ; sum_loss=18913.1]]
[T_total=11:31:52 | T_train=11:19:48 | T_epoch=02:13:46] Epoch 3, batch 4901 / 6666 (step 24898) loss=18837.2 (avg=0.7202) [[all losses: diffusion=0.0982917 ; kl=1.88368e+10 ; lpips=0.28178 ; repa=0.594462 ; sum_loss=18837.2]]
[T_total=11:34:36 | T_train=11:22:31 | T_epoch=02:16:30] Epoch 3, batch 5001 / 6666 (step 24998) loss=18761.8 (avg=0.7124) [[all losses: diffusion=0.0982289 ; kl=1.87614e+10 ; lpips=0.281528 ; repa=0.59423 ; sum_loss=18761.8]]
[T_total=11:37:20 | T_train=11:25:15 | T_epoch=02:19:14] Epoch 3, batch 5101 / 6666 (step 25098) loss=18687.1 (avg=0.7049) [[all losses: diffusion=0.0981701 ; kl=1.86867e+10 ; lpips=0.281271 ; repa=0.593998 ; sum_loss=18687.1]]
[T_total=11:40:03 | T_train=11:27:59 | T_epoch=02:21:58] Epoch 3, batch 5201 / 6666 (step 25198) loss=19658.9 (avg=5069) [[all losses: diffusion=0.0981981 ; kl=1.96586e+10 ; lpips=0.281312 ; repa=0.593926 ; sum_loss=19658.9]]
[T_total=11:42:47 | T_train=11:30:43 | T_epoch=02:24:41] Epoch 3, batch 5301 / 6666 (step 25298) loss=19581.5 (avg=4975) [[all losses: diffusion=0.0982617 ; kl=1.95811e+10 ; lpips=0.281473 ; repa=0.593912 ; sum_loss=19581.5]]
[T_total=11:45:31 | T_train=11:33:26 | T_epoch=02:27:25] Epoch 3, batch 5401 / 6666 (step 25398) loss=19504.4 (avg=4882) [[all losses: diffusion=0.0982029 ; kl=1.9504e+10 ; lpips=0.281229 ; repa=0.593683 ; sum_loss=19504.4]]
[T_total=11:48:14 | T_train=11:36:10 | T_epoch=02:30:08] Epoch 3, batch 5501 / 6666 (step 25498) loss=19427.9 (avg=4794) [[all losses: diffusion=0.0981459 ; kl=1.94275e+10 ; lpips=0.28097 ; repa=0.593447 ; sum_loss=19427.9]]
[T_total=11:50:58 | T_train=11:38:54 | T_epoch=02:32:53] Epoch 3, batch 5601 / 6666 (step 25598) loss=19352 (avg=4708) [[all losses: diffusion=0.098091 ; kl=1.93517e+10 ; lpips=0.280718 ; repa=0.593218 ; sum_loss=19352]]
[T_total=11:53:42 | T_train=11:41:38 | T_epoch=02:35:36] Epoch 3, batch 5701 / 6666 (step 25698) loss=19276.7 (avg=4626) [[all losses: diffusion=0.0980355 ; kl=1.92763e+10 ; lpips=0.280466 ; repa=0.592992 ; sum_loss=19276.7]]
[T_total=11:56:26 | T_train=11:44:21 | T_epoch=02:38:20] Epoch 3, batch 5801 / 6666 (step 25798) loss=19202 (avg=4546) [[all losses: diffusion=0.0979766 ; kl=1.92016e+10 ; lpips=0.280215 ; repa=0.592764 ; sum_loss=19202]]
[T_total=11:59:09 | T_train=11:47:05 | T_epoch=02:41:04] Epoch 3, batch 5901 / 6666 (step 25898) loss=19127.9 (avg=4469) [[all losses: diffusion=0.0979206 ; kl=1.91275e+10 ; lpips=0.279962 ; repa=0.592537 ; sum_loss=19127.9]]
[T_total=12:01:53 | T_train=11:49:49 | T_epoch=02:43:47] Epoch 3, batch 6001 / 6666 (step 25998) loss=19054.3 (avg=4394) [[all losses: diffusion=0.0978692 ; kl=1.90539e+10 ; lpips=0.279712 ; repa=0.592312 ; sum_loss=19054.3]]
[T_total=12:04:37 | T_train=11:52:33 | T_epoch=02:46:31] Epoch 3, batch 6101 / 6666 (step 26098) loss=18981.3 (avg=4322) [[all losses: diffusion=0.097807 ; kl=1.89809e+10 ; lpips=0.279467 ; repa=0.592086 ; sum_loss=18981.3]]
[T_total=12:07:21 | T_train=11:55:16 | T_epoch=02:49:15] Epoch 3, batch 6201 / 6666 (step 26198) loss=18908.9 (avg=4253) [[all losses: diffusion=0.0977478 ; kl=1.89085e+10 ; lpips=0.279228 ; repa=0.591866 ; sum_loss=18908.9]]
[T_total=12:10:04 | T_train=11:58:00 | T_epoch=02:51:58] Epoch 3, batch 6301 / 6666 (step 26298) loss=18837 (avg=4185) [[all losses: diffusion=0.09769 ; kl=1.88366e+10 ; lpips=0.278991 ; repa=0.591645 ; sum_loss=18837]]
[T_total=12:12:48 | T_train=12:00:44 | T_epoch=02:54:42] Epoch 3, batch 6401 / 6666 (step 26398) loss=18765.6 (avg=4120) [[all losses: diffusion=0.0976357 ; kl=1.87652e+10 ; lpips=0.278745 ; repa=0.591424 ; sum_loss=18765.6]]
[T_total=12:15:32 | T_train=12:03:27 | T_epoch=02:57:26] Epoch 3, batch 6501 / 6666 (step 26498) loss=18694.8 (avg=4056) [[all losses: diffusion=0.0975812 ; kl=1.86944e+10 ; lpips=0.278505 ; repa=0.591206 ; sum_loss=18694.8]]
[T_total=12:18:15 | T_train=12:06:11 | T_epoch=03:00:09] Epoch 3, batch 6601 / 6666 (step 26598) loss=18624.5 (avg=3995) [[all losses: diffusion=0.0975246 ; kl=1.86241e+10 ; lpips=0.278261 ; repa=0.590985 ; sum_loss=18624.5]]
[2025-10-25 16:31:23,501][main][INFO] - [T_total=12:20:02 | T_train=12:07:58 | T_epoch=03:01:56] End of epoch 3 (26664 steps) train loss 3955.96[2025-10-25 16:31:23,503][main][INFO] - [Epoch 3] All losses: [[diffusion=0.0853818 ; kl=3.95562e+09 ; lpips=0.225845 ; repa=0.542498]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:01<06:47,  1.57s/it]
Reconstructing from test set:   1%|          | 2/261 [00:02<04:45,  1.10s/it]
Reconstructing from test set:   1%|          | 3/261 [00:03<04:06,  1.05it/s]
Reconstructing from test set:   2%|▏         | 4/261 [00:03<03:47,  1.13it/s]
Reconstructing from test set:   2%|▏         | 5/261 [00:04<03:36,  1.18it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:05<03:30,  1.21it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:06<03:26,  1.23it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:07<03:23,  1.25it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:07<03:20,  1.26it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:08<03:18,  1.26it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:09<03:17,  1.27it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:10<03:15,  1.27it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:10<03:15,  1.27it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:11<03:13,  1.27it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:12<03:12,  1.28it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:13<03:11,  1.28it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:14<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:14<03:09,  1.28it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:15<03:08,  1.29it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:16<03:07,  1.29it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:17<03:06,  1.29it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:17<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:18<03:05,  1.28it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:19<03:04,  1.28it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:20<03:04,  1.28it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:22<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:23<03:01,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:24<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:24<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:25<02:59,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:26<02:58,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:27<02:57,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:28<02:56,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:28<02:55,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:29<02:55,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:30<02:54,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:31<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:32<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:32<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:33<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:34<02:50,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:35<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:36<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:37<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:38<02:46,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:39<02:46,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:39<02:45,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:40<02:44,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:41<02:44,  1.27it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:42<02:43,  1.27it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:42<02:42,  1.27it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:43<02:41,  1.27it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:44<02:40,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:45<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:46<02:39,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:47<02:37,  1.27it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:48<02:37,  1.27it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:49<02:35,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:50<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:50<02:34,  1.27it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:51<02:33,  1.28it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:52<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:53<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:53<02:30,  1.28it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:54<02:29,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:55<02:28,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:56<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:57<02:27,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:57<02:27,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:58<02:26,  1.28it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [00:59<02:25,  1.28it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [01:00<02:24,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:00<02:24,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:01<02:23,  1.28it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:02<02:22,  1.27it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:03<02:22,  1.27it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:04<02:21,  1.28it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:04<02:20,  1.27it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:05<02:20,  1.27it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:06<02:19,  1.27it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:07<02:18,  1.27it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:08<02:17,  1.27it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:08<02:16,  1.28it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:09<02:15,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:10<02:14,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:11<02:13,  1.28it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:11<02:12,  1.28it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:12<02:11,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:13<02:11,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:14<02:10,  1.28it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:15<02:09,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:15<02:08,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:16<02:08,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:17<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:18<02:06,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:18<02:06,  1.27it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:19<02:05,  1.27it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:20<02:04,  1.27it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:21<02:04,  1.27it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:22<02:03,  1.27it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:22<02:02,  1.27it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:23<02:02,  1.27it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:24<02:01,  1.27it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:25<02:00,  1.27it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:26<01:59,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:26<01:58,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:27<01:57,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:28<01:56,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:29<01:55,  1.28it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:29<01:55,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:30<01:54,  1.27it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:31<01:53,  1.28it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:32<01:52,  1.28it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:33<01:51,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:33<01:51,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:34<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:35<01:49,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:36<01:48,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:37<01:48,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:37<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:38<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:39<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:41<01:43,  1.27it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:42<01:42,  1.27it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:43<01:41,  1.28it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:44<01:41,  1.27it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:44<01:40,  1.27it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:45<01:40,  1.27it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:46<01:39,  1.27it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:47<01:38,  1.27it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:48<01:37,  1.27it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:48<01:36,  1.27it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:49<01:35,  1.27it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:50<01:34,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:51<01:33,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:51<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:52<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:53<01:31,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:54<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:55<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:55<01:29,  1.27it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:56<01:28,  1.27it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:57<01:27,  1.27it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:58<01:27,  1.27it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [01:59<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [01:59<01:25,  1.28it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:00<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:01<01:23,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:02<01:23,  1.27it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:02<01:22,  1.27it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:03<01:21,  1.27it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:04<01:21,  1.27it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:05<01:20,  1.27it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:06<01:19,  1.27it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:06<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:07<01:17,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:08<01:16,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:09<01:16,  1.27it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:10<01:15,  1.27it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:10<01:14,  1.27it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:11<01:13,  1.27it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:12<01:13,  1.27it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:13<01:12,  1.27it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:13<01:11,  1.27it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:14<01:10,  1.27it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:15<01:09,  1.27it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:16<01:09,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:17<01:08,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:17<01:07,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:18<01:06,  1.27it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:19<01:05,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:20<01:05,  1.28it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:20<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:21<01:03,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:22<01:02,  1.28it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:23<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:24<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:24<01:00,  1.28it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:25<00:59,  1.27it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:26<00:58,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:27<00:58,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:28<00:57,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:28<00:56,  1.27it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:29<00:55,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:30<00:54,  1.28it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:31<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:31<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:32<00:52,  1.28it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:33<00:51,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:34<00:50,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:35<00:50,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:35<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:36<00:48,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:37<00:47,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:38<00:46,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:38<00:46,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:39<00:45,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:40<00:44,  1.28it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:41<00:43,  1.27it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:42<00:43,  1.27it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:42<00:42,  1.28it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:43<00:41,  1.28it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:44<00:40,  1.28it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:45<00:40,  1.27it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:46<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:46<00:38,  1.27it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:47<00:37,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:49<00:36,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:49<00:35,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:50<00:34,  1.28it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:51<00:33,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:53<00:32,  1.28it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:53<00:31,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:54<00:30,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:55<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:57<00:27,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:58<00:26,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [02:59<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [03:00<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:00<00:24,  1.28it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:01<00:23,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:02<00:22,  1.29it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:03<00:21,  1.29it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:04<00:21,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:04<00:20,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:05<00:19,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:06<00:18,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:08<00:16,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:09<00:15,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:10<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:11<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:11<00:13,  1.29it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:12<00:12,  1.29it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:13<00:11,  1.29it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:14<00:10,  1.28it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:14<00:10,  1.28it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:15<00:09,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:16<00:08,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:17<00:07,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:18<00:07,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:18<00:06,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:19<00:05,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:20<00:04,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:21<00:03,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:21<00:03,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:22<00:02,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:23<00:01,  1.29it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:25<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:25<00:00,  1.27it/s]
[2025-10-25 16:34:51,652][main][INFO] - [Epoch 4] Test metrics: [[MSE=33.1 | MAE=0.1326 | LPIPS=0.2336 | PSNR=14.8 | SSIM=0.3397 | dreamsim=0.3658 | FID=48.75]][2025-10-25 16:34:51,653][main][INFO] - [Epoch 4] Best metrics: [[min_MSE=33.1 | min_MAE=0.1326 | min_LPIPS=0.2336 | max_PSNR=14.8 | max_SSIM=0.3397 | min_dreamsim=0.3658 | min_FID=48.75]][2025-10-25 16:34:51,655][main][DEBUG] - Writing images to disk...[2025-10-25 16:34:52,487][main][DEBUG] - Image(s) saved on disk[2025-10-25 16:34:52,740][main][INFO] - End of epoch timers: [T_train=12:07:58 | T_epoch=03:01:56 | T_eval=00:13:59 | T_total=12:23:31][2025-10-25 16:34:52,742][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-25 16:35:04,325][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-25 16:35:16,162][main][INFO] - ---

[2025-10-25 16:35:16,163][main][INFO] - [T_total=12:23:55 | T_train=12:07:58] Start epoch 4[T_total=12:23:57 | T_train=12:08:00 | T_epoch=00:00:02] Epoch 4, batch 1 / 6666 (step 26664) loss=18578.4 (avg=0.3283) [[all losses: diffusion=0.0974898 ; kl=1.8578e+10 ; lpips=0.278108 ; repa=0.590842 ; sum_loss=18578.4]]
[T_total=12:26:41 | T_train=12:10:44 | T_epoch=00:02:46] Epoch 4, batch 101 / 6666 (step 26764) loss=18509 (avg=0.3254) [[all losses: diffusion=0.0974369 ; kl=1.85086e+10 ; lpips=0.277869 ; repa=0.590622 ; sum_loss=18509]]
[T_total=12:29:25 | T_train=12:13:28 | T_epoch=00:05:30] Epoch 4, batch 201 / 6666 (step 26864) loss=18440.1 (avg=0.326) [[all losses: diffusion=0.0973851 ; kl=1.84397e+10 ; lpips=0.277636 ; repa=0.590407 ; sum_loss=18440.1]]
[T_total=12:32:09 | T_train=12:16:12 | T_epoch=00:08:14] Epoch 4, batch 301 / 6666 (step 26964) loss=18371.7 (avg=0.3254) [[all losses: diffusion=0.0973286 ; kl=1.83713e+10 ; lpips=0.2774 ; repa=0.590189 ; sum_loss=18371.7]]
[T_total=12:34:52 | T_train=12:18:55 | T_epoch=00:10:57] Epoch 4, batch 401 / 6666 (step 27064) loss=18303.8 (avg=0.3254) [[all losses: diffusion=0.0972777 ; kl=1.83035e+10 ; lpips=0.277165 ; repa=0.589972 ; sum_loss=18303.8]]
[T_total=12:37:36 | T_train=12:21:39 | T_epoch=00:13:41] Epoch 4, batch 501 / 6666 (step 27164) loss=18236.5 (avg=0.3253) [[all losses: diffusion=0.0972251 ; kl=1.82361e+10 ; lpips=0.276931 ; repa=0.589758 ; sum_loss=18236.5]]
[T_total=12:40:19 | T_train=12:24:22 | T_epoch=00:16:24] Epoch 4, batch 601 / 6666 (step 27264) loss=18169.6 (avg=0.3251) [[all losses: diffusion=0.0971728 ; kl=1.81692e+10 ; lpips=0.276695 ; repa=0.589544 ; sum_loss=18169.6]]
[T_total=12:43:03 | T_train=12:27:06 | T_epoch=00:19:08] Epoch 4, batch 701 / 6666 (step 27364) loss=18103.2 (avg=0.325) [[all losses: diffusion=0.0971202 ; kl=1.81028e+10 ; lpips=0.276464 ; repa=0.589332 ; sum_loss=18103.2]]
[T_total=12:45:47 | T_train=12:29:50 | T_epoch=00:21:52] Epoch 4, batch 801 / 6666 (step 27464) loss=18037.3 (avg=0.325) [[all losses: diffusion=0.0970698 ; kl=1.80369e+10 ; lpips=0.276233 ; repa=0.58912 ; sum_loss=18037.3]]
[T_total=12:48:30 | T_train=12:32:33 | T_epoch=00:24:35] Epoch 4, batch 901 / 6666 (step 27564) loss=17971.8 (avg=0.3247) [[all losses: diffusion=0.0970146 ; kl=1.79714e+10 ; lpips=0.276 ; repa=0.588909 ; sum_loss=17971.8]]
[T_total=12:51:14 | T_train=12:35:17 | T_epoch=00:27:19] Epoch 4, batch 1001 / 6666 (step 27664) loss=17906.9 (avg=0.3455) [[all losses: diffusion=0.0969929 ; kl=1.79065e+10 ; lpips=0.275859 ; repa=0.588747 ; sum_loss=17906.9]]
[T_total=12:53:58 | T_train=12:38:01 | T_epoch=00:30:03] Epoch 4, batch 1101 / 6666 (step 27764) loss=17846.5 (avg=103.5) [[all losses: diffusion=0.0970048 ; kl=1.78461e+10 ; lpips=0.2759 ; repa=0.588674 ; sum_loss=17846.5]]
[T_total=12:56:41 | T_train=12:40:44 | T_epoch=00:32:46] Epoch 4, batch 1201 / 6666 (step 27864) loss=17782.4 (avg=94.93) [[all losses: diffusion=0.0969509 ; kl=1.7782e+10 ; lpips=0.275676 ; repa=0.588466 ; sum_loss=17782.4]]
[T_total=12:59:25 | T_train=12:43:28 | T_epoch=00:35:30] Epoch 4, batch 1301 / 6666 (step 27964) loss=17718.8 (avg=87.66) [[all losses: diffusion=0.0969047 ; kl=1.77185e+10 ; lpips=0.275446 ; repa=0.588253 ; sum_loss=17718.8]]
[T_total=13:02:09 | T_train=12:46:12 | T_epoch=00:38:14] Epoch 4, batch 1401 / 6666 (step 28064) loss=17655.7 (avg=81.42) [[all losses: diffusion=0.096859 ; kl=1.76553e+10 ; lpips=0.275222 ; repa=0.588051 ; sum_loss=17655.7]]
[T_total=13:04:53 | T_train=12:48:56 | T_epoch=00:40:58] Epoch 4, batch 1501 / 6666 (step 28164) loss=17593 (avg=76.02) [[all losses: diffusion=0.0968138 ; kl=1.75926e+10 ; lpips=0.27499 ; repa=0.587837 ; sum_loss=17593]]
[T_total=13:07:36 | T_train=12:51:39 | T_epoch=00:43:41] Epoch 4, batch 1601 / 6666 (step 28264) loss=17530.8 (avg=71.29) [[all losses: diffusion=0.0967669 ; kl=1.75304e+10 ; lpips=0.274763 ; repa=0.58763 ; sum_loss=17530.8]]
[T_total=13:10:20 | T_train=12:54:23 | T_epoch=00:46:25] Epoch 4, batch 1701 / 6666 (step 28364) loss=17469.4 (avg=74.71) [[all losses: diffusion=0.0968381 ; kl=1.7469e+10 ; lpips=0.274959 ; repa=0.587653 ; sum_loss=17469.4]]
[T_total=13:13:04 | T_train=12:57:07 | T_epoch=00:49:08] Epoch 4, batch 1801 / 6666 (step 28464) loss=17408.1 (avg=70.59) [[all losses: diffusion=0.0968098 ; kl=1.74077e+10 ; lpips=0.274858 ; repa=0.587503 ; sum_loss=17408.1]]
[T_total=13:15:47 | T_train=12:59:50 | T_epoch=00:51:52] Epoch 4, batch 1901 / 6666 (step 28564) loss=17383.3 (avg=610.5) [[all losses: diffusion=0.0967743 ; kl=1.73829e+10 ; lpips=0.27468 ; repa=0.587325 ; sum_loss=17383.3]]
[T_total=13:18:31 | T_train=13:02:34 | T_epoch=00:54:36] Epoch 4, batch 2001 / 6666 (step 28664) loss=17322.6 (avg=580) [[all losses: diffusion=0.096729 ; kl=1.73223e+10 ; lpips=0.27446 ; repa=0.587124 ; sum_loss=17322.6]]
[T_total=13:21:15 | T_train=13:05:18 | T_epoch=00:57:20] Epoch 4, batch 2101 / 6666 (step 28764) loss=17262.4 (avg=552.4) [[all losses: diffusion=0.0966782 ; kl=1.7262e+10 ; lpips=0.274241 ; repa=0.58692 ; sum_loss=17262.4]]
[T_total=13:23:59 | T_train=13:08:02 | T_epoch=01:00:04] Epoch 4, batch 2201 / 6666 (step 28864) loss=17202.6 (avg=527.3) [[all losses: diffusion=0.0966282 ; kl=1.72022e+10 ; lpips=0.274022 ; repa=0.586717 ; sum_loss=17202.6]]
[T_total=13:26:42 | T_train=13:10:45 | T_epoch=01:02:47] Epoch 4, batch 2301 / 6666 (step 28964) loss=17143.2 (avg=504.4) [[all losses: diffusion=0.0965869 ; kl=1.71429e+10 ; lpips=0.273797 ; repa=0.586518 ; sum_loss=17143.2]]
[T_total=13:29:26 | T_train=13:13:29 | T_epoch=01:05:31] Epoch 4, batch 2401 / 6666 (step 29064) loss=17084.3 (avg=483.4) [[all losses: diffusion=0.0965401 ; kl=1.70839e+10 ; lpips=0.273571 ; repa=0.586314 ; sum_loss=17084.3]]
[T_total=13:32:10 | T_train=13:16:13 | T_epoch=01:08:15] Epoch 4, batch 2501 / 6666 (step 29164) loss=17041.9 (avg=653.6) [[all losses: diffusion=0.0965349 ; kl=1.70415e+10 ; lpips=0.273501 ; repa=0.586194 ; sum_loss=17041.9]]
[T_total=13:34:54 | T_train=13:18:57 | T_epoch=01:10:59] Epoch 4, batch 2601 / 6666 (step 29264) loss=16983.7 (avg=628.5) [[all losses: diffusion=0.0964839 ; kl=1.69833e+10 ; lpips=0.273301 ; repa=0.585996 ; sum_loss=16983.7]]
[T_total=13:37:37 | T_train=13:21:40 | T_epoch=01:13:42] Epoch 4, batch 2701 / 6666 (step 29364) loss=16925.9 (avg=605.4) [[all losses: diffusion=0.0964492 ; kl=1.69255e+10 ; lpips=0.27314 ; repa=0.585826 ; sum_loss=16925.9]]
[T_total=13:40:21 | T_train=13:24:24 | T_epoch=01:16:26] Epoch 4, batch 2801 / 6666 (step 29464) loss=16868.4 (avg=583.8) [[all losses: diffusion=0.0964131 ; kl=1.68681e+10 ; lpips=0.272957 ; repa=0.585644 ; sum_loss=16868.4]]
[T_total=13:43:05 | T_train=13:27:08 | T_epoch=01:19:10] Epoch 4, batch 2901 / 6666 (step 29564) loss=16811.4 (avg=563.7) [[all losses: diffusion=0.096367 ; kl=1.6811e+10 ; lpips=0.272742 ; repa=0.585445 ; sum_loss=16811.4]]
[T_total=13:45:48 | T_train=13:29:51 | T_epoch=01:21:53] Epoch 4, batch 3001 / 6666 (step 29664) loss=16754.7 (avg=544.9) [[all losses: diffusion=0.0963206 ; kl=1.67543e+10 ; lpips=0.272527 ; repa=0.585246 ; sum_loss=16754.7]]
[T_total=13:48:32 | T_train=13:32:35 | T_epoch=01:24:37] Epoch 4, batch 3101 / 6666 (step 29764) loss=16698.4 (avg=527.4) [[all losses: diffusion=0.0962685 ; kl=1.6698e+10 ; lpips=0.272324 ; repa=0.585051 ; sum_loss=16698.4]]
[T_total=13:51:16 | T_train=13:35:19 | T_epoch=01:27:21] Epoch 4, batch 3201 / 6666 (step 29864) loss=16642.5 (avg=510.9) [[all losses: diffusion=0.0962198 ; kl=1.66421e+10 ; lpips=0.272116 ; repa=0.584854 ; sum_loss=16642.5]]
[T_total=13:53:59 | T_train=13:38:03 | T_epoch=01:30:04] Epoch 4, batch 3301 / 6666 (step 29964) loss=16587 (avg=496) [[all losses: diffusion=0.0962283 ; kl=1.65867e+10 ; lpips=0.272102 ; repa=0.584766 ; sum_loss=16587]]
[T_total=13:56:43 | T_train=13:40:46 | T_epoch=01:32:48] Epoch 4, batch 3401 / 6666 (step 30064) loss=16531.9 (avg=481.4) [[all losses: diffusion=0.096184 ; kl=1.65315e+10 ; lpips=0.271901 ; repa=0.584577 ; sum_loss=16531.9]]
[T_total=13:59:27 | T_train=13:43:30 | T_epoch=01:35:32] Epoch 4, batch 3501 / 6666 (step 30164) loss=16477.1 (avg=467.7) [[all losses: diffusion=0.0961408 ; kl=1.64767e+10 ; lpips=0.271694 ; repa=0.584385 ; sum_loss=16477.1]]
[T_total=14:02:11 | T_train=13:46:14 | T_epoch=01:38:16] Epoch 4, batch 3601 / 6666 (step 30264) loss=16422.6 (avg=454.7) [[all losses: diffusion=0.0960951 ; kl=1.64222e+10 ; lpips=0.271484 ; repa=0.584189 ; sum_loss=16422.6]]
[T_total=14:04:54 | T_train=13:48:57 | T_epoch=01:40:59] Epoch 4, batch 3701 / 6666 (step 30364) loss=16368.5 (avg=442.4) [[all losses: diffusion=0.0960523 ; kl=1.63682e+10 ; lpips=0.271277 ; repa=0.584 ; sum_loss=16368.5]]
[T_total=14:07:38 | T_train=13:51:41 | T_epoch=01:43:43] Epoch 4, batch 3801 / 6666 (step 30464) loss=16314.8 (avg=430.8) [[all losses: diffusion=0.0960068 ; kl=1.63144e+10 ; lpips=0.271077 ; repa=0.583806 ; sum_loss=16314.8]]
[T_total=14:10:22 | T_train=13:54:25 | T_epoch=01:46:27] Epoch 4, batch 3901 / 6666 (step 30564) loss=16261.4 (avg=419.8) [[all losses: diffusion=0.0959667 ; kl=1.62611e+10 ; lpips=0.270869 ; repa=0.583613 ; sum_loss=16261.4]]
[T_total=14:13:05 | T_train=13:57:08 | T_epoch=01:49:10] Epoch 4, batch 4001 / 6666 (step 30664) loss=16225.3 (avg=539.1) [[all losses: diffusion=0.0959924 ; kl=1.6225e+10 ; lpips=0.270914 ; repa=0.583555 ; sum_loss=16225.3]]
[T_total=14:15:49 | T_train=13:59:52 | T_epoch=01:51:54] Epoch 4, batch 4101 / 6666 (step 30764) loss=16172.6 (avg=526) [[all losses: diffusion=0.0959493 ; kl=1.61722e+10 ; lpips=0.270729 ; repa=0.58337 ; sum_loss=16172.6]]
[T_total=14:18:33 | T_train=14:02:36 | T_epoch=01:54:38] Epoch 4, batch 4201 / 6666 (step 30864) loss=16120.2 (avg=513.5) [[all losses: diffusion=0.0959019 ; kl=1.61198e+10 ; lpips=0.270531 ; repa=0.583182 ; sum_loss=16120.2]]
[T_total=14:21:17 | T_train=14:05:20 | T_epoch=01:57:21] Epoch 4, batch 4301 / 6666 (step 30964) loss=16068.2 (avg=501.6) [[all losses: diffusion=0.0958611 ; kl=1.60678e+10 ; lpips=0.27037 ; repa=0.583015 ; sum_loss=16068.2]]
[T_total=14:24:00 | T_train=14:08:03 | T_epoch=02:00:05] Epoch 4, batch 4401 / 6666 (step 31064) loss=16016.4 (avg=490.2) [[all losses: diffusion=0.0958192 ; kl=1.60161e+10 ; lpips=0.270167 ; repa=0.582827 ; sum_loss=16016.4]]
[T_total=14:26:44 | T_train=14:10:47 | T_epoch=02:02:49] Epoch 4, batch 4501 / 6666 (step 31164) loss=15965 (avg=479.3) [[all losses: diffusion=0.0957731 ; kl=1.59647e+10 ; lpips=0.269973 ; repa=0.582642 ; sum_loss=15965]]
[T_total=14:29:28 | T_train=14:13:31 | T_epoch=02:05:33] Epoch 4, batch 4601 / 6666 (step 31264) loss=15914 (avg=468.9) [[all losses: diffusion=0.095729 ; kl=1.59136e+10 ; lpips=0.269773 ; repa=0.582458 ; sum_loss=15914]]
[T_total=14:32:12 | T_train=14:16:15 | T_epoch=02:08:17] Epoch 4, batch 4701 / 6666 (step 31364) loss=15863.2 (avg=458.9) [[all losses: diffusion=0.0956829 ; kl=1.58629e+10 ; lpips=0.269571 ; repa=0.582273 ; sum_loss=15863.2]]
[T_total=14:34:55 | T_train=14:18:58 | T_epoch=02:11:00] Epoch 4, batch 4801 / 6666 (step 31464) loss=15812.8 (avg=449.4) [[all losses: diffusion=0.0956427 ; kl=1.58125e+10 ; lpips=0.269371 ; repa=0.582089 ; sum_loss=15812.8]]
[T_total=14:37:39 | T_train=14:21:42 | T_epoch=02:13:44] Epoch 4, batch 4901 / 6666 (step 31564) loss=15762.7 (avg=440.2) [[all losses: diffusion=0.0956007 ; kl=1.57624e+10 ; lpips=0.269174 ; repa=0.581903 ; sum_loss=15762.7]]
[T_total=14:40:22 | T_train=14:24:25 | T_epoch=02:16:27] Epoch 4, batch 5001 / 6666 (step 31664) loss=15713 (avg=431.4) [[all losses: diffusion=0.0955614 ; kl=1.57126e+10 ; lpips=0.268972 ; repa=0.581719 ; sum_loss=15713]]
[T_total=14:43:06 | T_train=14:27:09 | T_epoch=02:19:11] Epoch 4, batch 5101 / 6666 (step 31764) loss=15663.5 (avg=423) [[all losses: diffusion=0.0955211 ; kl=1.56631e+10 ; lpips=0.268774 ; repa=0.581536 ; sum_loss=15663.5]]
[T_total=14:45:50 | T_train=14:29:53 | T_epoch=02:21:55] Epoch 4, batch 5201 / 6666 (step 31864) loss=15614.3 (avg=414.8) [[all losses: diffusion=0.0954809 ; kl=1.5614e+10 ; lpips=0.268576 ; repa=0.581354 ; sum_loss=15614.3]]
[T_total=14:48:34 | T_train=14:32:37 | T_epoch=02:24:39] Epoch 4, batch 5301 / 6666 (step 31964) loss=15565.5 (avg=407) [[all losses: diffusion=0.0954368 ; kl=1.55651e+10 ; lpips=0.268388 ; repa=0.581173 ; sum_loss=15565.5]]
[T_total=14:51:17 | T_train=14:35:20 | T_epoch=02:27:22] Epoch 4, batch 5401 / 6666 (step 32064) loss=15516.9 (avg=399.5) [[all losses: diffusion=0.0953982 ; kl=1.55166e+10 ; lpips=0.268197 ; repa=0.580994 ; sum_loss=15516.9]]
[T_total=14:54:01 | T_train=14:38:04 | T_epoch=02:30:06] Epoch 4, batch 5501 / 6666 (step 32164) loss=15468.7 (avg=392.2) [[all losses: diffusion=0.0953553 ; kl=1.54683e+10 ; lpips=0.268004 ; repa=0.580811 ; sum_loss=15468.7]]
[T_total=14:56:45 | T_train=14:40:48 | T_epoch=02:32:50] Epoch 4, batch 5601 / 6666 (step 32264) loss=15420.8 (avg=385.2) [[all losses: diffusion=0.095315 ; kl=1.54204e+10 ; lpips=0.267815 ; repa=0.580629 ; sum_loss=15420.8]]
[T_total=14:59:28 | T_train=14:43:31 | T_epoch=02:35:33] Epoch 4, batch 5701 / 6666 (step 32364) loss=15373.1 (avg=378.5) [[all losses: diffusion=0.0952713 ; kl=1.53727e+10 ; lpips=0.267633 ; repa=0.580451 ; sum_loss=15373.1]]
[T_total=15:02:12 | T_train=14:46:15 | T_epoch=02:38:17] Epoch 4, batch 5801 / 6666 (step 32464) loss=15325.8 (avg=372) [[all losses: diffusion=0.0952313 ; kl=1.53254e+10 ; lpips=0.267442 ; repa=0.580272 ; sum_loss=15325.8]]
[T_total=15:04:56 | T_train=14:48:59 | T_epoch=02:41:01] Epoch 4, batch 5901 / 6666 (step 32564) loss=15278.7 (avg=365.7) [[all losses: diffusion=0.0951897 ; kl=1.52783e+10 ; lpips=0.267258 ; repa=0.580093 ; sum_loss=15278.7]]
[T_total=15:07:40 | T_train=14:51:43 | T_epoch=02:43:45] Epoch 4, batch 6001 / 6666 (step 32664) loss=15231.9 (avg=359.6) [[all losses: diffusion=0.0951525 ; kl=1.52316e+10 ; lpips=0.267071 ; repa=0.579916 ; sum_loss=15231.9]]
[T_total=15:10:23 | T_train=14:54:26 | T_epoch=02:46:28] Epoch 4, batch 6101 / 6666 (step 32764) loss=15185.4 (avg=353.7) [[all losses: diffusion=0.0951131 ; kl=1.51851e+10 ; lpips=0.266886 ; repa=0.57974 ; sum_loss=15185.4]]
[T_total=15:13:07 | T_train=14:57:10 | T_epoch=02:49:12] Epoch 4, batch 6201 / 6666 (step 32864) loss=15139.2 (avg=348) [[all losses: diffusion=0.0950759 ; kl=1.51389e+10 ; lpips=0.2667 ; repa=0.579566 ; sum_loss=15139.2]]
[T_total=15:15:51 | T_train=14:59:54 | T_epoch=02:51:56] Epoch 4, batch 6301 / 6666 (step 32964) loss=15093.3 (avg=342.5) [[all losses: diffusion=0.095037 ; kl=1.50929e+10 ; lpips=0.266543 ; repa=0.579399 ; sum_loss=15093.3]]
[T_total=15:18:35 | T_train=15:02:38 | T_epoch=02:54:40] Epoch 4, batch 6401 / 6666 (step 33064) loss=15047.7 (avg=337.1) [[all losses: diffusion=0.0950007 ; kl=1.50473e+10 ; lpips=0.266352 ; repa=0.579223 ; sum_loss=15047.7]]
[T_total=15:21:18 | T_train=15:05:21 | T_epoch=02:57:23] Epoch 4, batch 6501 / 6666 (step 33164) loss=15002.3 (avg=331.9) [[all losses: diffusion=0.0949616 ; kl=1.50019e+10 ; lpips=0.266169 ; repa=0.57905 ; sum_loss=15002.3]]
[T_total=15:24:02 | T_train=15:08:05 | T_epoch=03:00:07] Epoch 4, batch 6601 / 6666 (step 33264) loss=14957.2 (avg=326.9) [[all losses: diffusion=0.0949218 ; kl=1.49568e+10 ; lpips=0.265992 ; repa=0.578877 ; sum_loss=14957.2]]
[2025-10-25 19:37:10,375][main][INFO] - [T_total=15:25:49 | T_train=15:09:52 | T_epoch=03:01:54] End of epoch 4 (33330 steps) train loss 323.739[2025-10-25 19:37:10,376][main][INFO] - [Epoch 4] All losses: [[diffusion=0.0845231 ; kl=3.23414e+08 ; lpips=0.216933 ; repa=0.530433]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:01<06:42,  1.55s/it]
Reconstructing from test set:   1%|          | 2/261 [00:02<04:43,  1.09s/it]
Reconstructing from test set:   1%|          | 3/261 [00:03<04:05,  1.05it/s]
Reconstructing from test set:   2%|▏         | 4/261 [00:03<03:46,  1.13it/s]
Reconstructing from test set:   2%|▏         | 5/261 [00:04<03:36,  1.18it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:05<03:30,  1.21it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:06<03:26,  1.23it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:07<03:22,  1.25it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:07<03:21,  1.25it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:08<03:18,  1.26it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:09<03:16,  1.27it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:10<03:15,  1.27it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:10<03:14,  1.27it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:11<03:13,  1.28it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:12<03:12,  1.28it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:13<03:11,  1.28it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:14<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:14<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:15<03:09,  1.28it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:16<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:17<03:08,  1.27it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:17<03:07,  1.27it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:18<03:06,  1.27it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:19<03:06,  1.27it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:20<03:05,  1.27it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:21<03:04,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:22<03:01,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:23<03:01,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:24<03:01,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:25<03:00,  1.27it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:25<02:59,  1.27it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:26<02:58,  1.27it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:27<02:58,  1.27it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:28<02:57,  1.27it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:28<02:56,  1.27it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:29<02:55,  1.27it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:30<02:55,  1.27it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:31<02:54,  1.27it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:32<02:53,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:32<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:33<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:34<02:50,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:35<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:36<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:37<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:38<02:46,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:39<02:46,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:39<02:45,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:40<02:44,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:41<02:43,  1.28it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:42<02:42,  1.28it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:43<02:41,  1.28it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:43<02:40,  1.28it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:44<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:45<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:47<02:37,  1.27it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:48<02:37,  1.27it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:49<02:36,  1.27it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:50<02:35,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:50<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:51<02:33,  1.28it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:52<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:53<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:54<02:31,  1.27it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:54<02:30,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:55<02:29,  1.27it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:56<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:57<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:57<02:27,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:58<02:26,  1.28it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [00:59<02:25,  1.28it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [01:00<02:24,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:01<02:23,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:01<02:23,  1.28it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:02<02:22,  1.28it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:03<02:21,  1.28it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:04<02:19,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:05<02:18,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:06<02:18,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:07<02:17,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:08<02:16,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:08<02:16,  1.28it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:09<02:15,  1.27it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:10<02:15,  1.27it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:11<02:13,  1.28it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:12<02:13,  1.28it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:12<02:12,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:13<02:11,  1.27it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:14<02:10,  1.28it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:15<02:09,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:15<02:09,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:16<02:08,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:17<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:18<02:07,  1.27it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:19<02:06,  1.28it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:19<02:05,  1.27it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:20<02:04,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:21<02:03,  1.28it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:22<02:02,  1.28it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:22<02:02,  1.28it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:23<02:00,  1.28it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:24<02:00,  1.28it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:25<01:59,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:26<01:58,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:26<01:58,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:27<01:57,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:28<01:56,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:29<01:55,  1.28it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:30<01:54,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:30<01:54,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:31<01:53,  1.28it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:32<01:52,  1.28it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:33<01:51,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:33<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:34<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:35<01:49,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:36<01:48,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:37<01:48,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:37<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:38<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:39<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:40<01:43,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:41<01:42,  1.28it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:42<01:42,  1.28it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:43<01:41,  1.28it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:44<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:44<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:45<01:39,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:46<01:38,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:47<01:37,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:47<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:48<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:49<01:35,  1.28it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:50<01:34,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:51<01:33,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:51<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:52<01:31,  1.28it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:53<01:30,  1.29it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:54<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:54<01:29,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:55<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:56<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:57<01:27,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:58<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [01:58<01:25,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [01:59<01:25,  1.28it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:00<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:01<01:23,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:03<01:21,  1.28it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:04<01:20,  1.28it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:05<01:19,  1.28it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:05<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:06<01:18,  1.27it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:07<01:17,  1.27it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:08<01:17,  1.27it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:09<01:16,  1.27it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:09<01:15,  1.28it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:10<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:11<01:13,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:12<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:12<01:11,  1.28it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:13<01:10,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:14<01:10,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:15<01:09,  1.28it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:16<01:08,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:16<01:07,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:17<01:06,  1.29it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:18<01:06,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:19<01:05,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:19<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:20<01:03,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:21<01:03,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:22<01:02,  1.28it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:23<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:23<01:00,  1.28it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:24<01:00,  1.28it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:25<00:59,  1.28it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:26<00:58,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:27<00:57,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:27<00:56,  1.29it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:28<00:55,  1.29it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:29<00:55,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:30<00:54,  1.28it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:30<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:31<00:52,  1.29it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:32<00:52,  1.28it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:33<00:51,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:34<00:50,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:34<00:49,  1.29it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:35<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:36<00:48,  1.29it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:37<00:47,  1.29it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:37<00:46,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:38<00:45,  1.29it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:39<00:45,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:40<00:44,  1.29it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:41<00:43,  1.29it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:41<00:42,  1.29it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:42<00:41,  1.29it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:43<00:41,  1.29it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:44<00:40,  1.29it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:44<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:45<00:38,  1.28it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:46<00:38,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:47<00:37,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:48<00:35,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:49<00:35,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:50<00:34,  1.28it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:51<00:33,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:51<00:32,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:53<00:31,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:54<00:30,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:55<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:55<00:28,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:57<00:27,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:58<00:26,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [02:58<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [02:59<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:00<00:24,  1.28it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:01<00:23,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:02<00:22,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:02<00:21,  1.28it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:03<00:21,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:04<00:20,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:05<00:19,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:06<00:18,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:06<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:08<00:16,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:09<00:15,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:09<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:10<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:11<00:13,  1.28it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:12<00:12,  1.29it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:13<00:11,  1.29it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:13<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:14<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:15<00:09,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:16<00:08,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:16<00:07,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:17<00:06,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:18<00:06,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:19<00:05,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:20<00:04,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:20<00:03,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:21<00:03,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:22<00:02,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:23<00:01,  1.29it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:23<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.28it/s]
[2025-10-25 19:40:38,194][main][INFO] - [Epoch 5] Test metrics: [[MSE=30.03 | MAE=0.1249 | LPIPS=0.2154 | PSNR=15.22 | SSIM=0.351 | dreamsim=0.3335 | FID=40.99]][2025-10-25 19:40:38,195][main][INFO] - [Epoch 5] Best metrics: [[min_MSE=30.03 | min_MAE=0.1249 | min_LPIPS=0.2154 | max_PSNR=15.22 | max_SSIM=0.351 | min_dreamsim=0.3335 | min_FID=40.99]][2025-10-25 19:40:38,196][main][DEBUG] - Writing images to disk...[2025-10-25 19:40:39,044][main][DEBUG] - Image(s) saved on disk[2025-10-25 19:40:39,245][main][INFO] - End of epoch timers: [T_train=15:09:52 | T_epoch=03:01:54 | T_eval=00:17:27 | T_total=15:29:18][2025-10-25 19:40:39,246][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-25 19:40:51,215][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-25 19:41:02,678][main][INFO] - ---

[2025-10-25 19:41:02,679][main][INFO] - [T_total=15:29:41 | T_train=15:09:52] Start epoch 5[T_total=15:29:44 | T_train=15:09:54 | T_epoch=00:00:02] Epoch 5, batch 1 / 6666 (step 33330) loss=14927.6 (avg=0.3104) [[all losses: diffusion=0.0948963 ; kl=1.49272e+10 ; lpips=0.265873 ; repa=0.57876 ; sum_loss=14927.6]]
[T_total=15:32:28 | T_train=15:12:38 | T_epoch=00:02:46] Epoch 5, batch 101 / 6666 (step 33430) loss=14882.9 (avg=0.3175) [[all losses: diffusion=0.0948589 ; kl=1.48826e+10 ; lpips=0.265693 ; repa=0.578587 ; sum_loss=14882.9]]
[T_total=15:35:11 | T_train=15:15:22 | T_epoch=00:05:30] Epoch 5, batch 201 / 6666 (step 33530) loss=14838.5 (avg=0.3231) [[all losses: diffusion=0.0948225 ; kl=1.48382e+10 ; lpips=0.265537 ; repa=0.578426 ; sum_loss=14838.5]]
[T_total=15:37:55 | T_train=15:18:06 | T_epoch=00:08:14] Epoch 5, batch 301 / 6666 (step 33630) loss=14794.4 (avg=0.3211) [[all losses: diffusion=0.0947847 ; kl=1.47941e+10 ; lpips=0.265358 ; repa=0.578254 ; sum_loss=14794.4]]
[T_total=15:40:39 | T_train=15:20:50 | T_epoch=00:10:58] Epoch 5, batch 401 / 6666 (step 33730) loss=14750.6 (avg=0.32) [[all losses: diffusion=0.0947475 ; kl=1.47502e+10 ; lpips=0.265177 ; repa=0.578083 ; sum_loss=14750.6]]
[T_total=15:43:23 | T_train=15:23:34 | T_epoch=00:13:42] Epoch 5, batch 501 / 6666 (step 33830) loss=14707 (avg=0.3191) [[all losses: diffusion=0.0947108 ; kl=1.47066e+10 ; lpips=0.264993 ; repa=0.57791 ; sum_loss=14707]]
[T_total=15:46:07 | T_train=15:26:18 | T_epoch=00:16:25] Epoch 5, batch 601 / 6666 (step 33930) loss=14663.6 (avg=0.3186) [[all losses: diffusion=0.0946742 ; kl=1.46633e+10 ; lpips=0.264812 ; repa=0.577741 ; sum_loss=14663.6]]
[T_total=15:48:51 | T_train=15:29:01 | T_epoch=00:19:09] Epoch 5, batch 701 / 6666 (step 34030) loss=14620.5 (avg=0.3187) [[all losses: diffusion=0.0946407 ; kl=1.46202e+10 ; lpips=0.264642 ; repa=0.577574 ; sum_loss=14620.5]]
[T_total=15:51:34 | T_train=15:31:45 | T_epoch=00:21:53] Epoch 5, batch 801 / 6666 (step 34130) loss=14577.7 (avg=0.3185) [[all losses: diffusion=0.0946064 ; kl=1.45773e+10 ; lpips=0.264467 ; repa=0.577402 ; sum_loss=14577.7]]
[T_total=15:54:18 | T_train=15:34:28 | T_epoch=00:24:36] Epoch 5, batch 901 / 6666 (step 34230) loss=14535.1 (avg=0.3182) [[all losses: diffusion=0.094567 ; kl=1.45347e+10 ; lpips=0.264292 ; repa=0.577234 ; sum_loss=14535.1]]
[T_total=15:57:01 | T_train=15:37:12 | T_epoch=00:27:20] Epoch 5, batch 1001 / 6666 (step 34330) loss=14492.8 (avg=0.3181) [[all losses: diffusion=0.0945328 ; kl=1.44924e+10 ; lpips=0.264116 ; repa=0.577066 ; sum_loss=14492.8]]
[T_total=15:59:45 | T_train=15:39:56 | T_epoch=00:30:04] Epoch 5, batch 1101 / 6666 (step 34430) loss=14450.7 (avg=0.3179) [[all losses: diffusion=0.094494 ; kl=1.44503e+10 ; lpips=0.263942 ; repa=0.576899 ; sum_loss=14450.7]]
[T_total=16:02:29 | T_train=15:42:40 | T_epoch=00:32:47] Epoch 5, batch 1201 / 6666 (step 34530) loss=14408.8 (avg=0.3177) [[all losses: diffusion=0.0944566 ; kl=1.44085e+10 ; lpips=0.26377 ; repa=0.576731 ; sum_loss=14408.8]]
[T_total=16:05:13 | T_train=15:45:24 | T_epoch=00:35:31] Epoch 5, batch 1301 / 6666 (step 34630) loss=14367.2 (avg=0.3175) [[all losses: diffusion=0.0944208 ; kl=1.43669e+10 ; lpips=0.263594 ; repa=0.576564 ; sum_loss=14367.2]]
[T_total=16:07:57 | T_train=15:48:07 | T_epoch=00:38:15] Epoch 5, batch 1401 / 6666 (step 34730) loss=14325.9 (avg=0.3173) [[all losses: diffusion=0.0943821 ; kl=1.43255e+10 ; lpips=0.263422 ; repa=0.576395 ; sum_loss=14325.9]]
[T_total=16:10:40 | T_train=15:50:51 | T_epoch=00:40:59] Epoch 5, batch 1501 / 6666 (step 34830) loss=14284.7 (avg=0.3171) [[all losses: diffusion=0.0943457 ; kl=1.42844e+10 ; lpips=0.26325 ; repa=0.576227 ; sum_loss=14284.7]]
[T_total=16:13:24 | T_train=15:53:35 | T_epoch=00:43:43] Epoch 5, batch 1601 / 6666 (step 34930) loss=14243.8 (avg=0.317) [[all losses: diffusion=0.0943132 ; kl=1.42435e+10 ; lpips=0.263072 ; repa=0.576063 ; sum_loss=14243.8]]
[T_total=16:16:08 | T_train=15:56:19 | T_epoch=00:46:27] Epoch 5, batch 1701 / 6666 (step 35030) loss=14203.2 (avg=0.3169) [[all losses: diffusion=0.0942746 ; kl=1.42028e+10 ; lpips=0.262907 ; repa=0.575899 ; sum_loss=14203.2]]
[T_total=16:18:52 | T_train=15:59:03 | T_epoch=00:49:11] Epoch 5, batch 1801 / 6666 (step 35130) loss=14162.8 (avg=0.3168) [[all losses: diffusion=0.0942388 ; kl=1.41624e+10 ; lpips=0.262735 ; repa=0.575733 ; sum_loss=14162.8]]
[T_total=16:21:36 | T_train=16:01:47 | T_epoch=00:51:55] Epoch 5, batch 1901 / 6666 (step 35230) loss=14122.6 (avg=0.3166) [[all losses: diffusion=0.0942007 ; kl=1.41222e+10 ; lpips=0.262568 ; repa=0.57557 ; sum_loss=14122.6]]
[T_total=16:24:20 | T_train=16:04:31 | T_epoch=00:54:38] Epoch 5, batch 2001 / 6666 (step 35330) loss=14082.6 (avg=0.3165) [[all losses: diffusion=0.094165 ; kl=1.40822e+10 ; lpips=0.262397 ; repa=0.575407 ; sum_loss=14082.6]]
[T_total=16:27:04 | T_train=16:07:14 | T_epoch=00:57:22] Epoch 5, batch 2101 / 6666 (step 35430) loss=14042.8 (avg=0.3164) [[all losses: diffusion=0.0941274 ; kl=1.40425e+10 ; lpips=0.26223 ; repa=0.575242 ; sum_loss=14042.8]]
[T_total=16:29:47 | T_train=16:09:58 | T_epoch=01:00:06] Epoch 5, batch 2201 / 6666 (step 35530) loss=14003.3 (avg=0.3163) [[all losses: diffusion=0.0940921 ; kl=1.40029e+10 ; lpips=0.262061 ; repa=0.575083 ; sum_loss=14003.3]]
[T_total=16:32:31 | T_train=16:12:42 | T_epoch=01:02:50] Epoch 5, batch 2301 / 6666 (step 35630) loss=13964 (avg=0.3162) [[all losses: diffusion=0.0940569 ; kl=1.39636e+10 ; lpips=0.261896 ; repa=0.574921 ; sum_loss=13964]]
[T_total=16:35:15 | T_train=16:15:26 | T_epoch=01:05:34] Epoch 5, batch 2401 / 6666 (step 35730) loss=13924.9 (avg=0.3161) [[all losses: diffusion=0.0940194 ; kl=1.39246e+10 ; lpips=0.261724 ; repa=0.574758 ; sum_loss=13924.9]]
[T_total=16:37:58 | T_train=16:18:09 | T_epoch=01:08:17] Epoch 5, batch 2501 / 6666 (step 35830) loss=16147.5 (avg=3.24e+04) [[all losses: diffusion=0.0940983 ; kl=1.61471e+10 ; lpips=0.261981 ; repa=0.574839 ; sum_loss=16147.5]]
[T_total=16:40:42 | T_train=16:20:53 | T_epoch=01:11:01] Epoch 5, batch 2601 / 6666 (step 35930) loss=16102.5 (avg=3.115e+04) [[all losses: diffusion=0.0940721 ; kl=1.61021e+10 ; lpips=0.261888 ; repa=0.574712 ; sum_loss=16102.5]]
[T_total=16:43:26 | T_train=16:23:37 | T_epoch=01:13:44] Epoch 5, batch 2701 / 6666 (step 36030) loss=16057.8 (avg=3e+04) [[all losses: diffusion=0.0940365 ; kl=1.60575e+10 ; lpips=0.261735 ; repa=0.574556 ; sum_loss=16057.8]]
[T_total=16:46:10 | T_train=16:26:20 | T_epoch=01:16:28] Epoch 5, batch 2801 / 6666 (step 36130) loss=16013.4 (avg=2.893e+04) [[all losses: diffusion=0.0940018 ; kl=1.6013e+10 ; lpips=0.261566 ; repa=0.574396 ; sum_loss=16013.4]]
[T_total=16:48:53 | T_train=16:29:04 | T_epoch=01:19:12] Epoch 5, batch 2901 / 6666 (step 36230) loss=15969.2 (avg=2.793e+04) [[all losses: diffusion=0.093965 ; kl=1.59688e+10 ; lpips=0.261403 ; repa=0.574236 ; sum_loss=15969.2]]
[T_total=16:51:38 | T_train=16:31:48 | T_epoch=01:21:56] Epoch 5, batch 3001 / 6666 (step 36330) loss=15925.2 (avg=2.7e+04) [[all losses: diffusion=0.0939453 ; kl=1.59249e+10 ; lpips=0.261283 ; repa=0.574105 ; sum_loss=15925.2]]
[T_total=16:54:21 | T_train=16:34:32 | T_epoch=01:24:40] Epoch 5, batch 3101 / 6666 (step 36430) loss=15881.5 (avg=2.613e+04) [[all losses: diffusion=0.0939135 ; kl=1.58812e+10 ; lpips=0.261117 ; repa=0.573948 ; sum_loss=15881.5]]
[T_total=16:57:05 | T_train=16:37:16 | T_epoch=01:27:24] Epoch 5, batch 3201 / 6666 (step 36530) loss=15838 (avg=2.531e+04) [[all losses: diffusion=0.0938779 ; kl=1.58377e+10 ; lpips=0.260952 ; repa=0.573788 ; sum_loss=15838]]
[T_total=16:59:49 | T_train=16:40:00 | T_epoch=01:30:07] Epoch 5, batch 3301 / 6666 (step 36630) loss=15794.8 (avg=2.455e+04) [[all losses: diffusion=0.0938435 ; kl=1.57944e+10 ; lpips=0.260793 ; repa=0.573632 ; sum_loss=15794.8]]
[T_total=17:02:33 | T_train=16:42:43 | T_epoch=01:32:51] Epoch 5, batch 3401 / 6666 (step 36730) loss=15751.8 (avg=2.382e+04) [[all losses: diffusion=0.0938109 ; kl=1.57514e+10 ; lpips=0.260632 ; repa=0.573476 ; sum_loss=15751.8]]
[T_total=17:05:16 | T_train=16:45:27 | T_epoch=01:35:35] Epoch 5, batch 3501 / 6666 (step 36830) loss=15709 (avg=2.314e+04) [[all losses: diffusion=0.0937789 ; kl=1.57087e+10 ; lpips=0.260474 ; repa=0.57332 ; sum_loss=15709]]
[T_total=17:08:00 | T_train=16:48:11 | T_epoch=01:38:18] Epoch 5, batch 3601 / 6666 (step 36930) loss=15666.5 (avg=2.25e+04) [[all losses: diffusion=0.0937445 ; kl=1.56661e+10 ; lpips=0.260313 ; repa=0.573163 ; sum_loss=15666.5]]
[T_total=17:10:44 | T_train=16:50:55 | T_epoch=01:41:02] Epoch 5, batch 3701 / 6666 (step 37030) loss=15624.2 (avg=2.189e+04) [[all losses: diffusion=0.0937118 ; kl=1.56238e+10 ; lpips=0.260154 ; repa=0.573005 ; sum_loss=15624.2]]
[T_total=17:13:28 | T_train=16:53:38 | T_epoch=01:43:46] Epoch 5, batch 3801 / 6666 (step 37130) loss=15582.1 (avg=2.132e+04) [[all losses: diffusion=0.0937032 ; kl=1.55818e+10 ; lpips=0.260064 ; repa=0.572894 ; sum_loss=15582.1]]
[T_total=17:16:11 | T_train=16:56:22 | T_epoch=01:46:30] Epoch 5, batch 3901 / 6666 (step 37230) loss=15540.3 (avg=2.077e+04) [[all losses: diffusion=0.0936669 ; kl=1.55399e+10 ; lpips=0.25991 ; repa=0.572737 ; sum_loss=15540.3]]
[T_total=17:18:55 | T_train=16:59:06 | T_epoch=01:49:14] Epoch 5, batch 4001 / 6666 (step 37330) loss=15498.6 (avg=2.025e+04) [[all losses: diffusion=0.0936318 ; kl=1.54983e+10 ; lpips=0.259755 ; repa=0.572584 ; sum_loss=15498.6]]
[T_total=17:21:39 | T_train=17:01:50 | T_epoch=01:51:58] Epoch 5, batch 4101 / 6666 (step 37430) loss=15457.2 (avg=1.976e+04) [[all losses: diffusion=0.0935963 ; kl=1.54569e+10 ; lpips=0.259594 ; repa=0.572429 ; sum_loss=15457.2]]
[T_total=17:24:23 | T_train=17:04:34 | T_epoch=01:54:41] Epoch 5, batch 4201 / 6666 (step 37530) loss=15416.1 (avg=1.929e+04) [[all losses: diffusion=0.0935655 ; kl=1.54157e+10 ; lpips=0.259432 ; repa=0.572276 ; sum_loss=15416.1]]
[T_total=17:27:07 | T_train=17:07:18 | T_epoch=01:57:25] Epoch 5, batch 4301 / 6666 (step 37630) loss=15375.1 (avg=1.884e+04) [[all losses: diffusion=0.0935307 ; kl=1.53747e+10 ; lpips=0.259291 ; repa=0.57213 ; sum_loss=15375.1]]
[T_total=17:29:51 | T_train=17:10:01 | T_epoch=02:00:09] Epoch 5, batch 4401 / 6666 (step 37730) loss=15334.3 (avg=1.841e+04) [[all losses: diffusion=0.0934957 ; kl=1.5334e+10 ; lpips=0.259139 ; repa=0.571977 ; sum_loss=15334.3]]
[T_total=17:32:35 | T_train=17:12:45 | T_epoch=02:02:53] Epoch 5, batch 4501 / 6666 (step 37830) loss=15293.8 (avg=1.8e+04) [[all losses: diffusion=0.093463 ; kl=1.52934e+10 ; lpips=0.258988 ; repa=0.571826 ; sum_loss=15293.8]]
[T_total=17:35:18 | T_train=17:15:29 | T_epoch=02:05:37] Epoch 5, batch 4601 / 6666 (step 37930) loss=15253.5 (avg=1.761e+04) [[all losses: diffusion=0.0934322 ; kl=1.52531e+10 ; lpips=0.258831 ; repa=0.571674 ; sum_loss=15253.5]]
[T_total=17:38:02 | T_train=17:18:13 | T_epoch=02:08:20] Epoch 5, batch 4701 / 6666 (step 38030) loss=15213.4 (avg=1.724e+04) [[all losses: diffusion=0.0933982 ; kl=1.5213e+10 ; lpips=0.25868 ; repa=0.571522 ; sum_loss=15213.4]]
[T_total=17:40:46 | T_train=17:20:57 | T_epoch=02:11:04] Epoch 5, batch 4801 / 6666 (step 38130) loss=15173.5 (avg=1.688e+04) [[all losses: diffusion=0.093365 ; kl=1.51731e+10 ; lpips=0.258528 ; repa=0.571374 ; sum_loss=15173.5]]
[T_total=17:43:30 | T_train=17:23:40 | T_epoch=02:13:48] Epoch 5, batch 4901 / 6666 (step 38230) loss=15133.8 (avg=1.653e+04) [[all losses: diffusion=0.0933296 ; kl=1.51334e+10 ; lpips=0.258379 ; repa=0.571226 ; sum_loss=15133.8]]
[T_total=17:46:13 | T_train=17:26:24 | T_epoch=02:16:32] Epoch 5, batch 5001 / 6666 (step 38330) loss=15094.3 (avg=1.62e+04) [[all losses: diffusion=0.0932946 ; kl=1.5094e+10 ; lpips=0.258228 ; repa=0.571078 ; sum_loss=15094.3]]
[T_total=17:48:57 | T_train=17:29:08 | T_epoch=02:19:16] Epoch 5, batch 5101 / 6666 (step 38430) loss=15055 (avg=1.588e+04) [[all losses: diffusion=0.0932758 ; kl=1.50547e+10 ; lpips=0.258103 ; repa=0.570946 ; sum_loss=15055]]
[T_total=17:51:41 | T_train=17:31:51 | T_epoch=02:21:59] Epoch 5, batch 5201 / 6666 (step 38530) loss=15016.1 (avg=1.558e+04) [[all losses: diffusion=0.093334 ; kl=1.50157e+10 ; lpips=0.258295 ; repa=0.570992 ; sum_loss=15016.1]]
[T_total=17:54:25 | T_train=17:34:35 | T_epoch=02:24:43] Epoch 5, batch 5301 / 6666 (step 38630) loss=14977.2 (avg=1.529e+04) [[all losses: diffusion=0.0933029 ; kl=1.49768e+10 ; lpips=0.258152 ; repa=0.570846 ; sum_loss=14977.2]]
[T_total=17:57:08 | T_train=17:37:19 | T_epoch=02:27:27] Epoch 5, batch 5401 / 6666 (step 38730) loss=14938.5 (avg=1.5e+04) [[all losses: diffusion=0.0932684 ; kl=1.49382e+10 ; lpips=0.258002 ; repa=0.570699 ; sum_loss=14938.5]]
[T_total=17:59:52 | T_train=17:40:03 | T_epoch=02:30:10] Epoch 5, batch 5501 / 6666 (step 38830) loss=14900.1 (avg=1.473e+04) [[all losses: diffusion=0.0932361 ; kl=1.48997e+10 ; lpips=0.257854 ; repa=0.570552 ; sum_loss=14900.1]]
[T_total=18:02:36 | T_train=17:42:46 | T_epoch=02:32:54] Epoch 5, batch 5601 / 6666 (step 38930) loss=14861.8 (avg=1.447e+04) [[all losses: diffusion=0.0932062 ; kl=1.48614e+10 ; lpips=0.257702 ; repa=0.570404 ; sum_loss=14861.8]]
[T_total=18:05:19 | T_train=17:45:30 | T_epoch=02:35:38] Epoch 5, batch 5701 / 6666 (step 39030) loss=14823.7 (avg=1.421e+04) [[all losses: diffusion=0.0931738 ; kl=1.48234e+10 ; lpips=0.257549 ; repa=0.570255 ; sum_loss=14823.7]]
[T_total=18:08:03 | T_train=17:48:14 | T_epoch=02:38:21] Epoch 5, batch 5801 / 6666 (step 39130) loss=14785.8 (avg=1.397e+04) [[all losses: diffusion=0.0931444 ; kl=1.47855e+10 ; lpips=0.257401 ; repa=0.570107 ; sum_loss=14785.8]]
[T_total=18:10:47 | T_train=17:50:58 | T_epoch=02:41:05] Epoch 5, batch 5901 / 6666 (step 39230) loss=14748.2 (avg=1.373e+04) [[all losses: diffusion=0.0931147 ; kl=1.47478e+10 ; lpips=0.257254 ; repa=0.569962 ; sum_loss=14748.2]]
[T_total=18:13:31 | T_train=17:53:41 | T_epoch=02:43:49] Epoch 5, batch 6001 / 6666 (step 39330) loss=14710.7 (avg=1.35e+04) [[all losses: diffusion=0.0930845 ; kl=1.47103e+10 ; lpips=0.25711 ; repa=0.569819 ; sum_loss=14710.7]]
[T_total=18:16:15 | T_train=17:56:25 | T_epoch=02:46:33] Epoch 5, batch 6101 / 6666 (step 39430) loss=14673.3 (avg=1.328e+04) [[all losses: diffusion=0.093071 ; kl=1.4673e+10 ; lpips=0.257031 ; repa=0.569708 ; sum_loss=14673.3]]
[T_total=18:18:58 | T_train=17:59:09 | T_epoch=02:49:17] Epoch 5, batch 6201 / 6666 (step 39530) loss=14636.2 (avg=1.307e+04) [[all losses: diffusion=0.0930519 ; kl=1.46359e+10 ; lpips=0.256937 ; repa=0.56959 ; sum_loss=14636.2]]
[T_total=18:21:42 | T_train=18:01:53 | T_epoch=02:52:00] Epoch 5, batch 6301 / 6666 (step 39630) loss=14599.3 (avg=1.286e+04) [[all losses: diffusion=0.09302 ; kl=1.45989e+10 ; lpips=0.256792 ; repa=0.569447 ; sum_loss=14599.3]]
[T_total=18:24:26 | T_train=18:04:37 | T_epoch=02:54:44] Epoch 5, batch 6401 / 6666 (step 39730) loss=14562.6 (avg=1.266e+04) [[all losses: diffusion=0.0929858 ; kl=1.45622e+10 ; lpips=0.256651 ; repa=0.569304 ; sum_loss=14562.6]]
[T_total=18:27:10 | T_train=18:07:21 | T_epoch=02:57:28] Epoch 5, batch 6501 / 6666 (step 39830) loss=14526 (avg=1.246e+04) [[all losses: diffusion=0.0929575 ; kl=1.45256e+10 ; lpips=0.256505 ; repa=0.569162 ; sum_loss=14526]]
[T_total=18:29:54 | T_train=18:10:04 | T_epoch=03:00:12] Epoch 5, batch 6601 / 6666 (step 39930) loss=14489.6 (avg=1.228e+04) [[all losses: diffusion=0.0929309 ; kl=1.44893e+10 ; lpips=0.25636 ; repa=0.569021 ; sum_loss=14489.6]]
[2025-10-25 22:43:01,888][main][INFO] - [T_total=18:31:40 | T_train=18:11:51 | T_epoch=03:01:59] End of epoch 5 (39996 steps) train loss 12156.3[2025-10-25 22:43:01,890][main][INFO] - [Epoch 5] All losses: [[diffusion=0.0830056 ; kl=1.2156e+10 ; lpips=0.208209 ; repa=0.519751]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:01<06:33,  1.51s/it]
Reconstructing from test set:   1%|          | 2/261 [00:02<04:41,  1.09s/it]
Reconstructing from test set:   1%|          | 3/261 [00:03<04:03,  1.06it/s]
Reconstructing from test set:   2%|▏         | 4/261 [00:03<03:46,  1.14it/s]
Reconstructing from test set:   2%|▏         | 5/261 [00:04<03:36,  1.18it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:05<03:29,  1.22it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:06<03:25,  1.23it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:06<03:22,  1.25it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:07<03:20,  1.26it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:08<03:18,  1.26it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:09<03:17,  1.27it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:10<03:16,  1.27it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:10<03:14,  1.27it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:11<03:13,  1.28it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:12<03:12,  1.28it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:13<03:11,  1.28it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:14<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:14<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:15<03:09,  1.28it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:16<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:17<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:17<03:07,  1.28it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:18<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:19<03:05,  1.28it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:20<03:04,  1.28it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:22<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:23<03:01,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:24<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:24<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:25<02:59,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:26<02:58,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:27<02:57,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:28<02:57,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:28<02:56,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:29<02:55,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:30<02:54,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:31<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:32<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:32<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:33<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:34<02:50,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:35<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:36<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:37<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:38<02:46,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:39<02:45,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:39<02:45,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:40<02:44,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:41<02:43,  1.28it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:42<02:42,  1.28it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:42<02:41,  1.28it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:43<02:41,  1.28it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:44<02:40,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:45<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:47<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:48<02:36,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:49<02:36,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:50<02:35,  1.27it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:50<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:51<02:33,  1.28it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:52<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:53<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:53<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:54<02:30,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:55<02:29,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:56<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:57<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:57<02:26,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:58<02:25,  1.28it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [00:59<02:24,  1.28it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [01:00<02:24,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:00<02:23,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:01<02:23,  1.28it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:02<02:22,  1.28it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:03<02:21,  1.28it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:04<02:19,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:05<02:18,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:06<02:17,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:07<02:17,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:07<02:16,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:08<02:15,  1.28it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:09<02:15,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:10<02:14,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:11<02:13,  1.28it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:11<02:12,  1.28it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:12<02:12,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:13<02:11,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:14<02:10,  1.28it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:15<02:10,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:15<02:09,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:16<02:08,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:17<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:18<02:06,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:18<02:06,  1.28it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:19<02:05,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:20<02:04,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:21<02:03,  1.28it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:22<02:02,  1.28it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:22<02:01,  1.28it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:23<02:01,  1.28it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:24<02:00,  1.28it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:25<01:59,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:25<01:59,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:26<01:58,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:27<01:57,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:28<01:56,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:29<01:55,  1.28it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:29<01:54,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:30<01:53,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:31<01:53,  1.28it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:32<01:52,  1.28it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:32<01:51,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:33<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:34<01:49,  1.28it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:35<01:49,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:36<01:48,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:36<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:37<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:38<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:39<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:41<01:43,  1.28it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:42<01:42,  1.28it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:43<01:41,  1.28it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:43<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:44<01:39,  1.28it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:45<01:39,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:46<01:38,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:47<01:37,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:47<01:37,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:48<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:49<01:35,  1.28it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:50<01:34,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:50<01:33,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:51<01:33,  1.28it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:52<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:53<01:31,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:54<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:54<01:29,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:55<01:29,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:56<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:57<01:27,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:58<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [01:58<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [01:59<01:25,  1.28it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:00<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:01<01:23,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:01<01:22,  1.28it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:02<01:21,  1.28it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:03<01:21,  1.28it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:04<01:20,  1.28it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:05<01:19,  1.28it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:05<01:19,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:06<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:07<01:17,  1.27it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:08<01:16,  1.27it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:08<01:16,  1.27it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:09<01:15,  1.27it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:10<01:14,  1.27it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:11<01:13,  1.27it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:12<01:13,  1.27it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:12<01:12,  1.27it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:13<01:11,  1.27it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:14<01:10,  1.27it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:15<01:09,  1.27it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:16<01:09,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:16<01:08,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:17<01:07,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:18<01:06,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:19<01:05,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:19<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:20<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:21<01:03,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:22<01:02,  1.28it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:23<01:01,  1.27it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:23<01:01,  1.27it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:24<01:00,  1.27it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:25<00:59,  1.28it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:26<00:58,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:26<00:57,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:27<00:57,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:28<00:56,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:29<00:55,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:30<00:54,  1.28it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:30<00:54,  1.27it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:31<00:53,  1.27it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:32<00:52,  1.28it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:33<00:51,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:34<00:50,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:34<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:35<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:36<00:48,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:37<00:47,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:37<00:46,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:38<00:46,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:39<00:45,  1.27it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:40<00:44,  1.27it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:41<00:43,  1.27it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:41<00:43,  1.27it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:42<00:42,  1.28it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:43<00:41,  1.28it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:44<00:40,  1.28it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:44<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:45<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:46<00:38,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:47<00:37,  1.27it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:48<00:36,  1.27it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:49<00:35,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:50<00:34,  1.28it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:51<00:33,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:53<00:31,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:54<00:30,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:55<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:55<00:28,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:57<00:27,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:58<00:26,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [02:59<00:25,  1.27it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [02:59<00:25,  1.27it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:00<00:24,  1.27it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:01<00:23,  1.27it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:02<00:22,  1.26it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:03<00:22,  1.26it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:03<00:21,  1.26it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:04<00:20,  1.27it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:05<00:19,  1.27it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:06<00:18,  1.27it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:07<00:18,  1.27it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:07<00:17,  1.27it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:08<00:16,  1.27it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:09<00:15,  1.27it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:10<00:14,  1.27it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:10<00:14,  1.27it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:11<00:13,  1.27it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:12<00:12,  1.27it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:13<00:11,  1.27it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:14<00:10,  1.27it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:14<00:10,  1.27it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:15<00:09,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:16<00:08,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:17<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:17<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:18<00:06,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:19<00:05,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:20<00:04,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:21<00:03,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:21<00:03,  1.28it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:22<00:02,  1.28it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:23<00:01,  1.28it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.27it/s]
[2025-10-25 22:46:30,029][main][INFO] - [Epoch 6] Test metrics: [[MSE=27.85 | MAE=0.1193 | LPIPS=0.2014 | PSNR=15.55 | SSIM=0.3646 | dreamsim=0.3094 | FID=35.9]][2025-10-25 22:46:30,031][main][INFO] - [Epoch 6] Best metrics: [[min_MSE=27.85 | min_MAE=0.1193 | min_LPIPS=0.2014 | max_PSNR=15.55 | max_SSIM=0.3646 | min_dreamsim=0.3094 | min_FID=35.9]][2025-10-25 22:46:30,032][main][DEBUG] - Writing images to disk...[2025-10-25 22:46:30,873][main][DEBUG] - Image(s) saved on disk[2025-10-25 22:46:31,075][main][INFO] - End of epoch timers: [T_train=18:11:51 | T_epoch=03:01:59 | T_eval=00:20:57 | T_total=18:35:09][2025-10-25 22:46:31,076][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-25 22:46:42,024][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-25 22:46:54,114][main][INFO] - ---

[2025-10-25 22:46:54,115][main][INFO] - [T_total=18:35:32 | T_train=18:11:51] Start epoch 6[T_total=18:35:35 | T_train=18:11:54 | T_epoch=00:00:02] Epoch 6, batch 1 / 6666 (step 39996) loss=14465.7 (avg=0.3116) [[all losses: diffusion=0.0929146 ; kl=1.44653e+10 ; lpips=0.256263 ; repa=0.568926 ; sum_loss=14465.7]]
[T_total=18:38:19 | T_train=18:14:37 | T_epoch=00:02:46] Epoch 6, batch 101 / 6666 (step 40096) loss=14429.6 (avg=0.3132) [[all losses: diffusion=0.0928837 ; kl=1.44293e+10 ; lpips=0.25613 ; repa=0.568787 ; sum_loss=14429.6]]
[T_total=18:41:03 | T_train=18:17:21 | T_epoch=00:05:30] Epoch 6, batch 201 / 6666 (step 40196) loss=14393.7 (avg=0.3121) [[all losses: diffusion=0.0928548 ; kl=1.43934e+10 ; lpips=0.255989 ; repa=0.568646 ; sum_loss=14393.7]]
[T_total=18:43:46 | T_train=18:20:05 | T_epoch=00:08:13] Epoch 6, batch 301 / 6666 (step 40296) loss=14358 (avg=0.3115) [[all losses: diffusion=0.0928228 ; kl=1.43577e+10 ; lpips=0.255852 ; repa=0.568506 ; sum_loss=14358]]
[T_total=18:46:30 | T_train=18:22:49 | T_epoch=00:10:57] Epoch 6, batch 401 / 6666 (step 40396) loss=14322.5 (avg=0.3112) [[all losses: diffusion=0.0927969 ; kl=1.43221e+10 ; lpips=0.255704 ; repa=0.568366 ; sum_loss=14322.5]]
[T_total=18:49:14 | T_train=18:25:32 | T_epoch=00:13:41] Epoch 6, batch 501 / 6666 (step 40496) loss=14287.1 (avg=0.3107) [[all losses: diffusion=0.0927651 ; kl=1.42868e+10 ; lpips=0.255561 ; repa=0.568225 ; sum_loss=14287.1]]
[T_total=18:51:57 | T_train=18:28:16 | T_epoch=00:16:24] Epoch 6, batch 601 / 6666 (step 40596) loss=14251.9 (avg=0.3105) [[all losses: diffusion=0.0927352 ; kl=1.42516e+10 ; lpips=0.255417 ; repa=0.568085 ; sum_loss=14251.9]]
[T_total=18:54:41 | T_train=18:31:00 | T_epoch=00:19:08] Epoch 6, batch 701 / 6666 (step 40696) loss=14216.9 (avg=0.3103) [[all losses: diffusion=0.0927065 ; kl=1.42165e+10 ; lpips=0.255273 ; repa=0.567943 ; sum_loss=14216.9]]
[T_total=18:57:24 | T_train=18:33:43 | T_epoch=00:21:51] Epoch 6, batch 801 / 6666 (step 40796) loss=14183.4 (avg=70.81) [[all losses: diffusion=0.0927389 ; kl=1.41831e+10 ; lpips=0.25533 ; repa=0.567917 ; sum_loss=14183.4]]
[T_total=19:00:08 | T_train=18:36:27 | T_epoch=00:24:35] Epoch 6, batch 901 / 6666 (step 40896) loss=14148.8 (avg=63.03) [[all losses: diffusion=0.0927451 ; kl=1.41484e+10 ; lpips=0.255356 ; repa=0.567869 ; sum_loss=14148.8]]
[T_total=19:02:52 | T_train=18:39:10 | T_epoch=00:27:19] Epoch 6, batch 1001 / 6666 (step 40996) loss=14114.2 (avg=56.76) [[all losses: diffusion=0.0927178 ; kl=1.41139e+10 ; lpips=0.255213 ; repa=0.567729 ; sum_loss=14114.2]]
[T_total=19:05:35 | T_train=18:41:54 | T_epoch=00:30:02] Epoch 6, batch 1101 / 6666 (step 41096) loss=14079.9 (avg=51.63) [[all losses: diffusion=0.0926893 ; kl=1.40795e+10 ; lpips=0.255072 ; repa=0.567591 ; sum_loss=14079.9]]
[T_total=19:08:19 | T_train=18:44:38 | T_epoch=00:32:46] Epoch 6, batch 1201 / 6666 (step 41196) loss=14045.7 (avg=47.36) [[all losses: diffusion=0.0926581 ; kl=1.40454e+10 ; lpips=0.254936 ; repa=0.567453 ; sum_loss=14045.7]]
[T_total=19:11:03 | T_train=18:47:22 | T_epoch=00:35:30] Epoch 6, batch 1301 / 6666 (step 41296) loss=14011.7 (avg=43.74) [[all losses: diffusion=0.092632 ; kl=1.40114e+10 ; lpips=0.254795 ; repa=0.567318 ; sum_loss=14011.7]]
[T_total=19:13:47 | T_train=18:50:06 | T_epoch=00:38:14] Epoch 6, batch 1401 / 6666 (step 41396) loss=13977.9 (avg=40.64) [[all losses: diffusion=0.0926049 ; kl=1.39775e+10 ; lpips=0.254653 ; repa=0.567179 ; sum_loss=13977.9]]
[T_total=19:16:31 | T_train=18:52:49 | T_epoch=00:40:58] Epoch 6, batch 1501 / 6666 (step 41496) loss=13944.2 (avg=37.96) [[all losses: diffusion=0.0925773 ; kl=1.39438e+10 ; lpips=0.254512 ; repa=0.567043 ; sum_loss=13944.2]]
[T_total=19:19:14 | T_train=18:55:33 | T_epoch=00:43:41] Epoch 6, batch 1601 / 6666 (step 41596) loss=13910.7 (avg=35.6) [[all losses: diffusion=0.0925499 ; kl=1.39103e+10 ; lpips=0.254374 ; repa=0.566907 ; sum_loss=13910.7]]
[T_total=19:21:58 | T_train=18:58:17 | T_epoch=00:46:25] Epoch 6, batch 1701 / 6666 (step 41696) loss=13877.3 (avg=33.53) [[all losses: diffusion=0.0925224 ; kl=1.38769e+10 ; lpips=0.254232 ; repa=0.566768 ; sum_loss=13877.3]]
[T_total=19:24:42 | T_train=19:01:01 | T_epoch=00:49:09] Epoch 6, batch 1801 / 6666 (step 41796) loss=13844.1 (avg=31.69) [[all losses: diffusion=0.0924922 ; kl=1.38437e+10 ; lpips=0.254094 ; repa=0.56663 ; sum_loss=13844.1]]
[T_total=19:27:26 | T_train=19:03:45 | T_epoch=00:51:53] Epoch 6, batch 1901 / 6666 (step 41896) loss=13811.1 (avg=30.03) [[all losses: diffusion=0.0924626 ; kl=1.38107e+10 ; lpips=0.253956 ; repa=0.566495 ; sum_loss=13811.1]]
[T_total=19:30:10 | T_train=19:06:29 | T_epoch=00:54:37] Epoch 6, batch 2001 / 6666 (step 41996) loss=13778.2 (avg=28.55) [[all losses: diffusion=0.0924365 ; kl=1.37778e+10 ; lpips=0.253819 ; repa=0.566358 ; sum_loss=13778.2]]
[T_total=19:32:54 | T_train=19:09:13 | T_epoch=00:57:21] Epoch 6, batch 2101 / 6666 (step 42096) loss=13745.4 (avg=27.2) [[all losses: diffusion=0.0924074 ; kl=1.37451e+10 ; lpips=0.253683 ; repa=0.566222 ; sum_loss=13745.4]]
[T_total=19:35:38 | T_train=19:11:56 | T_epoch=01:00:05] Epoch 6, batch 2201 / 6666 (step 42196) loss=13712.9 (avg=25.99) [[all losses: diffusion=0.0923837 ; kl=1.37125e+10 ; lpips=0.253571 ; repa=0.566098 ; sum_loss=13712.9]]
[T_total=19:38:22 | T_train=19:14:40 | T_epoch=01:02:49] Epoch 6, batch 2301 / 6666 (step 42296) loss=13680.5 (avg=24.87) [[all losses: diffusion=0.0923545 ; kl=1.36801e+10 ; lpips=0.25344 ; repa=0.565961 ; sum_loss=13680.5]]
[T_total=19:41:06 | T_train=19:17:24 | T_epoch=01:05:33] Epoch 6, batch 2401 / 6666 (step 42396) loss=13648.2 (avg=23.85) [[all losses: diffusion=0.0923274 ; kl=1.36478e+10 ; lpips=0.253303 ; repa=0.565827 ; sum_loss=13648.2]]
[T_total=19:43:49 | T_train=19:20:08 | T_epoch=01:08:16] Epoch 6, batch 2501 / 6666 (step 42496) loss=13616.1 (avg=22.91) [[all losses: diffusion=0.0922981 ; kl=1.36157e+10 ; lpips=0.253168 ; repa=0.565691 ; sum_loss=13616.1]]
[T_total=19:46:33 | T_train=19:22:52 | T_epoch=01:11:00] Epoch 6, batch 2601 / 6666 (step 42596) loss=13648.3 (avg=1073) [[all losses: diffusion=0.0923271 ; kl=1.36479e+10 ; lpips=0.253189 ; repa=0.565649 ; sum_loss=13648.3]]
[T_total=19:49:17 | T_train=19:25:35 | T_epoch=01:13:44] Epoch 6, batch 2701 / 6666 (step 42696) loss=13616.3 (avg=1033) [[all losses: diffusion=0.0923237 ; kl=1.36159e+10 ; lpips=0.253193 ; repa=0.565589 ; sum_loss=13616.3]]
[T_total=19:52:00 | T_train=19:28:19 | T_epoch=01:16:27] Epoch 6, batch 2801 / 6666 (step 42796) loss=13584.6 (avg=998.1) [[all losses: diffusion=0.0923247 ; kl=1.35843e+10 ; lpips=0.253159 ; repa=0.565514 ; sum_loss=13584.6]]
[T_total=19:54:44 | T_train=19:31:02 | T_epoch=01:19:11] Epoch 6, batch 2901 / 6666 (step 42896) loss=13552.9 (avg=963.7) [[all losses: diffusion=0.0923004 ; kl=1.35526e+10 ; lpips=0.253047 ; repa=0.565393 ; sum_loss=13552.9]]
[T_total=19:57:28 | T_train=19:33:46 | T_epoch=01:21:55] Epoch 6, batch 3001 / 6666 (step 42996) loss=13521.4 (avg=931.6) [[all losses: diffusion=0.0922761 ; kl=1.35211e+10 ; lpips=0.252924 ; repa=0.565266 ; sum_loss=13521.4]]
[T_total=20:00:11 | T_train=19:36:30 | T_epoch=01:24:38] Epoch 6, batch 3101 / 6666 (step 43096) loss=13490.1 (avg=901.5) [[all losses: diffusion=0.0922463 ; kl=1.34897e+10 ; lpips=0.252797 ; repa=0.565136 ; sum_loss=13490.1]]
[T_total=20:02:55 | T_train=19:39:13 | T_epoch=01:27:22] Epoch 6, batch 3201 / 6666 (step 43196) loss=13458.8 (avg=873.4) [[all losses: diffusion=0.0922191 ; kl=1.34585e+10 ; lpips=0.252662 ; repa=0.565002 ; sum_loss=13458.8]]
[T_total=20:05:39 | T_train=19:41:57 | T_epoch=01:30:06] Epoch 6, batch 3301 / 6666 (step 43296) loss=13430.5 (avg=882.9) [[all losses: diffusion=0.0922713 ; kl=1.34301e+10 ; lpips=0.252769 ; repa=0.565015 ; sum_loss=13430.5]]
[T_total=20:08:23 | T_train=19:44:41 | T_epoch=01:32:50] Epoch 6, batch 3401 / 6666 (step 43396) loss=13399.5 (avg=857) [[all losses: diffusion=0.0922607 ; kl=1.33992e+10 ; lpips=0.252709 ; repa=0.564926 ; sum_loss=13399.5]]
[T_total=20:11:06 | T_train=19:47:25 | T_epoch=01:35:33] Epoch 6, batch 3501 / 6666 (step 43496) loss=13368.7 (avg=832.5) [[all losses: diffusion=0.0922335 ; kl=1.33684e+10 ; lpips=0.252581 ; repa=0.564798 ; sum_loss=13368.7]]
[T_total=20:13:50 | T_train=19:50:09 | T_epoch=01:38:17] Epoch 6, batch 3601 / 6666 (step 43596) loss=13338.2 (avg=810.7) [[all losses: diffusion=0.0922279 ; kl=1.33378e+10 ; lpips=0.252516 ; repa=0.564706 ; sum_loss=13338.2]]
[T_total=20:16:34 | T_train=19:52:53 | T_epoch=01:41:01] Epoch 6, batch 3701 / 6666 (step 43696) loss=13307.7 (avg=788.8) [[all losses: diffusion=0.0922024 ; kl=1.33073e+10 ; lpips=0.252384 ; repa=0.564576 ; sum_loss=13307.7]]
[T_total=20:19:18 | T_train=19:55:36 | T_epoch=01:43:45] Epoch 6, batch 3801 / 6666 (step 43796) loss=13277.3 (avg=768.1) [[all losses: diffusion=0.0921748 ; kl=1.32769e+10 ; lpips=0.252254 ; repa=0.564449 ; sum_loss=13277.3]]
[T_total=20:22:02 | T_train=19:58:20 | T_epoch=01:46:29] Epoch 6, batch 3901 / 6666 (step 43896) loss=13247 (avg=748.4) [[all losses: diffusion=0.0921464 ; kl=1.32467e+10 ; lpips=0.252128 ; repa=0.56432 ; sum_loss=13247]]
[T_total=20:24:46 | T_train=20:01:04 | T_epoch=01:49:13] Epoch 6, batch 4001 / 6666 (step 43996) loss=13216.9 (avg=729.7) [[all losses: diffusion=0.0921193 ; kl=1.32166e+10 ; lpips=0.251996 ; repa=0.564192 ; sum_loss=13216.9]]
[T_total=20:27:29 | T_train=20:03:48 | T_epoch=01:51:56] Epoch 6, batch 4101 / 6666 (step 44096) loss=13186.9 (avg=711.9) [[all losses: diffusion=0.0920894 ; kl=1.31866e+10 ; lpips=0.25187 ; repa=0.564063 ; sum_loss=13186.9]]
[T_total=20:30:13 | T_train=20:06:31 | T_epoch=01:54:40] Epoch 6, batch 4201 / 6666 (step 44196) loss=13157.1 (avg=695) [[all losses: diffusion=0.0920615 ; kl=1.31567e+10 ; lpips=0.251742 ; repa=0.563934 ; sum_loss=13157.1]]
[T_total=20:32:57 | T_train=20:09:15 | T_epoch=01:57:24] Epoch 6, batch 4301 / 6666 (step 44296) loss=13127.4 (avg=678.8) [[all losses: diffusion=0.0920343 ; kl=1.3127e+10 ; lpips=0.251612 ; repa=0.563806 ; sum_loss=13127.4]]
[T_total=20:35:40 | T_train=20:11:59 | T_epoch=02:00:07] Epoch 6, batch 4401 / 6666 (step 44396) loss=13097.8 (avg=663.5) [[all losses: diffusion=0.0920658 ; kl=1.30975e+10 ; lpips=0.251678 ; repa=0.56379 ; sum_loss=13097.8]]
[T_total=20:38:24 | T_train=20:14:42 | T_epoch=02:02:51] Epoch 6, batch 4501 / 6666 (step 44496) loss=13075.1 (avg=714.5) [[all losses: diffusion=0.0920662 ; kl=1.30747e+10 ; lpips=0.251649 ; repa=0.563722 ; sum_loss=13075.1]]
[T_total=20:41:08 | T_train=20:17:26 | T_epoch=02:05:35] Epoch 6, batch 4601 / 6666 (step 44596) loss=13045.7 (avg=699) [[all losses: diffusion=0.0920435 ; kl=1.30454e+10 ; lpips=0.251542 ; repa=0.563606 ; sum_loss=13045.7]]
[T_total=20:43:52 | T_train=20:20:10 | T_epoch=02:08:19] Epoch 6, batch 4701 / 6666 (step 44696) loss=13016.6 (avg=684.1) [[all losses: diffusion=0.0920183 ; kl=1.30162e+10 ; lpips=0.251417 ; repa=0.563481 ; sum_loss=13016.6]]
[T_total=20:46:35 | T_train=20:22:54 | T_epoch=02:11:02] Epoch 6, batch 4801 / 6666 (step 44796) loss=12987.5 (avg=669.9) [[all losses: diffusion=0.0919908 ; kl=1.29871e+10 ; lpips=0.251294 ; repa=0.563356 ; sum_loss=12987.5]]
[T_total=20:49:19 | T_train=20:25:38 | T_epoch=02:13:46] Epoch 6, batch 4901 / 6666 (step 44896) loss=12958.6 (avg=656.2) [[all losses: diffusion=0.0919652 ; kl=1.29582e+10 ; lpips=0.251166 ; repa=0.563231 ; sum_loss=12958.6]]
[T_total=20:52:03 | T_train=20:28:21 | T_epoch=02:16:30] Epoch 6, batch 5001 / 6666 (step 44996) loss=12929.8 (avg=643.1) [[all losses: diffusion=0.0919391 ; kl=1.29294e+10 ; lpips=0.251043 ; repa=0.563103 ; sum_loss=12929.8]]
[T_total=20:54:47 | T_train=20:31:05 | T_epoch=02:19:14] Epoch 6, batch 5101 / 6666 (step 45096) loss=12901.1 (avg=630.5) [[all losses: diffusion=0.0919177 ; kl=1.29008e+10 ; lpips=0.250928 ; repa=0.562984 ; sum_loss=12901.1]]
[T_total=20:57:31 | T_train=20:33:49 | T_epoch=02:21:58] Epoch 6, batch 5201 / 6666 (step 45196) loss=12872.6 (avg=618.4) [[all losses: diffusion=0.091892 ; kl=1.28722e+10 ; lpips=0.2508 ; repa=0.562859 ; sum_loss=12872.6]]
[T_total=21:00:15 | T_train=20:36:33 | T_epoch=02:24:42] Epoch 6, batch 5301 / 6666 (step 45296) loss=12844.1 (avg=606.7) [[all losses: diffusion=0.0918656 ; kl=1.28438e+10 ; lpips=0.250681 ; repa=0.562736 ; sum_loss=12844.1]]
[T_total=21:02:58 | T_train=20:39:17 | T_epoch=02:27:25] Epoch 6, batch 5401 / 6666 (step 45396) loss=12815.9 (avg=595.5) [[all losses: diffusion=0.0918424 ; kl=1.28155e+10 ; lpips=0.250558 ; repa=0.56261 ; sum_loss=12815.9]]
[T_total=21:05:42 | T_train=20:42:01 | T_epoch=02:30:09] Epoch 6, batch 5501 / 6666 (step 45496) loss=12787.7 (avg=584.7) [[all losses: diffusion=0.0918137 ; kl=1.27873e+10 ; lpips=0.250436 ; repa=0.562486 ; sum_loss=12787.7]]
[T_total=21:08:26 | T_train=20:44:44 | T_epoch=02:32:53] Epoch 6, batch 5601 / 6666 (step 45596) loss=12759.6 (avg=574.3) [[all losses: diffusion=0.0917869 ; kl=1.27593e+10 ; lpips=0.250311 ; repa=0.56236 ; sum_loss=12759.6]]
[T_total=21:11:09 | T_train=20:47:28 | T_epoch=02:35:36] Epoch 6, batch 5701 / 6666 (step 45696) loss=12731.7 (avg=564.2) [[all losses: diffusion=0.0917586 ; kl=1.27314e+10 ; lpips=0.250189 ; repa=0.562237 ; sum_loss=12731.7]]
[T_total=21:13:53 | T_train=20:50:12 | T_epoch=02:38:20] Epoch 6, batch 5801 / 6666 (step 45796) loss=12703.9 (avg=554.5) [[all losses: diffusion=0.0917339 ; kl=1.27036e+10 ; lpips=0.250066 ; repa=0.562112 ; sum_loss=12703.9]]
[T_total=21:16:37 | T_train=20:52:56 | T_epoch=02:41:04] Epoch 6, batch 5901 / 6666 (step 45896) loss=12676.2 (avg=545.1) [[all losses: diffusion=0.0917062 ; kl=1.26759e+10 ; lpips=0.249946 ; repa=0.561988 ; sum_loss=12676.2]]
[T_total=21:19:21 | T_train=20:55:39 | T_epoch=02:43:48] Epoch 6, batch 6001 / 6666 (step 45996) loss=12648.7 (avg=536) [[all losses: diffusion=0.0916767 ; kl=1.26483e+10 ; lpips=0.249831 ; repa=0.561865 ; sum_loss=12648.7]]
[T_total=21:22:04 | T_train=20:58:23 | T_epoch=02:46:31] Epoch 6, batch 6101 / 6666 (step 46096) loss=12621.2 (avg=527.2) [[all losses: diffusion=0.0916525 ; kl=1.26209e+10 ; lpips=0.249707 ; repa=0.561742 ; sum_loss=12621.2]]
[T_total=21:24:48 | T_train=21:01:07 | T_epoch=02:49:15] Epoch 6, batch 6201 / 6666 (step 46196) loss=12593.9 (avg=518.7) [[all losses: diffusion=0.0916269 ; kl=1.25936e+10 ; lpips=0.249585 ; repa=0.561619 ; sum_loss=12593.9]]
[T_total=21:27:32 | T_train=21:03:50 | T_epoch=02:51:59] Epoch 6, batch 6301 / 6666 (step 46296) loss=12566.7 (avg=510.5) [[all losses: diffusion=0.0915993 ; kl=1.25664e+10 ; lpips=0.249465 ; repa=0.561499 ; sum_loss=12566.7]]
[T_total=21:30:16 | T_train=21:06:34 | T_epoch=02:54:43] Epoch 6, batch 6401 / 6666 (step 46396) loss=12539.6 (avg=502.5) [[all losses: diffusion=0.0915726 ; kl=1.25393e+10 ; lpips=0.249347 ; repa=0.561377 ; sum_loss=12539.6]]
[T_total=21:32:59 | T_train=21:09:18 | T_epoch=02:57:26] Epoch 6, batch 6501 / 6666 (step 46496) loss=12512.7 (avg=494.8) [[all losses: diffusion=0.0915468 ; kl=1.25123e+10 ; lpips=0.249225 ; repa=0.561255 ; sum_loss=12512.7]]
[T_total=21:35:43 | T_train=21:12:02 | T_epoch=03:00:10] Epoch 6, batch 6601 / 6666 (step 46596) loss=12485.8 (avg=487.3) [[all losses: diffusion=0.0915207 ; kl=1.24855e+10 ; lpips=0.249104 ; repa=0.561136 ; sum_loss=12485.8]]
[2025-10-26 01:48:51,311][main][INFO] - [T_total=21:37:30 | T_train=21:13:48 | T_epoch=03:01:57] End of epoch 6 (46662 steps) train loss 482.556[2025-10-26 01:48:51,313][main][INFO] - [Epoch 6] All losses: [[diffusion=0.0830356 ; kl=4.82242e+08 ; lpips=0.205621 ; repa=0.513838]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:01<06:29,  1.50s/it]
Reconstructing from test set:   1%|          | 2/261 [00:02<04:40,  1.08s/it]
Reconstructing from test set:   1%|          | 3/261 [00:03<04:03,  1.06it/s]
Reconstructing from test set:   2%|▏         | 4/261 [00:03<03:46,  1.13it/s]
Reconstructing from test set:   2%|▏         | 5/261 [00:04<03:36,  1.18it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:05<03:30,  1.21it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:06<03:25,  1.23it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:06<03:23,  1.25it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:07<03:20,  1.26it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:08<03:19,  1.26it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:09<03:18,  1.26it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:10<03:17,  1.26it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:10<03:15,  1.27it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:11<03:14,  1.27it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:12<03:12,  1.28it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:13<03:11,  1.28it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:14<03:11,  1.28it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:14<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:15<03:09,  1.28it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:16<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:17<03:07,  1.28it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:17<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:18<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:19<03:04,  1.28it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:20<03:03,  1.28it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:21<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:22<03:01,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:23<03:00,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:24<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:24<02:59,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:25<02:58,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:26<02:57,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:27<02:56,  1.29it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:28<02:55,  1.29it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:28<02:54,  1.29it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:29<02:54,  1.29it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:30<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:31<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:31<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:32<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:33<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:34<02:50,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:35<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:36<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:37<02:46,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:38<02:45,  1.29it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:38<02:45,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:39<02:44,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:40<02:43,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:41<02:43,  1.28it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:42<02:42,  1.28it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:42<02:41,  1.28it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:43<02:40,  1.28it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:44<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:45<02:38,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:47<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:48<02:36,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:49<02:35,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:49<02:35,  1.27it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:50<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:51<02:33,  1.28it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:52<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:53<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:53<02:30,  1.28it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:54<02:30,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:55<02:29,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:56<02:29,  1.27it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:56<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:57<02:27,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:58<02:26,  1.27it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [00:59<02:25,  1.28it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [01:00<02:24,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:00<02:23,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:01<02:23,  1.28it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:02<02:22,  1.28it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:03<02:21,  1.28it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:04<02:19,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:05<02:19,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:06<02:18,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:07<02:18,  1.27it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:07<02:17,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:08<02:16,  1.28it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:09<02:15,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:10<02:14,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:11<02:14,  1.28it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:11<02:13,  1.27it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:12<02:12,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:13<02:11,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:14<02:10,  1.28it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:15<02:09,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:15<02:08,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:16<02:08,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:17<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:18<02:06,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:18<02:05,  1.28it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:19<02:05,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:20<02:04,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:21<02:03,  1.28it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:22<02:03,  1.28it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:22<02:02,  1.28it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:23<02:01,  1.28it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:24<02:00,  1.27it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:25<02:00,  1.27it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:25<01:59,  1.27it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:26<01:58,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:27<01:57,  1.27it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:28<01:56,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:29<01:55,  1.28it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:29<01:54,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:30<01:54,  1.27it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:31<01:53,  1.27it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:32<01:52,  1.27it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:33<01:52,  1.27it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:33<01:51,  1.27it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:34<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:35<01:49,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:36<01:48,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:36<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:37<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:38<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:39<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:41<01:43,  1.27it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:42<01:42,  1.28it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:43<01:41,  1.28it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:43<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:44<01:40,  1.27it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:45<01:39,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:46<01:38,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:47<01:37,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:47<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:48<01:36,  1.27it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:49<01:35,  1.27it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:50<01:35,  1.27it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:51<01:34,  1.27it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:51<01:33,  1.28it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:52<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:53<01:31,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:54<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:54<01:29,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:55<01:29,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:56<01:28,  1.27it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:57<01:27,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:58<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [01:58<01:25,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [01:59<01:25,  1.28it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:00<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:01<01:23,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:01<01:22,  1.28it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:03<01:21,  1.28it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:04<01:20,  1.28it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:05<01:20,  1.27it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:05<01:19,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:06<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:07<01:17,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:08<01:16,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:09<01:15,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:09<01:15,  1.28it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:10<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:11<01:13,  1.27it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:12<01:13,  1.27it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:12<01:12,  1.27it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:13<01:11,  1.27it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:14<01:10,  1.27it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:15<01:09,  1.28it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:16<01:09,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:16<01:08,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:17<01:07,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:18<01:06,  1.27it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:19<01:06,  1.27it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:20<01:05,  1.27it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:20<01:04,  1.27it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:21<01:03,  1.27it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:22<01:02,  1.27it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:23<01:02,  1.27it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:23<01:01,  1.27it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:24<01:00,  1.27it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:25<00:59,  1.27it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:26<00:59,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:27<00:58,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:27<00:57,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:28<00:56,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:29<00:55,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:30<00:54,  1.28it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:31<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:31<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:32<00:52,  1.28it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:33<00:51,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:34<00:50,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:34<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:35<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:36<00:48,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:37<00:47,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:38<00:47,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:38<00:46,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:39<00:45,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:40<00:44,  1.28it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:41<00:43,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:41<00:43,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:42<00:42,  1.28it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:43<00:41,  1.28it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:44<00:40,  1.27it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:45<00:40,  1.27it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:45<00:39,  1.27it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:46<00:38,  1.27it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:47<00:37,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:48<00:36,  1.27it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:49<00:36,  1.27it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:49<00:35,  1.27it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:50<00:34,  1.27it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:51<00:33,  1.27it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:52<00:32,  1.27it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:52<00:32,  1.27it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:53<00:31,  1.27it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:54<00:30,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:55<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:56<00:29,  1.27it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:56<00:28,  1.27it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:57<00:27,  1.27it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:58<00:26,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [02:59<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [03:00<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:00<00:24,  1.27it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:01<00:23,  1.27it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:02<00:22,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:03<00:21,  1.28it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:03<00:21,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:04<00:20,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:05<00:19,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:06<00:18,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:08<00:16,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:09<00:15,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:10<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:10<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:11<00:13,  1.28it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:12<00:12,  1.28it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:13<00:11,  1.28it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:14<00:10,  1.28it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:14<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:15<00:09,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:16<00:08,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:17<00:07,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:17<00:07,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:18<00:06,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:19<00:05,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:20<00:04,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:21<00:03,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:21<00:03,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:22<00:02,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:23<00:01,  1.28it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.27it/s]
[2025-10-26 01:52:19,589][main][INFO] - [Epoch 7] Test metrics: [[MSE=26.51 | MAE=0.1158 | LPIPS=0.1917 | PSNR=15.77 | SSIM=0.3748 | dreamsim=0.2921 | FID=31.73]][2025-10-26 01:52:19,591][main][INFO] - [Epoch 7] Best metrics: [[min_MSE=26.51 | min_MAE=0.1158 | min_LPIPS=0.1917 | max_PSNR=15.77 | max_SSIM=0.3748 | min_dreamsim=0.2921 | min_FID=31.73]][2025-10-26 01:52:19,592][main][DEBUG] - Writing images to disk...[2025-10-26 01:52:20,433][main][DEBUG] - Image(s) saved on disk[2025-10-26 01:52:20,683][main][INFO] - End of epoch timers: [T_train=21:13:48 | T_epoch=03:01:57 | T_eval=00:24:26 | T_total=21:40:59][2025-10-26 01:52:20,685][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-26 01:52:32,727][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-26 01:52:44,520][main][INFO] - ---

[2025-10-26 01:52:44,521][main][INFO] - [T_total=21:41:23 | T_train=21:13:48] Start epoch 7[T_total=21:41:25 | T_train=21:13:51 | T_epoch=00:00:02] Epoch 7, batch 1 / 6666 (step 46662) loss=12468.2 (avg=0.3002) [[all losses: diffusion=0.0915032 ; kl=1.24678e+10 ; lpips=0.249028 ; repa=0.561056 ; sum_loss=12468.2]]
[T_total=21:44:09 | T_train=21:16:34 | T_epoch=00:02:46] Epoch 7, batch 101 / 6666 (step 46762) loss=12441.7 (avg=72.54) [[all losses: diffusion=0.0915034 ; kl=1.24413e+10 ; lpips=0.248994 ; repa=0.560985 ; sum_loss=12441.7]]
[T_total=21:46:53 | T_train=21:19:18 | T_epoch=00:05:29] Epoch 7, batch 201 / 6666 (step 46862) loss=12415.1 (avg=36.6) [[all losses: diffusion=0.0914758 ; kl=1.24148e+10 ; lpips=0.248884 ; repa=0.560866 ; sum_loss=12415.1]]
[T_total=21:49:36 | T_train=21:22:02 | T_epoch=00:08:13] Epoch 7, batch 301 / 6666 (step 46962) loss=12388.7 (avg=24.54) [[all losses: diffusion=0.0914523 ; kl=1.23883e+10 ; lpips=0.248768 ; repa=0.560748 ; sum_loss=12388.7]]
[T_total=21:52:20 | T_train=21:24:46 | T_epoch=00:10:57] Epoch 7, batch 401 / 6666 (step 47062) loss=12362.3 (avg=18.5) [[all losses: diffusion=0.0914262 ; kl=1.2362e+10 ; lpips=0.24865 ; repa=0.56063 ; sum_loss=12362.3]]
[T_total=21:55:04 | T_train=21:27:30 | T_epoch=00:13:41] Epoch 7, batch 501 / 6666 (step 47162) loss=12336.1 (avg=14.87) [[all losses: diffusion=0.0914028 ; kl=1.23358e+10 ; lpips=0.248529 ; repa=0.560508 ; sum_loss=12336.1]]
[T_total=21:57:48 | T_train=21:30:13 | T_epoch=00:16:25] Epoch 7, batch 601 / 6666 (step 47262) loss=12310 (avg=12.44) [[all losses: diffusion=0.0913789 ; kl=1.23097e+10 ; lpips=0.248409 ; repa=0.560387 ; sum_loss=12310]]
[T_total=22:00:32 | T_train=21:32:58 | T_epoch=00:19:09] Epoch 7, batch 701 / 6666 (step 47362) loss=12284 (avg=10.71) [[all losses: diffusion=0.0913547 ; kl=1.22837e+10 ; lpips=0.248289 ; repa=0.560269 ; sum_loss=12284]]
[T_total=22:03:16 | T_train=21:35:41 | T_epoch=00:21:52] Epoch 7, batch 801 / 6666 (step 47462) loss=12258.2 (avg=9.413) [[all losses: diffusion=0.0913299 ; kl=1.22578e+10 ; lpips=0.248169 ; repa=0.56015 ; sum_loss=12258.2]]
[T_total=22:05:59 | T_train=21:38:25 | T_epoch=00:24:36] Epoch 7, batch 901 / 6666 (step 47562) loss=12232.4 (avg=8.402) [[all losses: diffusion=0.0913087 ; kl=1.2232e+10 ; lpips=0.248047 ; repa=0.560031 ; sum_loss=12232.4]]
[T_total=22:08:43 | T_train=21:41:08 | T_epoch=00:27:20] Epoch 7, batch 1001 / 6666 (step 47662) loss=12206.7 (avg=7.593) [[all losses: diffusion=0.0912865 ; kl=1.22064e+10 ; lpips=0.24793 ; repa=0.559912 ; sum_loss=12206.7]]
[T_total=22:11:27 | T_train=21:43:52 | T_epoch=00:30:03] Epoch 7, batch 1101 / 6666 (step 47762) loss=12181.2 (avg=6.931) [[all losses: diffusion=0.0912624 ; kl=1.21808e+10 ; lpips=0.247813 ; repa=0.559793 ; sum_loss=12181.2]]
[T_total=22:14:11 | T_train=21:46:36 | T_epoch=00:32:47] Epoch 7, batch 1201 / 6666 (step 47862) loss=12155.7 (avg=6.379) [[all losses: diffusion=0.0912381 ; kl=1.21554e+10 ; lpips=0.247699 ; repa=0.559678 ; sum_loss=12155.7]]
[T_total=22:16:54 | T_train=21:49:19 | T_epoch=00:35:31] Epoch 7, batch 1301 / 6666 (step 47962) loss=12130.4 (avg=5.912) [[all losses: diffusion=0.091213 ; kl=1.213e+10 ; lpips=0.247583 ; repa=0.55956 ; sum_loss=12130.4]]
[T_total=22:19:38 | T_train=21:52:03 | T_epoch=00:38:14] Epoch 7, batch 1401 / 6666 (step 48062) loss=12105.1 (avg=5.512) [[all losses: diffusion=0.0911887 ; kl=1.21048e+10 ; lpips=0.247469 ; repa=0.559444 ; sum_loss=12105.1]]
[T_total=22:22:21 | T_train=21:54:47 | T_epoch=00:40:58] Epoch 7, batch 1501 / 6666 (step 48162) loss=12080 (avg=5.165) [[all losses: diffusion=0.0911653 ; kl=1.20797e+10 ; lpips=0.247353 ; repa=0.559329 ; sum_loss=12080]]
[T_total=22:25:05 | T_train=21:57:30 | T_epoch=00:43:42] Epoch 7, batch 1601 / 6666 (step 48262) loss=12055 (avg=4.861) [[all losses: diffusion=0.0911439 ; kl=1.20546e+10 ; lpips=0.247237 ; repa=0.559214 ; sum_loss=12055]]
[T_total=22:27:49 | T_train=22:00:14 | T_epoch=00:46:26] Epoch 7, batch 1701 / 6666 (step 48362) loss=12030.1 (avg=4.593) [[all losses: diffusion=0.0911215 ; kl=1.20297e+10 ; lpips=0.24712 ; repa=0.559098 ; sum_loss=12030.1]]
[T_total=22:30:33 | T_train=22:02:58 | T_epoch=00:49:09] Epoch 7, batch 1801 / 6666 (step 48462) loss=12005.2 (avg=4.355) [[all losses: diffusion=0.0910967 ; kl=1.20049e+10 ; lpips=0.247006 ; repa=0.558982 ; sum_loss=12005.2]]
[T_total=22:33:17 | T_train=22:05:42 | T_epoch=00:51:53] Epoch 7, batch 1901 / 6666 (step 48562) loss=11980.5 (avg=4.142) [[all losses: diffusion=0.0910734 ; kl=1.19802e+10 ; lpips=0.246892 ; repa=0.558866 ; sum_loss=11980.5]]
[T_total=22:36:01 | T_train=22:08:26 | T_epoch=00:54:37] Epoch 7, batch 2001 / 6666 (step 48662) loss=11955.9 (avg=3.95) [[all losses: diffusion=0.0910515 ; kl=1.19555e+10 ; lpips=0.246781 ; repa=0.55875 ; sum_loss=11955.9]]
[T_total=22:38:44 | T_train=22:11:09 | T_epoch=00:57:21] Epoch 7, batch 2101 / 6666 (step 48762) loss=11931.4 (avg=4.074) [[all losses: diffusion=0.0910621 ; kl=1.1931e+10 ; lpips=0.246784 ; repa=0.558698 ; sum_loss=11931.4]]
[T_total=22:41:28 | T_train=22:13:53 | T_epoch=01:00:05] Epoch 7, batch 2201 / 6666 (step 48862) loss=11907 (avg=3.903) [[all losses: diffusion=0.0910413 ; kl=1.19066e+10 ; lpips=0.246669 ; repa=0.558584 ; sum_loss=11907]]
[T_total=22:44:12 | T_train=22:16:37 | T_epoch=01:02:48] Epoch 7, batch 2301 / 6666 (step 48962) loss=11882.7 (avg=3.747) [[all losses: diffusion=0.0910195 ; kl=1.18823e+10 ; lpips=0.246558 ; repa=0.55847 ; sum_loss=11882.7]]
[T_total=22:46:55 | T_train=22:19:21 | T_epoch=01:05:32] Epoch 7, batch 2401 / 6666 (step 49062) loss=11858.4 (avg=3.603) [[all losses: diffusion=0.090998 ; kl=1.18581e+10 ; lpips=0.246446 ; repa=0.558356 ; sum_loss=11858.4]]
[T_total=22:49:39 | T_train=22:22:04 | T_epoch=01:08:16] Epoch 7, batch 2501 / 6666 (step 49162) loss=11834.3 (avg=3.471) [[all losses: diffusion=0.0909733 ; kl=1.1834e+10 ; lpips=0.246337 ; repa=0.558241 ; sum_loss=11834.3]]
[T_total=22:52:23 | T_train=22:24:48 | T_epoch=01:11:00] Epoch 7, batch 2601 / 6666 (step 49262) loss=11810.3 (avg=3.35) [[all losses: diffusion=0.0909507 ; kl=1.18099e+10 ; lpips=0.246228 ; repa=0.558126 ; sum_loss=11810.3]]
[T_total=22:55:07 | T_train=22:27:32 | T_epoch=01:13:43] Epoch 7, batch 2701 / 6666 (step 49362) loss=11786.4 (avg=3.237) [[all losses: diffusion=0.0909271 ; kl=1.1786e+10 ; lpips=0.246117 ; repa=0.558011 ; sum_loss=11786.4]]
[T_total=22:57:51 | T_train=22:30:16 | T_epoch=01:16:27] Epoch 7, batch 2801 / 6666 (step 49462) loss=11762.5 (avg=3.132) [[all losses: diffusion=0.0909043 ; kl=1.17622e+10 ; lpips=0.246005 ; repa=0.557898 ; sum_loss=11762.5]]
[T_total=23:00:34 | T_train=22:33:00 | T_epoch=01:19:11] Epoch 7, batch 2901 / 6666 (step 49562) loss=11738.8 (avg=3.035) [[all losses: diffusion=0.090885 ; kl=1.17385e+10 ; lpips=0.245891 ; repa=0.557786 ; sum_loss=11738.8]]
[T_total=23:03:18 | T_train=22:35:43 | T_epoch=01:21:55] Epoch 7, batch 3001 / 6666 (step 49662) loss=11715.2 (avg=2.943) [[all losses: diffusion=0.0908624 ; kl=1.17148e+10 ; lpips=0.245779 ; repa=0.55767 ; sum_loss=11715.2]]
[T_total=23:06:02 | T_train=22:38:27 | T_epoch=01:24:38] Epoch 7, batch 3101 / 6666 (step 49762) loss=11691.6 (avg=2.858) [[all losses: diffusion=0.0908418 ; kl=1.16913e+10 ; lpips=0.245666 ; repa=0.557557 ; sum_loss=11691.6]]
[T_total=23:08:46 | T_train=22:41:11 | T_epoch=01:27:22] Epoch 7, batch 3201 / 6666 (step 49862) loss=11668.2 (avg=2.778) [[all losses: diffusion=0.0908173 ; kl=1.16678e+10 ; lpips=0.245562 ; repa=0.557447 ; sum_loss=11668.2]]
[T_total=23:11:29 | T_train=22:43:54 | T_epoch=01:30:06] Epoch 7, batch 3301 / 6666 (step 49962) loss=11644.8 (avg=2.703) [[all losses: diffusion=0.0907946 ; kl=1.16445e+10 ; lpips=0.245453 ; repa=0.557335 ; sum_loss=11644.8]]
[T_total=23:14:13 | T_train=22:46:38 | T_epoch=01:32:49] Epoch 7, batch 3401 / 6666 (step 50062) loss=11621.6 (avg=2.633) [[all losses: diffusion=0.0907744 ; kl=1.16212e+10 ; lpips=0.245343 ; repa=0.557221 ; sum_loss=11621.6]]
[T_total=23:16:56 | T_train=22:49:22 | T_epoch=01:35:33] Epoch 7, batch 3501 / 6666 (step 50162) loss=11598.4 (avg=2.67) [[all losses: diffusion=0.0908088 ; kl=1.15981e+10 ; lpips=0.245414 ; repa=0.557223 ; sum_loss=11598.4]]
[T_total=23:19:40 | T_train=22:52:05 | T_epoch=01:38:17] Epoch 7, batch 3601 / 6666 (step 50262) loss=11575.3 (avg=2.605) [[all losses: diffusion=0.0907892 ; kl=1.1575e+10 ; lpips=0.245348 ; repa=0.55713 ; sum_loss=11575.3]]
[T_total=23:22:24 | T_train=22:54:49 | T_epoch=01:41:00] Epoch 7, batch 3701 / 6666 (step 50362) loss=11552.4 (avg=2.543) [[all losses: diffusion=0.0907676 ; kl=1.1552e+10 ; lpips=0.245238 ; repa=0.557018 ; sum_loss=11552.4]]
[T_total=23:25:08 | T_train=22:57:33 | T_epoch=01:43:44] Epoch 7, batch 3801 / 6666 (step 50462) loss=11529.5 (avg=2.484) [[all losses: diffusion=0.0907471 ; kl=1.15291e+10 ; lpips=0.24513 ; repa=0.556907 ; sum_loss=11529.5]]
[T_total=23:27:51 | T_train=23:00:16 | T_epoch=01:46:28] Epoch 7, batch 3901 / 6666 (step 50562) loss=11506.7 (avg=2.428) [[all losses: diffusion=0.0907279 ; kl=1.15063e+10 ; lpips=0.245027 ; repa=0.5568 ; sum_loss=11506.7]]
[T_total=23:30:35 | T_train=23:03:00 | T_epoch=01:49:12] Epoch 7, batch 4001 / 6666 (step 50662) loss=11483.9 (avg=2.375) [[all losses: diffusion=0.0907071 ; kl=1.14836e+10 ; lpips=0.244931 ; repa=0.556695 ; sum_loss=11483.9]]
[T_total=23:33:19 | T_train=23:05:44 | T_epoch=01:51:55] Epoch 7, batch 4101 / 6666 (step 50762) loss=11461.3 (avg=2.325) [[all losses: diffusion=0.0906858 ; kl=1.1461e+10 ; lpips=0.244825 ; repa=0.556583 ; sum_loss=11461.3]]
[T_total=23:36:02 | T_train=23:08:28 | T_epoch=01:54:39] Epoch 7, batch 4201 / 6666 (step 50862) loss=11438.8 (avg=2.276) [[all losses: diffusion=0.0906659 ; kl=1.14384e+10 ; lpips=0.244714 ; repa=0.556473 ; sum_loss=11438.8]]
[T_total=23:38:46 | T_train=23:11:11 | T_epoch=01:57:23] Epoch 7, batch 4301 / 6666 (step 50962) loss=11416.3 (avg=2.231) [[all losses: diffusion=0.0906455 ; kl=1.1416e+10 ; lpips=0.244607 ; repa=0.556363 ; sum_loss=11416.3]]
[T_total=23:41:30 | T_train=23:13:55 | T_epoch=02:00:06] Epoch 7, batch 4401 / 6666 (step 51062) loss=11394 (avg=2.187) [[all losses: diffusion=0.0906297 ; kl=1.13936e+10 ; lpips=0.244517 ; repa=0.556264 ; sum_loss=11394]]
[T_total=23:44:13 | T_train=23:16:39 | T_epoch=02:02:50] Epoch 7, batch 4501 / 6666 (step 51162) loss=11371.7 (avg=2.145) [[all losses: diffusion=0.090607 ; kl=1.13714e+10 ; lpips=0.244409 ; repa=0.556156 ; sum_loss=11371.7]]
[T_total=23:46:57 | T_train=23:19:22 | T_epoch=02:05:34] Epoch 7, batch 4601 / 6666 (step 51262) loss=11349.5 (avg=2.105) [[all losses: diffusion=0.0905873 ; kl=1.13492e+10 ; lpips=0.244303 ; repa=0.556049 ; sum_loss=11349.5]]
[T_total=23:49:41 | T_train=23:22:06 | T_epoch=02:08:17] Epoch 7, batch 4701 / 6666 (step 51362) loss=11327.4 (avg=2.067) [[all losses: diffusion=0.0905668 ; kl=1.13271e+10 ; lpips=0.244196 ; repa=0.555941 ; sum_loss=11327.4]]
[T_total=23:52:25 | T_train=23:24:50 | T_epoch=02:11:01] Epoch 7, batch 4801 / 6666 (step 51462) loss=11305.4 (avg=2.03) [[all losses: diffusion=0.0905455 ; kl=1.13051e+10 ; lpips=0.244091 ; repa=0.55583 ; sum_loss=11305.4]]
[T_total=23:55:08 | T_train=23:27:34 | T_epoch=02:13:45] Epoch 7, batch 4901 / 6666 (step 51562) loss=11283.5 (avg=1.995) [[all losses: diffusion=0.0905228 ; kl=1.12832e+10 ; lpips=0.243987 ; repa=0.555723 ; sum_loss=11283.5]]
[T_total=23:57:52 | T_train=23:30:17 | T_epoch=02:16:29] Epoch 7, batch 5001 / 6666 (step 51662) loss=11261.7 (avg=1.961) [[all losses: diffusion=0.090502 ; kl=1.12613e+10 ; lpips=0.243881 ; repa=0.555615 ; sum_loss=11261.7]]
[T_total=24:00:36 | T_train=23:33:01 | T_epoch=02:19:13] Epoch 7, batch 5101 / 6666 (step 51762) loss=11239.9 (avg=2.077) [[all losses: diffusion=0.090491 ; kl=1.12396e+10 ; lpips=0.243808 ; repa=0.555527 ; sum_loss=11239.9]]
[T_total=24:03:20 | T_train=23:35:45 | T_epoch=02:21:56] Epoch 7, batch 5201 / 6666 (step 51862) loss=11218.5 (avg=4.304) [[all losses: diffusion=0.0905242 ; kl=1.12181e+10 ; lpips=0.2439 ; repa=0.555533 ; sum_loss=11218.5]]
[T_total=24:06:03 | T_train=23:38:29 | T_epoch=02:24:40] Epoch 7, batch 5301 / 6666 (step 51962) loss=11196.9 (avg=4.228) [[all losses: diffusion=0.0905027 ; kl=1.11965e+10 ; lpips=0.2438 ; repa=0.555427 ; sum_loss=11196.9]]
[T_total=24:08:47 | T_train=23:41:12 | T_epoch=02:27:23] Epoch 7, batch 5401 / 6666 (step 52062) loss=11175.4 (avg=4.156) [[all losses: diffusion=0.0904849 ; kl=1.1175e+10 ; lpips=0.243706 ; repa=0.555326 ; sum_loss=11175.4]]
[T_total=24:11:31 | T_train=23:43:56 | T_epoch=02:30:07] Epoch 7, batch 5501 / 6666 (step 52162) loss=11154 (avg=4.086) [[all losses: diffusion=0.0904636 ; kl=1.11536e+10 ; lpips=0.243605 ; repa=0.555221 ; sum_loss=11154]]
[T_total=24:14:15 | T_train=23:46:40 | T_epoch=02:32:51] Epoch 7, batch 5601 / 6666 (step 52262) loss=11132.6 (avg=4.249) [[all losses: diffusion=0.0904664 ; kl=1.11323e+10 ; lpips=0.243587 ; repa=0.555168 ; sum_loss=11132.6]]
[T_total=24:16:58 | T_train=23:49:24 | T_epoch=02:35:35] Epoch 7, batch 5701 / 6666 (step 52362) loss=11111.4 (avg=4.185) [[all losses: diffusion=0.0904557 ; kl=1.1111e+10 ; lpips=0.243517 ; repa=0.555081 ; sum_loss=11111.4]]
[T_total=24:19:42 | T_train=23:52:07 | T_epoch=02:38:19] Epoch 7, batch 5801 / 6666 (step 52462) loss=11090.2 (avg=4.118) [[all losses: diffusion=0.0904331 ; kl=1.10899e+10 ; lpips=0.243415 ; repa=0.554976 ; sum_loss=11090.2]]
[T_total=24:22:26 | T_train=23:54:51 | T_epoch=02:41:02] Epoch 7, batch 5901 / 6666 (step 52562) loss=11069.1 (avg=4.053) [[all losses: diffusion=0.0904115 ; kl=1.10688e+10 ; lpips=0.243313 ; repa=0.55487 ; sum_loss=11069.1]]
[T_total=24:25:10 | T_train=23:57:35 | T_epoch=02:43:46] Epoch 7, batch 6001 / 6666 (step 52662) loss=11048.1 (avg=3.991) [[all losses: diffusion=0.0903905 ; kl=1.10477e+10 ; lpips=0.243213 ; repa=0.554763 ; sum_loss=11048.1]]
[T_total=24:27:53 | T_train=24:00:19 | T_epoch=02:46:30] Epoch 7, batch 6101 / 6666 (step 52762) loss=11027.2 (avg=3.93) [[all losses: diffusion=0.0903683 ; kl=1.10268e+10 ; lpips=0.243115 ; repa=0.554658 ; sum_loss=11027.2]]
[T_total=24:30:37 | T_train=24:03:02 | T_epoch=02:49:14] Epoch 7, batch 6201 / 6666 (step 52862) loss=11006.3 (avg=3.872) [[all losses: diffusion=0.0903489 ; kl=1.10059e+10 ; lpips=0.243012 ; repa=0.554554 ; sum_loss=11006.3]]
[T_total=24:33:21 | T_train=24:05:46 | T_epoch=02:51:57] Epoch 7, batch 6301 / 6666 (step 52962) loss=10985.5 (avg=3.815) [[all losses: diffusion=0.090326 ; kl=1.09852e+10 ; lpips=0.242914 ; repa=0.55445 ; sum_loss=10985.5]]
[T_total=24:36:04 | T_train=24:08:30 | T_epoch=02:54:41] Epoch 7, batch 6401 / 6666 (step 53062) loss=10964.8 (avg=3.76) [[all losses: diffusion=0.0903063 ; kl=1.09645e+10 ; lpips=0.242813 ; repa=0.554346 ; sum_loss=10964.8]]
[T_total=24:38:48 | T_train=24:11:13 | T_epoch=02:57:25] Epoch 7, batch 6501 / 6666 (step 53162) loss=10944.2 (avg=3.726) [[all losses: diffusion=0.0903317 ; kl=1.09438e+10 ; lpips=0.242919 ; repa=0.554363 ; sum_loss=10944.2]]
[T_total=24:41:32 | T_train=24:13:57 | T_epoch=03:00:08] Epoch 7, batch 6601 / 6666 (step 53262) loss=10923.6 (avg=3.675) [[all losses: diffusion=0.0903285 ; kl=1.09233e+10 ; lpips=0.242885 ; repa=0.554299 ; sum_loss=10923.6]]
[2025-10-26 04:54:39,982][main][INFO] - [T_total=24:43:18 | T_train=24:15:44 | T_epoch=03:01:55] End of epoch 7 (53328 steps) train loss 3.64232[2025-10-26 04:54:39,984][main][INFO] - [Epoch 7] All losses: [[diffusion=0.0819999 ; kl=3.33403e+06 ; lpips=0.199347 ; repa=0.506464]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:01<06:35,  1.52s/it]
Reconstructing from test set:   1%|          | 2/261 [00:02<04:41,  1.09s/it]
Reconstructing from test set:   1%|          | 3/261 [00:03<04:04,  1.05it/s]
Reconstructing from test set:   2%|▏         | 4/261 [00:03<03:47,  1.13it/s]
Reconstructing from test set:   2%|▏         | 5/261 [00:04<03:37,  1.18it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:05<03:30,  1.21it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:06<03:26,  1.23it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:07<03:23,  1.24it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:07<03:21,  1.25it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:08<03:20,  1.25it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:09<03:18,  1.26it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:10<03:17,  1.26it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:10<03:16,  1.26it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:11<03:14,  1.27it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:12<03:13,  1.27it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:13<03:12,  1.27it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:14<03:11,  1.28it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:14<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:15<03:09,  1.28it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:16<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:17<03:07,  1.28it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:18<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:18<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:19<03:05,  1.28it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:20<03:04,  1.28it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:22<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:23<03:01,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:24<03:01,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:25<03:00,  1.27it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:25<02:59,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:26<02:58,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:27<02:58,  1.27it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:28<02:57,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:28<02:56,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:29<02:55,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:30<02:54,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:31<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:32<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:32<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:33<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:34<02:50,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:36<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:36<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:37<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:38<02:46,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:39<02:45,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:39<02:44,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:40<02:43,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:41<02:43,  1.28it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:42<02:42,  1.28it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:43<02:41,  1.28it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:43<02:40,  1.28it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:44<02:40,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:45<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:46<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:47<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:48<02:36,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:49<02:36,  1.27it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:50<02:35,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:50<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:51<02:33,  1.28it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:52<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:53<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:53<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:54<02:30,  1.27it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:55<02:30,  1.27it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:56<02:29,  1.27it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:57<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:57<02:27,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:58<02:26,  1.28it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [00:59<02:25,  1.27it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [01:00<02:24,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:01<02:23,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:01<02:23,  1.28it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:02<02:22,  1.28it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:03<02:21,  1.28it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:05<02:19,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:06<02:18,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:07<02:17,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:08<02:17,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:08<02:16,  1.28it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:09<02:15,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:10<02:14,  1.27it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:11<02:14,  1.27it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:12<02:13,  1.27it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:12<02:12,  1.27it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:13<02:11,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:14<02:10,  1.28it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:15<02:10,  1.27it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:15<02:09,  1.27it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:16<02:08,  1.27it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:17<02:08,  1.27it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:18<02:07,  1.27it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:19<02:06,  1.28it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:19<02:05,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:20<02:04,  1.27it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:21<02:03,  1.27it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:22<02:03,  1.27it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:22<02:02,  1.27it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:23<02:01,  1.27it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:24<02:00,  1.28it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:25<02:00,  1.27it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:26<01:59,  1.27it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:26<01:58,  1.27it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:27<01:58,  1.27it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:28<01:57,  1.27it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:29<01:56,  1.27it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:30<01:55,  1.27it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:30<01:54,  1.27it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:31<01:53,  1.27it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:32<01:53,  1.27it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:33<01:52,  1.27it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:34<01:52,  1.27it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:34<01:51,  1.27it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:35<01:50,  1.27it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:36<01:49,  1.27it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:37<01:48,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:37<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:38<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:39<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:41<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:41<01:43,  1.27it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:42<01:42,  1.28it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:43<01:41,  1.28it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:44<01:41,  1.28it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:44<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:45<01:39,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:46<01:38,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:47<01:37,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:48<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:48<01:36,  1.27it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:49<01:35,  1.28it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:50<01:34,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:51<01:33,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:52<01:33,  1.28it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:52<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:53<01:31,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:54<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:55<01:29,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:55<01:29,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:56<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:57<01:27,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:58<01:27,  1.27it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [01:59<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [01:59<01:25,  1.28it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:00<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:01<01:23,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:02<01:21,  1.28it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:03<01:21,  1.28it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:04<01:20,  1.28it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:05<01:19,  1.28it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:06<01:19,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:06<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:07<01:17,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:08<01:16,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:09<01:15,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:09<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:10<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:11<01:13,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:12<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:13<01:11,  1.28it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:13<01:11,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:14<01:10,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:15<01:09,  1.28it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:16<01:08,  1.29it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:17<01:07,  1.29it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:17<01:06,  1.29it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:18<01:06,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:19<01:05,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:20<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:20<01:04,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:21<01:03,  1.28it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:22<01:02,  1.28it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:23<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:24<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:24<01:00,  1.28it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:25<00:59,  1.28it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:26<00:58,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:27<00:58,  1.28it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:27<00:57,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:28<00:56,  1.27it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:29<00:55,  1.27it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:30<00:54,  1.27it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:31<00:54,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:31<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:32<00:52,  1.28it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:33<00:51,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:34<00:50,  1.27it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:35<00:50,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:35<00:49,  1.27it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:36<00:48,  1.27it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:37<00:47,  1.27it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:38<00:47,  1.27it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:38<00:46,  1.27it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:39<00:45,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:40<00:44,  1.27it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:41<00:44,  1.27it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:42<00:43,  1.27it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:42<00:42,  1.27it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:43<00:41,  1.27it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:44<00:40,  1.28it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:45<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:46<00:38,  1.28it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:46<00:38,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:47<00:37,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:49<00:36,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:49<00:35,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:50<00:34,  1.28it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:51<00:33,  1.27it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:53<00:32,  1.27it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:53<00:31,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:54<00:30,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:55<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:57<00:27,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:58<00:26,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [02:59<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [03:00<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:00<00:24,  1.28it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:01<00:23,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:02<00:22,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:03<00:21,  1.28it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:04<00:21,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:04<00:20,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:05<00:19,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:06<00:18,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:08<00:16,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:09<00:15,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:10<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:11<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:11<00:13,  1.28it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:12<00:12,  1.28it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:13<00:11,  1.28it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:14<00:10,  1.28it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:14<00:10,  1.28it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:15<00:09,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:16<00:08,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:17<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:18<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:18<00:06,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:19<00:05,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:20<00:04,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:21<00:03,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:21<00:03,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:22<00:02,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:23<00:01,  1.29it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:25<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:25<00:00,  1.27it/s]
[2025-10-26 04:58:08,198][main][INFO] - [Epoch 8] Test metrics: [[MSE=25.63 | MAE=0.1131 | LPIPS=0.1843 | PSNR=15.91 | SSIM=0.3824 | dreamsim=0.2779 | FID=28.82]][2025-10-26 04:58:08,200][main][INFO] - [Epoch 8] Best metrics: [[min_MSE=25.63 | min_MAE=0.1131 | min_LPIPS=0.1843 | max_PSNR=15.91 | max_SSIM=0.3824 | min_dreamsim=0.2779 | min_FID=28.82]][2025-10-26 04:58:08,201][main][DEBUG] - Writing images to disk...[2025-10-26 04:58:09,050][main][DEBUG] - Image(s) saved on disk[2025-10-26 04:58:09,290][main][INFO] - End of epoch timers: [T_train=24:15:44 | T_epoch=03:01:55 | T_eval=00:27:55 | T_total=24:46:48][2025-10-26 04:58:09,291][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-26 04:58:20,610][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-26 04:58:31,832][main][INFO] - ---

[2025-10-26 04:58:31,833][main][INFO] - [T_total=24:47:10 | T_train=24:15:44] Start epoch 8[T_total=24:47:13 | T_train=24:15:46 | T_epoch=00:00:02] Epoch 8, batch 1 / 6666 (step 53328) loss=10910.1 (avg=0.306) [[all losses: diffusion=0.0903156 ; kl=1.09098e+10 ; lpips=0.242818 ; repa=0.554232 ; sum_loss=10910.1]]
[T_total=24:49:56 | T_train=24:18:30 | T_epoch=00:02:46] Epoch 8, batch 101 / 6666 (step 53428) loss=10889.7 (avg=0.3012) [[all losses: diffusion=0.0902966 ; kl=1.08894e+10 ; lpips=0.242716 ; repa=0.554129 ; sum_loss=10889.7]]
[T_total=24:52:40 | T_train=24:21:14 | T_epoch=00:05:30] Epoch 8, batch 201 / 6666 (step 53528) loss=10869.4 (avg=0.3005) [[all losses: diffusion=0.0902756 ; kl=1.0869e+10 ; lpips=0.242615 ; repa=0.554024 ; sum_loss=10869.4]]
[T_total=24:55:24 | T_train=24:23:57 | T_epoch=00:08:13] Epoch 8, batch 301 / 6666 (step 53628) loss=10849.1 (avg=0.3003) [[all losses: diffusion=0.0902551 ; kl=1.08487e+10 ; lpips=0.242513 ; repa=0.55392 ; sum_loss=10849.1]]
[T_total=24:58:08 | T_train=24:26:41 | T_epoch=00:10:57] Epoch 8, batch 401 / 6666 (step 53728) loss=10828.9 (avg=0.3002) [[all losses: diffusion=0.0902336 ; kl=1.08286e+10 ; lpips=0.242415 ; repa=0.553816 ; sum_loss=10828.9]]
[T_total=25:00:51 | T_train=24:29:25 | T_epoch=00:13:41] Epoch 8, batch 501 / 6666 (step 53828) loss=10808.8 (avg=0.3003) [[all losses: diffusion=0.0902133 ; kl=1.08084e+10 ; lpips=0.242316 ; repa=0.553712 ; sum_loss=10808.8]]
[T_total=25:03:35 | T_train=24:32:08 | T_epoch=00:16:24] Epoch 8, batch 601 / 6666 (step 53928) loss=10788.7 (avg=0.3002) [[all losses: diffusion=0.0901931 ; kl=1.07884e+10 ; lpips=0.242216 ; repa=0.553609 ; sum_loss=10788.7]]
[T_total=25:06:19 | T_train=24:34:52 | T_epoch=00:19:08] Epoch 8, batch 701 / 6666 (step 54028) loss=10768.8 (avg=0.3002) [[all losses: diffusion=0.0901716 ; kl=1.07684e+10 ; lpips=0.242118 ; repa=0.553507 ; sum_loss=10768.8]]
[T_total=25:09:03 | T_train=24:37:36 | T_epoch=00:21:52] Epoch 8, batch 801 / 6666 (step 54128) loss=10748.9 (avg=0.3001) [[all losses: diffusion=0.0901506 ; kl=1.07485e+10 ; lpips=0.242021 ; repa=0.553403 ; sum_loss=10748.9]]
[T_total=25:11:46 | T_train=24:40:20 | T_epoch=00:24:36] Epoch 8, batch 901 / 6666 (step 54228) loss=10729.1 (avg=0.3002) [[all losses: diffusion=0.0901319 ; kl=1.07287e+10 ; lpips=0.241921 ; repa=0.5533 ; sum_loss=10729.1]]
[T_total=25:14:30 | T_train=24:43:04 | T_epoch=00:27:20] Epoch 8, batch 1001 / 6666 (step 54328) loss=10709.3 (avg=0.3) [[all losses: diffusion=0.0901103 ; kl=1.0709e+10 ; lpips=0.24182 ; repa=0.553198 ; sum_loss=10709.3]]
[T_total=25:17:14 | T_train=24:45:48 | T_epoch=00:30:04] Epoch 8, batch 1101 / 6666 (step 54428) loss=10689.6 (avg=0.3) [[all losses: diffusion=0.0900923 ; kl=1.06893e+10 ; lpips=0.241718 ; repa=0.553097 ; sum_loss=10689.6]]
[T_total=25:19:59 | T_train=24:48:32 | T_epoch=00:32:48] Epoch 8, batch 1201 / 6666 (step 54528) loss=10670 (avg=0.3) [[all losses: diffusion=0.0900733 ; kl=1.06697e+10 ; lpips=0.241619 ; repa=0.552994 ; sum_loss=10670]]
[T_total=25:22:42 | T_train=24:51:16 | T_epoch=00:35:32] Epoch 8, batch 1301 / 6666 (step 54628) loss=10650.5 (avg=0.6788) [[all losses: diffusion=0.090107 ; kl=1.06502e+10 ; lpips=0.241735 ; repa=0.553022 ; sum_loss=10650.5]]
[T_total=25:25:26 | T_train=24:54:00 | T_epoch=00:38:15] Epoch 8, batch 1401 / 6666 (step 54728) loss=10631.1 (avg=0.6529) [[all losses: diffusion=0.0900893 ; kl=1.06307e+10 ; lpips=0.241661 ; repa=0.552933 ; sum_loss=10631.1]]
[T_total=25:28:10 | T_train=24:56:43 | T_epoch=00:40:59] Epoch 8, batch 1501 / 6666 (step 54828) loss=10612.4 (avg=25.98) [[all losses: diffusion=0.0901021 ; kl=1.0612e+10 ; lpips=0.241674 ; repa=0.552898 ; sum_loss=10612.4]]
[T_total=25:30:54 | T_train=24:59:27 | T_epoch=00:43:43] Epoch 8, batch 1601 / 6666 (step 54928) loss=10593 (avg=24.46) [[all losses: diffusion=0.0901003 ; kl=1.05927e+10 ; lpips=0.241657 ; repa=0.552846 ; sum_loss=10593]]
[T_total=25:33:37 | T_train=25:02:11 | T_epoch=00:46:27] Epoch 8, batch 1701 / 6666 (step 55028) loss=10573.8 (avg=23.04) [[all losses: diffusion=0.0900791 ; kl=1.05734e+10 ; lpips=0.241564 ; repa=0.552748 ; sum_loss=10573.8]]
[T_total=25:36:21 | T_train=25:04:55 | T_epoch=00:49:10] Epoch 8, batch 1801 / 6666 (step 55128) loss=10554.6 (avg=22.54) [[all losses: diffusion=0.0900797 ; kl=1.05543e+10 ; lpips=0.241536 ; repa=0.552691 ; sum_loss=10554.6]]
[T_total=25:39:05 | T_train=25:07:38 | T_epoch=00:51:54] Epoch 8, batch 1901 / 6666 (step 55228) loss=10535.5 (avg=21.37) [[all losses: diffusion=0.0900626 ; kl=1.05352e+10 ; lpips=0.241439 ; repa=0.552591 ; sum_loss=10535.5]]
[T_total=25:41:48 | T_train=25:10:22 | T_epoch=00:54:38] Epoch 8, batch 2001 / 6666 (step 55328) loss=10516.5 (avg=20.32) [[all losses: diffusion=0.0900431 ; kl=1.05161e+10 ; lpips=0.241347 ; repa=0.552491 ; sum_loss=10516.5]]
[T_total=25:44:32 | T_train=25:13:06 | T_epoch=00:57:22] Epoch 8, batch 2101 / 6666 (step 55428) loss=10497.5 (avg=19.37) [[all losses: diffusion=0.0900234 ; kl=1.04972e+10 ; lpips=0.241251 ; repa=0.55239 ; sum_loss=10497.5]]
[T_total=25:47:16 | T_train=25:15:50 | T_epoch=01:00:05] Epoch 8, batch 2201 / 6666 (step 55528) loss=10478.6 (avg=18.5) [[all losses: diffusion=0.0900012 ; kl=1.04783e+10 ; lpips=0.241158 ; repa=0.552292 ; sum_loss=10478.6]]
[T_total=25:50:00 | T_train=25:18:33 | T_epoch=01:02:49] Epoch 8, batch 2301 / 6666 (step 55628) loss=10459.8 (avg=17.71) [[all losses: diffusion=0.0899814 ; kl=1.04594e+10 ; lpips=0.241064 ; repa=0.552191 ; sum_loss=10459.8]]
[T_total=25:52:44 | T_train=25:21:17 | T_epoch=01:05:33] Epoch 8, batch 2401 / 6666 (step 55728) loss=10441 (avg=16.98) [[all losses: diffusion=0.0899596 ; kl=1.04407e+10 ; lpips=0.240975 ; repa=0.552095 ; sum_loss=10441]]
[T_total=25:55:27 | T_train=25:24:01 | T_epoch=01:08:17] Epoch 8, batch 2501 / 6666 (step 55828) loss=10422.3 (avg=16.32) [[all losses: diffusion=0.0899399 ; kl=1.0422e+10 ; lpips=0.240881 ; repa=0.551995 ; sum_loss=10422.3]]
[T_total=25:58:11 | T_train=25:26:44 | T_epoch=01:11:00] Epoch 8, batch 2601 / 6666 (step 55928) loss=10403.7 (avg=15.7) [[all losses: diffusion=0.0899208 ; kl=1.04033e+10 ; lpips=0.240785 ; repa=0.551894 ; sum_loss=10403.7]]
[T_total=26:00:55 | T_train=25:29:28 | T_epoch=01:13:44] Epoch 8, batch 2701 / 6666 (step 56028) loss=10385.1 (avg=15.13) [[all losses: diffusion=0.0899036 ; kl=1.03848e+10 ; lpips=0.240689 ; repa=0.551794 ; sum_loss=10385.1]]
[T_total=26:03:39 | T_train=25:32:12 | T_epoch=01:16:28] Epoch 8, batch 2801 / 6666 (step 56128) loss=10366.6 (avg=14.6) [[all losses: diffusion=0.0898832 ; kl=1.03663e+10 ; lpips=0.240594 ; repa=0.551697 ; sum_loss=10366.6]]
[T_total=26:06:18 | T_train=25:34:52 | T_epoch=01:19:08] Epoch 8, batch 2901 / 6666 (step 56228) loss=10348.2 (avg=14.11) [[all losses: diffusion=0.0898652 ; kl=1.03478e+10 ; lpips=0.240502 ; repa=0.551599 ; sum_loss=10348.2]]
[T_total=26:09:02 | T_train=25:37:35 | T_epoch=01:21:51] Epoch 8, batch 3001 / 6666 (step 56328) loss=10329.8 (avg=13.65) [[all losses: diffusion=0.0898459 ; kl=1.03294e+10 ; lpips=0.240413 ; repa=0.551502 ; sum_loss=10329.8]]
[T_total=26:11:46 | T_train=25:40:19 | T_epoch=01:24:35] Epoch 8, batch 3101 / 6666 (step 56428) loss=10311.5 (avg=13.22) [[all losses: diffusion=0.0898266 ; kl=1.03111e+10 ; lpips=0.240317 ; repa=0.551404 ; sum_loss=10311.5]]
[T_total=26:14:29 | T_train=25:43:03 | T_epoch=01:27:19] Epoch 8, batch 3201 / 6666 (step 56528) loss=10293.2 (avg=12.81) [[all losses: diffusion=0.0898061 ; kl=1.02929e+10 ; lpips=0.240224 ; repa=0.551306 ; sum_loss=10293.2]]
[T_total=26:17:13 | T_train=25:45:47 | T_epoch=01:30:02] Epoch 8, batch 3301 / 6666 (step 56628) loss=10275.1 (avg=12.45) [[all losses: diffusion=0.0898168 ; kl=1.02747e+10 ; lpips=0.24027 ; repa=0.551296 ; sum_loss=10275.1]]
[T_total=26:19:57 | T_train=25:48:30 | T_epoch=01:32:46] Epoch 8, batch 3401 / 6666 (step 56728) loss=10257 (avg=12.09) [[all losses: diffusion=0.0898078 ; kl=1.02566e+10 ; lpips=0.240222 ; repa=0.551222 ; sum_loss=10257]]
[T_total=26:22:40 | T_train=25:51:14 | T_epoch=01:35:30] Epoch 8, batch 3501 / 6666 (step 56828) loss=10238.9 (avg=11.76) [[all losses: diffusion=0.0897899 ; kl=1.02386e+10 ; lpips=0.240134 ; repa=0.551128 ; sum_loss=10238.9]]
[T_total=26:25:24 | T_train=25:53:58 | T_epoch=01:38:14] Epoch 8, batch 3601 / 6666 (step 56928) loss=10220.9 (avg=11.44) [[all losses: diffusion=0.0897727 ; kl=1.02206e+10 ; lpips=0.240038 ; repa=0.551031 ; sum_loss=10220.9]]
[T_total=26:28:08 | T_train=25:56:42 | T_epoch=01:40:57] Epoch 8, batch 3701 / 6666 (step 57028) loss=10203 (avg=11.14) [[all losses: diffusion=0.0897542 ; kl=1.02027e+10 ; lpips=0.239944 ; repa=0.550935 ; sum_loss=10203]]
[T_total=26:30:52 | T_train=25:59:25 | T_epoch=01:43:41] Epoch 8, batch 3801 / 6666 (step 57128) loss=10185.1 (avg=10.85) [[all losses: diffusion=0.0897367 ; kl=1.01848e+10 ; lpips=0.239852 ; repa=0.550838 ; sum_loss=10185.1]]
[T_total=26:33:36 | T_train=26:02:09 | T_epoch=01:46:25] Epoch 8, batch 3901 / 6666 (step 57228) loss=10167.4 (avg=10.58) [[all losses: diffusion=0.0897155 ; kl=1.0167e+10 ; lpips=0.239761 ; repa=0.550741 ; sum_loss=10167.4]]
[T_total=26:36:20 | T_train=26:04:53 | T_epoch=01:49:09] Epoch 8, batch 4001 / 6666 (step 57328) loss=10149.6 (avg=10.33) [[all losses: diffusion=0.0896984 ; kl=1.01493e+10 ; lpips=0.239667 ; repa=0.550646 ; sum_loss=10149.6]]
[T_total=26:39:03 | T_train=26:07:37 | T_epoch=01:51:53] Epoch 8, batch 4101 / 6666 (step 57428) loss=10131.9 (avg=10.08) [[all losses: diffusion=0.08968 ; kl=1.01316e+10 ; lpips=0.239583 ; repa=0.550554 ; sum_loss=10131.9]]
[T_total=26:41:47 | T_train=26:10:20 | T_epoch=01:54:36] Epoch 8, batch 4201 / 6666 (step 57528) loss=10114.3 (avg=9.849) [[all losses: diffusion=0.0896614 ; kl=1.0114e+10 ; lpips=0.239491 ; repa=0.550458 ; sum_loss=10114.3]]
[T_total=26:44:31 | T_train=26:13:04 | T_epoch=01:57:20] Epoch 8, batch 4301 / 6666 (step 57628) loss=10096.8 (avg=9.627) [[all losses: diffusion=0.089644 ; kl=1.00964e+10 ; lpips=0.239401 ; repa=0.550363 ; sum_loss=10096.8]]
[T_total=26:47:14 | T_train=26:15:48 | T_epoch=02:00:04] Epoch 8, batch 4401 / 6666 (step 57728) loss=10079.3 (avg=9.415) [[all losses: diffusion=0.0896256 ; kl=1.00789e+10 ; lpips=0.239312 ; repa=0.55027 ; sum_loss=10079.3]]
[T_total=26:49:58 | T_train=26:18:32 | T_epoch=02:02:48] Epoch 8, batch 4501 / 6666 (step 57828) loss=10061.9 (avg=9.212) [[all losses: diffusion=0.0896063 ; kl=1.00615e+10 ; lpips=0.239223 ; repa=0.550175 ; sum_loss=10061.9]]
[T_total=26:52:42 | T_train=26:21:15 | T_epoch=02:05:31] Epoch 8, batch 4601 / 6666 (step 57928) loss=10044.5 (avg=9.019) [[all losses: diffusion=0.0895873 ; kl=1.00441e+10 ; lpips=0.239135 ; repa=0.550081 ; sum_loss=10044.5]]
[T_total=26:55:26 | T_train=26:23:59 | T_epoch=02:08:15] Epoch 8, batch 4701 / 6666 (step 58028) loss=10027.2 (avg=8.833) [[all losses: diffusion=0.0895671 ; kl=1.00268e+10 ; lpips=0.239045 ; repa=0.549988 ; sum_loss=10027.2]]
[T_total=26:58:09 | T_train=26:26:43 | T_epoch=02:10:59] Epoch 8, batch 4801 / 6666 (step 58128) loss=10009.9 (avg=8.655) [[all losses: diffusion=0.0895493 ; kl=1.00096e+10 ; lpips=0.238955 ; repa=0.549895 ; sum_loss=10009.9]]
[T_total=27:00:53 | T_train=26:29:26 | T_epoch=02:13:42] Epoch 8, batch 4901 / 6666 (step 58228) loss=9992.75 (avg=8.485) [[all losses: diffusion=0.0895299 ; kl=9.9924e+09 ; lpips=0.238864 ; repa=0.549801 ; sum_loss=9992.75]]
[T_total=27:03:37 | T_train=26:32:10 | T_epoch=02:16:26] Epoch 8, batch 5001 / 6666 (step 58328) loss=9975.62 (avg=8.321) [[all losses: diffusion=0.0895117 ; kl=9.97527e+09 ; lpips=0.238771 ; repa=0.549706 ; sum_loss=9975.62]]
[T_total=27:06:20 | T_train=26:34:54 | T_epoch=02:19:10] Epoch 8, batch 5101 / 6666 (step 58428) loss=9958.54 (avg=8.164) [[all losses: diffusion=0.0894944 ; kl=9.9582e+09 ; lpips=0.238679 ; repa=0.549612 ; sum_loss=9958.54]]
[T_total=27:09:04 | T_train=26:37:37 | T_epoch=02:21:53] Epoch 8, batch 5201 / 6666 (step 58528) loss=9941.53 (avg=8.013) [[all losses: diffusion=0.0894763 ; kl=9.94118e+09 ; lpips=0.238594 ; repa=0.54952 ; sum_loss=9941.53]]
[T_total=27:11:48 | T_train=26:40:21 | T_epoch=02:24:37] Epoch 8, batch 5301 / 6666 (step 58628) loss=9924.57 (avg=7.867) [[all losses: diffusion=0.089459 ; kl=9.92423e+09 ; lpips=0.23851 ; repa=0.549429 ; sum_loss=9924.57]]
[T_total=27:14:31 | T_train=26:43:05 | T_epoch=02:27:21] Epoch 8, batch 5401 / 6666 (step 58728) loss=9907.67 (avg=7.727) [[all losses: diffusion=0.0894394 ; kl=9.90733e+09 ; lpips=0.238422 ; repa=0.549335 ; sum_loss=9907.67]]
[T_total=27:17:15 | T_train=26:45:48 | T_epoch=02:30:04] Epoch 8, batch 5501 / 6666 (step 58828) loss=9890.83 (avg=7.592) [[all losses: diffusion=0.0894204 ; kl=9.89049e+09 ; lpips=0.238339 ; repa=0.549244 ; sum_loss=9890.83]]
[T_total=27:19:59 | T_train=26:48:32 | T_epoch=02:32:48] Epoch 8, batch 5601 / 6666 (step 58928) loss=9874.05 (avg=7.462) [[all losses: diffusion=0.0894011 ; kl=9.8737e+09 ; lpips=0.238253 ; repa=0.549151 ; sum_loss=9874.05]]
[T_total=27:22:42 | T_train=26:51:16 | T_epoch=02:35:32] Epoch 8, batch 5701 / 6666 (step 59028) loss=9857.32 (avg=7.336) [[all losses: diffusion=0.0893825 ; kl=9.85698e+09 ; lpips=0.238169 ; repa=0.549057 ; sum_loss=9857.32]]
[T_total=27:25:26 | T_train=26:53:59 | T_epoch=02:38:15] Epoch 8, batch 5801 / 6666 (step 59128) loss=9840.65 (avg=7.215) [[all losses: diffusion=0.0893628 ; kl=9.84031e+09 ; lpips=0.238084 ; repa=0.548965 ; sum_loss=9840.65]]
[T_total=27:28:10 | T_train=26:56:43 | T_epoch=02:40:59] Epoch 8, batch 5901 / 6666 (step 59228) loss=9824.04 (avg=7.098) [[all losses: diffusion=0.0893455 ; kl=9.82369e+09 ; lpips=0.237995 ; repa=0.548874 ; sum_loss=9824.04]]
[T_total=27:30:53 | T_train=26:59:27 | T_epoch=02:43:43] Epoch 8, batch 6001 / 6666 (step 59328) loss=9807.48 (avg=6.984) [[all losses: diffusion=0.0893266 ; kl=9.80713e+09 ; lpips=0.237908 ; repa=0.548782 ; sum_loss=9807.48]]
[T_total=27:33:37 | T_train=27:02:10 | T_epoch=02:46:26] Epoch 8, batch 6101 / 6666 (step 59428) loss=9790.98 (avg=6.875) [[all losses: diffusion=0.0893118 ; kl=9.79063e+09 ; lpips=0.23782 ; repa=0.548688 ; sum_loss=9790.98]]
[T_total=27:36:21 | T_train=27:04:54 | T_epoch=02:49:10] Epoch 8, batch 6201 / 6666 (step 59528) loss=9774.53 (avg=6.769) [[all losses: diffusion=0.0892956 ; kl=9.77419e+09 ; lpips=0.237732 ; repa=0.548595 ; sum_loss=9774.53]]
[T_total=27:39:04 | T_train=27:07:38 | T_epoch=02:51:54] Epoch 8, batch 6301 / 6666 (step 59628) loss=9758.14 (avg=6.666) [[all losses: diffusion=0.089279 ; kl=9.75779e+09 ; lpips=0.237645 ; repa=0.548502 ; sum_loss=9758.14]]
[T_total=27:41:48 | T_train=27:10:22 | T_epoch=02:54:37] Epoch 8, batch 6401 / 6666 (step 59728) loss=9741.8 (avg=6.566) [[all losses: diffusion=0.0892618 ; kl=9.74146e+09 ; lpips=0.23756 ; repa=0.54841 ; sum_loss=9741.8]]
[T_total=27:44:32 | T_train=27:13:05 | T_epoch=02:57:21] Epoch 8, batch 6501 / 6666 (step 59828) loss=9725.52 (avg=6.47) [[all losses: diffusion=0.0892465 ; kl=9.72517e+09 ; lpips=0.23747 ; repa=0.548319 ; sum_loss=9725.52]]
[T_total=27:47:16 | T_train=27:15:49 | T_epoch=03:00:05] Epoch 8, batch 6601 / 6666 (step 59928) loss=9709.29 (avg=6.376) [[all losses: diffusion=0.0892291 ; kl=9.70895e+09 ; lpips=0.237383 ; repa=0.548227 ; sum_loss=9709.29]]
[2025-10-26 08:00:23,920][main][INFO] - [T_total=27:49:02 | T_train=27:17:36 | T_epoch=03:01:52] End of epoch 8 (59994 steps) train loss 6.31716[2025-10-26 08:00:23,922][main][INFO] - [Epoch 8] All losses: [[diffusion=0.0804194 ; kl=6.01513e+06 ; lpips=0.19339 ; repa=0.49965]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:01<06:25,  1.48s/it]
Reconstructing from test set:   1%|          | 2/261 [00:02<04:37,  1.07s/it]
Reconstructing from test set:   1%|          | 3/261 [00:03<04:03,  1.06it/s]
Reconstructing from test set:   2%|▏         | 4/261 [00:03<03:46,  1.14it/s]
Reconstructing from test set:   2%|▏         | 5/261 [00:04<03:36,  1.18it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:05<03:29,  1.21it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:06<03:25,  1.24it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:06<03:21,  1.25it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:07<03:20,  1.26it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:08<03:18,  1.27it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:09<03:16,  1.27it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:10<03:15,  1.27it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:10<03:14,  1.28it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:11<03:13,  1.28it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:12<03:12,  1.28it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:13<03:11,  1.28it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:13<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:14<03:09,  1.28it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:15<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:16<03:07,  1.29it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:17<03:06,  1.29it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:17<03:05,  1.29it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:18<03:05,  1.29it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:19<03:04,  1.29it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:20<03:03,  1.29it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:20<03:03,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:21<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:22<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:23<03:01,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:24<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:24<02:59,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:25<02:59,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:26<02:58,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:27<02:57,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:28<02:56,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:28<02:54,  1.29it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:29<02:54,  1.29it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:30<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:31<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:31<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:32<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:33<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:34<02:50,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:35<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:36<02:47,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:37<02:46,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:38<02:46,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:38<02:45,  1.28it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:39<02:45,  1.27it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:40<02:44,  1.27it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:41<02:43,  1.28it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:42<02:42,  1.28it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:42<02:41,  1.29it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:43<02:40,  1.28it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:44<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:45<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:45<02:38,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:46<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:47<02:36,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:48<02:36,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:49<02:35,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:49<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:50<02:33,  1.29it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:51<02:32,  1.29it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:52<02:31,  1.29it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:52<02:30,  1.29it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:53<02:30,  1.28it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:54<02:29,  1.29it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:55<02:28,  1.28it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:56<02:27,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:56<02:26,  1.29it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:57<02:26,  1.29it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:58<02:25,  1.29it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [00:59<02:24,  1.29it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [00:59<02:23,  1.29it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:00<02:22,  1.29it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:01<02:22,  1.28it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:02<02:22,  1.28it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:03<02:21,  1.28it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:03<02:21,  1.28it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:04<02:20,  1.27it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:05<02:20,  1.27it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:06<02:19,  1.27it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:07<02:18,  1.27it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:07<02:17,  1.27it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:08<02:16,  1.27it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:09<02:15,  1.27it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:10<02:14,  1.27it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:10<02:14,  1.27it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:11<02:13,  1.27it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:12<02:12,  1.27it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:13<02:11,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:14<02:11,  1.27it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:14<02:10,  1.27it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:15<02:09,  1.27it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:16<02:08,  1.27it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:17<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:18<02:06,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:18<02:05,  1.28it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:19<02:05,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:20<02:04,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:21<02:03,  1.28it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:21<02:02,  1.28it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:22<02:01,  1.28it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:23<02:01,  1.28it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:24<02:00,  1.28it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:25<01:59,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:25<01:58,  1.28it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:26<01:57,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:27<01:57,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:28<01:56,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:28<01:55,  1.28it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:29<01:54,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:30<01:54,  1.28it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:31<01:53,  1.28it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:32<01:53,  1.27it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:32<01:52,  1.27it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:33<01:51,  1.27it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:34<01:50,  1.27it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:35<01:49,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:36<01:48,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:36<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:37<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:38<01:45,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:39<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:39<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:40<01:43,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:41<01:43,  1.28it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:42<01:42,  1.28it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:43<01:41,  1.28it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:43<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:44<01:40,  1.28it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:45<01:39,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:46<01:38,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:46<01:37,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:47<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:48<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:49<01:35,  1.28it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:50<01:34,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:50<01:33,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:51<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:52<01:31,  1.28it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:53<01:30,  1.29it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:53<01:30,  1.29it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:54<01:29,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:55<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:56<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:57<01:27,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:57<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [01:58<01:25,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [01:59<01:25,  1.28it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:00<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:00<01:23,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:01<01:22,  1.28it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:03<01:21,  1.28it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:04<01:20,  1.28it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:04<01:19,  1.28it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:05<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:06<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:07<01:17,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:08<01:16,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:08<01:15,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:09<01:15,  1.28it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:10<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:11<01:13,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:11<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:12<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:13<01:11,  1.27it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:14<01:10,  1.27it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:15<01:10,  1.27it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:15<01:09,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:16<01:08,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:17<01:07,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:18<01:06,  1.27it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:19<01:05,  1.27it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:19<01:05,  1.27it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:20<01:04,  1.27it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:21<01:03,  1.27it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:22<01:02,  1.27it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:22<01:02,  1.27it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:23<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:24<01:00,  1.28it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:25<00:59,  1.27it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:26<00:58,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:26<00:58,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:27<00:57,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:28<00:56,  1.27it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:29<00:55,  1.27it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:29<00:54,  1.28it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:30<00:54,  1.27it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:31<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:32<00:52,  1.28it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:33<00:51,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:33<00:50,  1.27it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:34<00:50,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:35<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:36<00:48,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:37<00:47,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:37<00:46,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:38<00:46,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:39<00:45,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:40<00:44,  1.28it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:40<00:43,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:41<00:43,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:42<00:42,  1.28it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:43<00:41,  1.27it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:44<00:40,  1.27it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:44<00:40,  1.27it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:45<00:39,  1.27it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:46<00:38,  1.27it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:47<00:37,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:49<00:35,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:50<00:34,  1.28it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:51<00:33,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:51<00:32,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:53<00:31,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:54<00:30,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:55<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:55<00:28,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:57<00:27,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:58<00:26,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [02:58<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [02:59<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:00<00:24,  1.28it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:01<00:23,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:02<00:22,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:02<00:21,  1.28it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:03<00:20,  1.29it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:04<00:20,  1.29it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:05<00:19,  1.29it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:05<00:18,  1.29it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:06<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:08<00:16,  1.29it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:09<00:15,  1.29it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:09<00:14,  1.29it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:10<00:14,  1.29it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:11<00:13,  1.29it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:12<00:12,  1.29it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:12<00:11,  1.29it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:13<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:14<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:15<00:09,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:16<00:08,  1.29it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:16<00:07,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:17<00:06,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:18<00:06,  1.29it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:19<00:05,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:19<00:04,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:20<00:03,  1.29it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:21<00:03,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:22<00:02,  1.29it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:23<00:01,  1.29it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:23<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.29it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.28it/s]
[2025-10-26 08:03:51,684][main][INFO] - [Epoch 9] Test metrics: [[MSE=25.07 | MAE=0.1114 | LPIPS=0.1784 | PSNR=16.01 | SSIM=0.3893 | dreamsim=0.2661 | FID=26.23]][2025-10-26 08:03:51,686][main][INFO] - [Epoch 9] Best metrics: [[min_MSE=25.07 | min_MAE=0.1114 | min_LPIPS=0.1784 | max_PSNR=16.01 | max_SSIM=0.3893 | min_dreamsim=0.2661 | min_FID=26.23]][2025-10-26 08:03:51,687][main][DEBUG] - Writing images to disk...[2025-10-26 08:03:52,532][main][DEBUG] - Image(s) saved on disk[2025-10-26 08:03:52,734][main][INFO] - End of epoch timers: [T_train=27:17:36 | T_epoch=03:01:52 | T_eval=00:31:23 | T_total=27:52:31][2025-10-26 08:03:52,735][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-26 08:04:04,116][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-26 08:04:15,898][main][INFO] - ---

[2025-10-26 08:04:15,899][main][INFO] - [T_total=27:52:54 | T_train=27:17:36] Start epoch 9[T_total=27:52:57 | T_train=27:17:38 | T_epoch=00:00:02] Epoch 9, batch 1 / 6666 (step 59994) loss=9698.61 (avg=0.297) [[all losses: diffusion=0.0892157 ; kl=9.69827e+09 ; lpips=0.237326 ; repa=0.548167 ; sum_loss=9698.61]]
[T_total=27:55:41 | T_train=27:20:22 | T_epoch=00:02:46] Epoch 9, batch 101 / 6666 (step 60094) loss=9682.47 (avg=0.2955) [[all losses: diffusion=0.0891986 ; kl=9.68213e+09 ; lpips=0.237236 ; repa=0.548075 ; sum_loss=9682.47]]
[T_total=27:58:24 | T_train=27:23:06 | T_epoch=00:05:30] Epoch 9, batch 201 / 6666 (step 60194) loss=9666.39 (avg=0.2959) [[all losses: diffusion=0.0891796 ; kl=9.66604e+09 ; lpips=0.237151 ; repa=0.547984 ; sum_loss=9666.39]]
[T_total=28:01:08 | T_train=27:25:49 | T_epoch=00:08:13] Epoch 9, batch 301 / 6666 (step 60294) loss=9650.36 (avg=0.2962) [[all losses: diffusion=0.0891638 ; kl=9.65001e+09 ; lpips=0.237063 ; repa=0.547894 ; sum_loss=9650.36]]
[T_total=28:03:51 | T_train=27:28:33 | T_epoch=00:10:57] Epoch 9, batch 401 / 6666 (step 60394) loss=9634.38 (avg=0.2962) [[all losses: diffusion=0.0891458 ; kl=9.63403e+09 ; lpips=0.236977 ; repa=0.547802 ; sum_loss=9634.38]]
[T_total=28:06:35 | T_train=27:31:17 | T_epoch=00:13:40] Epoch 9, batch 501 / 6666 (step 60494) loss=9618.45 (avg=0.2962) [[all losses: diffusion=0.0891283 ; kl=9.61811e+09 ; lpips=0.236891 ; repa=0.547711 ; sum_loss=9618.45]]
[T_total=28:09:19 | T_train=27:34:00 | T_epoch=00:16:24] Epoch 9, batch 601 / 6666 (step 60594) loss=9638.91 (avg=3663) [[all losses: diffusion=0.0891829 ; kl=9.63856e+09 ; lpips=0.237081 ; repa=0.547788 ; sum_loss=9638.91]]
[T_total=28:12:02 | T_train=27:36:44 | T_epoch=00:19:08] Epoch 9, batch 701 / 6666 (step 60694) loss=9623.03 (avg=3141) [[all losses: diffusion=0.0891718 ; kl=9.62268e+09 ; lpips=0.237028 ; repa=0.547716 ; sum_loss=9623.03]]
[T_total=28:14:46 | T_train=27:39:27 | T_epoch=00:21:51] Epoch 9, batch 801 / 6666 (step 60794) loss=9607.47 (avg=2769) [[all losses: diffusion=0.0892377 ; kl=9.60713e+09 ; lpips=0.237236 ; repa=0.547808 ; sum_loss=9607.47]]
[T_total=28:17:30 | T_train=27:42:11 | T_epoch=00:24:35] Epoch 9, batch 901 / 6666 (step 60894) loss=9591.69 (avg=2462) [[all losses: diffusion=0.0892254 ; kl=9.59135e+09 ; lpips=0.237171 ; repa=0.54773 ; sum_loss=9591.69]]
[T_total=28:20:14 | T_train=27:44:55 | T_epoch=00:27:19] Epoch 9, batch 1001 / 6666 (step 60994) loss=9575.97 (avg=2216) [[all losses: diffusion=0.089209 ; kl=9.57563e+09 ; lpips=0.237087 ; repa=0.547641 ; sum_loss=9575.97]]
[T_total=28:22:57 | T_train=27:47:39 | T_epoch=00:30:03] Epoch 9, batch 1101 / 6666 (step 61094) loss=9560.3 (avg=2015) [[all losses: diffusion=0.0891926 ; kl=9.55995e+09 ; lpips=0.237001 ; repa=0.547552 ; sum_loss=9560.3]]
[T_total=28:25:41 | T_train=27:50:23 | T_epoch=00:32:46] Epoch 9, batch 1201 / 6666 (step 61194) loss=9544.67 (avg=1847) [[all losses: diffusion=0.0891741 ; kl=9.54433e+09 ; lpips=0.236921 ; repa=0.547464 ; sum_loss=9544.67]]
[T_total=28:28:25 | T_train=27:53:06 | T_epoch=00:35:30] Epoch 9, batch 1301 / 6666 (step 61294) loss=9529.1 (avg=1705) [[all losses: diffusion=0.0891561 ; kl=9.52876e+09 ; lpips=0.236837 ; repa=0.547375 ; sum_loss=9529.1]]
[T_total=28:31:09 | T_train=27:55:50 | T_epoch=00:38:14] Epoch 9, batch 1401 / 6666 (step 61394) loss=9513.58 (avg=1583) [[all losses: diffusion=0.0891372 ; kl=9.51324e+09 ; lpips=0.236754 ; repa=0.547285 ; sum_loss=9513.58]]
[T_total=28:33:52 | T_train=27:58:34 | T_epoch=00:40:58] Epoch 9, batch 1501 / 6666 (step 61494) loss=9498.11 (avg=1478) [[all losses: diffusion=0.0891198 ; kl=9.49777e+09 ; lpips=0.236672 ; repa=0.547197 ; sum_loss=9498.11]]
[T_total=28:36:36 | T_train=28:01:17 | T_epoch=00:43:41] Epoch 9, batch 1601 / 6666 (step 61594) loss=9482.69 (avg=1386) [[all losses: diffusion=0.0891027 ; kl=9.48235e+09 ; lpips=0.236588 ; repa=0.547108 ; sum_loss=9482.69]]
[T_total=28:39:20 | T_train=28:04:01 | T_epoch=00:46:25] Epoch 9, batch 1701 / 6666 (step 61694) loss=9467.32 (avg=1304) [[all losses: diffusion=0.0890848 ; kl=9.46698e+09 ; lpips=0.236505 ; repa=0.547021 ; sum_loss=9467.32]]
[T_total=28:42:04 | T_train=28:06:45 | T_epoch=00:49:09] Epoch 9, batch 1801 / 6666 (step 61794) loss=9452 (avg=1232) [[all losses: diffusion=0.0890671 ; kl=9.45166e+09 ; lpips=0.236422 ; repa=0.546931 ; sum_loss=9452]]
[T_total=28:44:47 | T_train=28:09:29 | T_epoch=00:51:53] Epoch 9, batch 1901 / 6666 (step 61894) loss=9436.73 (avg=1167) [[all losses: diffusion=0.0890504 ; kl=9.43639e+09 ; lpips=0.236337 ; repa=0.546841 ; sum_loss=9436.73]]
[T_total=28:47:31 | T_train=28:12:13 | T_epoch=00:54:36] Epoch 9, batch 2001 / 6666 (step 61994) loss=9421.57 (avg=1110) [[all losses: diffusion=0.0890628 ; kl=9.42122e+09 ; lpips=0.236363 ; repa=0.546822 ; sum_loss=9421.57]]
[T_total=28:50:15 | T_train=28:14:56 | T_epoch=00:57:20] Epoch 9, batch 2101 / 6666 (step 62094) loss=9406.4 (avg=1058) [[all losses: diffusion=0.0890462 ; kl=9.40605e+09 ; lpips=0.23628 ; repa=0.546734 ; sum_loss=9406.4]]
[T_total=28:52:58 | T_train=28:17:40 | T_epoch=01:00:04] Epoch 9, batch 2201 / 6666 (step 62194) loss=9391.27 (avg=1010) [[all losses: diffusion=0.0890286 ; kl=9.39093e+09 ; lpips=0.236199 ; repa=0.546647 ; sum_loss=9391.27]]
[T_total=28:55:42 | T_train=28:20:23 | T_epoch=01:02:47] Epoch 9, batch 2301 / 6666 (step 62294) loss=9376.2 (avg=965.7) [[all losses: diffusion=0.0890121 ; kl=9.37585e+09 ; lpips=0.236116 ; repa=0.546558 ; sum_loss=9376.2]]
[T_total=28:58:26 | T_train=28:23:07 | T_epoch=01:05:31] Epoch 9, batch 2401 / 6666 (step 62394) loss=9361.17 (avg=925.5) [[all losses: diffusion=0.0889956 ; kl=9.36083e+09 ; lpips=0.236034 ; repa=0.54647 ; sum_loss=9361.17]]
[T_total=29:01:09 | T_train=28:25:51 | T_epoch=01:08:15] Epoch 9, batch 2501 / 6666 (step 62494) loss=9346.29 (avg=891) [[all losses: diffusion=0.088999 ; kl=9.34595e+09 ; lpips=0.236007 ; repa=0.546419 ; sum_loss=9346.29]]
[T_total=29:03:53 | T_train=28:28:35 | T_epoch=01:10:59] Epoch 9, batch 2601 / 6666 (step 62594) loss=9331.36 (avg=856.7) [[all losses: diffusion=0.0889851 ; kl=9.33102e+09 ; lpips=0.235942 ; repa=0.546342 ; sum_loss=9331.36]]
[T_total=29:06:37 | T_train=28:31:19 | T_epoch=01:13:42] Epoch 9, batch 2701 / 6666 (step 62694) loss=9316.48 (avg=825) [[all losses: diffusion=0.0889685 ; kl=9.31613e+09 ; lpips=0.235861 ; repa=0.546255 ; sum_loss=9316.48]]
[T_total=29:09:21 | T_train=28:34:03 | T_epoch=01:16:26] Epoch 9, batch 2801 / 6666 (step 62794) loss=9301.64 (avg=795.6) [[all losses: diffusion=0.0889511 ; kl=9.3013e+09 ; lpips=0.23578 ; repa=0.546169 ; sum_loss=9301.64]]
[T_total=29:12:05 | T_train=28:36:46 | T_epoch=01:19:10] Epoch 9, batch 2901 / 6666 (step 62894) loss=9286.85 (avg=768.2) [[all losses: diffusion=0.0889352 ; kl=9.28651e+09 ; lpips=0.235696 ; repa=0.546082 ; sum_loss=9286.85]]
[T_total=29:14:48 | T_train=28:39:30 | T_epoch=01:21:54] Epoch 9, batch 3001 / 6666 (step 62994) loss=9272.11 (avg=742.6) [[all losses: diffusion=0.0889209 ; kl=9.27177e+09 ; lpips=0.235629 ; repa=0.546002 ; sum_loss=9272.11]]
[T_total=29:17:32 | T_train=28:42:14 | T_epoch=01:24:37] Epoch 9, batch 3101 / 6666 (step 63094) loss=9257.42 (avg=718.6) [[all losses: diffusion=0.0889044 ; kl=9.25707e+09 ; lpips=0.235547 ; repa=0.545915 ; sum_loss=9257.42]]
[T_total=29:20:16 | T_train=28:44:57 | T_epoch=01:27:21] Epoch 9, batch 3201 / 6666 (step 63194) loss=9242.77 (avg=696.2) [[all losses: diffusion=0.0888901 ; kl=9.24242e+09 ; lpips=0.235478 ; repa=0.545835 ; sum_loss=9242.77]]
[T_total=29:22:59 | T_train=28:47:41 | T_epoch=01:30:05] Epoch 9, batch 3301 / 6666 (step 63294) loss=9228.17 (avg=675.1) [[all losses: diffusion=0.0888918 ; kl=9.22782e+09 ; lpips=0.235463 ; repa=0.54579 ; sum_loss=9228.17]]
[T_total=29:25:43 | T_train=28:50:25 | T_epoch=01:32:48] Epoch 9, batch 3401 / 6666 (step 63394) loss=9213.61 (avg=655.3) [[all losses: diffusion=0.088904 ; kl=9.21327e+09 ; lpips=0.235482 ; repa=0.545766 ; sum_loss=9213.61]]
[T_total=29:28:27 | T_train=28:53:09 | T_epoch=01:35:32] Epoch 9, batch 3501 / 6666 (step 63494) loss=9199.1 (avg=636.6) [[all losses: diffusion=0.0888928 ; kl=9.19876e+09 ; lpips=0.235414 ; repa=0.545687 ; sum_loss=9199.1]]
[T_total=29:31:11 | T_train=28:55:52 | T_epoch=01:38:16] Epoch 9, batch 3601 / 6666 (step 63594) loss=9184.64 (avg=618.9) [[all losses: diffusion=0.088876 ; kl=9.18429e+09 ; lpips=0.23534 ; repa=0.545604 ; sum_loss=9184.64]]
[T_total=29:33:55 | T_train=28:58:36 | T_epoch=01:41:00] Epoch 9, batch 3701 / 6666 (step 63694) loss=9170.22 (avg=602.2) [[all losses: diffusion=0.0888597 ; kl=9.16987e+09 ; lpips=0.23526 ; repa=0.54552 ; sum_loss=9170.22]]
[T_total=29:36:38 | T_train=29:01:20 | T_epoch=01:43:44] Epoch 9, batch 3801 / 6666 (step 63794) loss=9155.84 (avg=586.4) [[all losses: diffusion=0.0888423 ; kl=9.1555e+09 ; lpips=0.235182 ; repa=0.545435 ; sum_loss=9155.84]]
[T_total=29:39:22 | T_train=29:04:03 | T_epoch=01:46:27] Epoch 9, batch 3901 / 6666 (step 63894) loss=9141.51 (avg=571.4) [[all losses: diffusion=0.0888258 ; kl=9.14117e+09 ; lpips=0.235103 ; repa=0.54535 ; sum_loss=9141.51]]
[T_total=29:42:06 | T_train=29:06:47 | T_epoch=01:49:11] Epoch 9, batch 4001 / 6666 (step 63994) loss=9127.23 (avg=557.1) [[all losses: diffusion=0.0888093 ; kl=9.12689e+09 ; lpips=0.235023 ; repa=0.545265 ; sum_loss=9127.23]]
[T_total=29:44:49 | T_train=29:09:31 | T_epoch=01:51:55] Epoch 9, batch 4101 / 6666 (step 64094) loss=9112.99 (avg=543.5) [[all losses: diffusion=0.0887943 ; kl=9.11265e+09 ; lpips=0.234944 ; repa=0.54518 ; sum_loss=9112.99]]
[T_total=29:47:33 | T_train=29:12:15 | T_epoch=01:54:38] Epoch 9, batch 4201 / 6666 (step 64194) loss=9098.79 (avg=530.6) [[all losses: diffusion=0.0887774 ; kl=9.09845e+09 ; lpips=0.234865 ; repa=0.545095 ; sum_loss=9098.79]]
[T_total=29:50:17 | T_train=29:14:58 | T_epoch=01:57:22] Epoch 9, batch 4301 / 6666 (step 64294) loss=9084.64 (avg=518.2) [[all losses: diffusion=0.0887627 ; kl=9.0843e+09 ; lpips=0.234785 ; repa=0.545011 ; sum_loss=9084.64]]
[T_total=29:53:01 | T_train=29:17:42 | T_epoch=02:00:06] Epoch 9, batch 4401 / 6666 (step 64394) loss=9070.54 (avg=506.5) [[all losses: diffusion=0.0887485 ; kl=9.07019e+09 ; lpips=0.234705 ; repa=0.544928 ; sum_loss=9070.54]]
[T_total=29:55:44 | T_train=29:20:26 | T_epoch=02:02:50] Epoch 9, batch 4501 / 6666 (step 64494) loss=9056.47 (avg=495.2) [[all losses: diffusion=0.0887315 ; kl=9.05613e+09 ; lpips=0.234627 ; repa=0.544844 ; sum_loss=9056.47]]
[T_total=29:58:28 | T_train=29:23:10 | T_epoch=02:05:33] Epoch 9, batch 4601 / 6666 (step 64594) loss=9042.45 (avg=484.5) [[all losses: diffusion=0.0887155 ; kl=9.04211e+09 ; lpips=0.234545 ; repa=0.544759 ; sum_loss=9042.45]]
[T_total=30:01:12 | T_train=29:25:53 | T_epoch=02:08:17] Epoch 9, batch 4701 / 6666 (step 64694) loss=9028.48 (avg=474.2) [[all losses: diffusion=0.0887 ; kl=9.02813e+09 ; lpips=0.234466 ; repa=0.544674 ; sum_loss=9028.48]]
[T_total=30:03:56 | T_train=29:28:37 | T_epoch=02:11:01] Epoch 9, batch 4801 / 6666 (step 64794) loss=9014.54 (avg=464.3) [[all losses: diffusion=0.0886835 ; kl=9.0142e+09 ; lpips=0.234388 ; repa=0.544591 ; sum_loss=9014.54]]
[T_total=30:06:40 | T_train=29:31:21 | T_epoch=02:13:45] Epoch 9, batch 4901 / 6666 (step 64894) loss=9000.65 (avg=454.8) [[all losses: diffusion=0.0886685 ; kl=9.00031e+09 ; lpips=0.234308 ; repa=0.544507 ; sum_loss=9000.65]]
[T_total=30:09:23 | T_train=29:34:05 | T_epoch=02:16:29] Epoch 9, batch 5001 / 6666 (step 64994) loss=8986.8 (avg=445.7) [[all losses: diffusion=0.088653 ; kl=8.98646e+09 ; lpips=0.234229 ; repa=0.544422 ; sum_loss=8986.8]]
[T_total=30:12:07 | T_train=29:36:48 | T_epoch=02:19:12] Epoch 9, batch 5101 / 6666 (step 65094) loss=8973 (avg=437) [[all losses: diffusion=0.0886364 ; kl=8.97266e+09 ; lpips=0.23415 ; repa=0.544339 ; sum_loss=8973]]
[T_total=30:14:51 | T_train=29:39:32 | T_epoch=02:21:56] Epoch 9, batch 5201 / 6666 (step 65194) loss=8959.24 (avg=428.6) [[all losses: diffusion=0.0886211 ; kl=8.95889e+09 ; lpips=0.234071 ; repa=0.544256 ; sum_loss=8959.24]]
[T_total=30:17:34 | T_train=29:42:16 | T_epoch=02:24:40] Epoch 9, batch 5301 / 6666 (step 65294) loss=8945.52 (avg=420.5) [[all losses: diffusion=0.088605 ; kl=8.94517e+09 ; lpips=0.233995 ; repa=0.544173 ; sum_loss=8945.52]]
[T_total=30:20:18 | T_train=29:44:59 | T_epoch=02:27:23] Epoch 9, batch 5401 / 6666 (step 65394) loss=8931.84 (avg=412.8) [[all losses: diffusion=0.0885876 ; kl=8.9315e+09 ; lpips=0.23392 ; repa=0.544089 ; sum_loss=8931.84]]
[T_total=30:23:02 | T_train=29:47:43 | T_epoch=02:30:07] Epoch 9, batch 5501 / 6666 (step 65494) loss=8918.2 (avg=405.3) [[all losses: diffusion=0.0885724 ; kl=8.91786e+09 ; lpips=0.233849 ; repa=0.544009 ; sum_loss=8918.2]]
[T_total=30:25:46 | T_train=29:50:27 | T_epoch=02:32:51] Epoch 9, batch 5601 / 6666 (step 65594) loss=8904.6 (avg=398) [[all losses: diffusion=0.0885586 ; kl=8.90426e+09 ; lpips=0.233773 ; repa=0.543928 ; sum_loss=8904.6]]
[T_total=30:28:29 | T_train=29:53:11 | T_epoch=02:35:35] Epoch 9, batch 5701 / 6666 (step 65694) loss=8891.08 (avg=391.4) [[all losses: diffusion=0.0885724 ; kl=8.89074e+09 ; lpips=0.2338 ; repa=0.543911 ; sum_loss=8891.08]]
[T_total=30:31:13 | T_train=29:55:54 | T_epoch=02:38:18] Epoch 9, batch 5801 / 6666 (step 65794) loss=8877.56 (avg=384.6) [[all losses: diffusion=0.0885622 ; kl=8.87722e+09 ; lpips=0.233738 ; repa=0.543837 ; sum_loss=8877.56]]
[T_total=30:33:57 | T_train=29:58:38 | T_epoch=02:41:02] Epoch 9, batch 5901 / 6666 (step 65894) loss=8864.09 (avg=378.1) [[all losses: diffusion=0.0885798 ; kl=8.86375e+09 ; lpips=0.233779 ; repa=0.543829 ; sum_loss=8864.09]]
[T_total=30:36:40 | T_train=30:01:22 | T_epoch=02:43:46] Epoch 9, batch 6001 / 6666 (step 65994) loss=8850.66 (avg=371.8) [[all losses: diffusion=0.0885656 ; kl=8.85032e+09 ; lpips=0.233704 ; repa=0.543748 ; sum_loss=8850.66]]
[T_total=30:39:24 | T_train=30:04:05 | T_epoch=02:46:29] Epoch 9, batch 6101 / 6666 (step 66094) loss=8837.28 (avg=365.8) [[all losses: diffusion=0.0885726 ; kl=8.83694e+09 ; lpips=0.23371 ; repa=0.543716 ; sum_loss=8837.28]]
[T_total=30:42:08 | T_train=30:06:49 | T_epoch=02:49:13] Epoch 9, batch 6201 / 6666 (step 66194) loss=8823.93 (avg=359.9) [[all losses: diffusion=0.0885585 ; kl=8.82359e+09 ; lpips=0.233634 ; repa=0.543635 ; sum_loss=8823.93]]
[T_total=30:44:52 | T_train=30:09:33 | T_epoch=02:51:57] Epoch 9, batch 6301 / 6666 (step 66294) loss=8810.62 (avg=354.2) [[all losses: diffusion=0.0885445 ; kl=8.81028e+09 ; lpips=0.233559 ; repa=0.543554 ; sum_loss=8810.62]]
[T_total=30:47:36 | T_train=30:12:17 | T_epoch=02:54:41] Epoch 9, batch 6401 / 6666 (step 66394) loss=8797.35 (avg=348.7) [[all losses: diffusion=0.0885302 ; kl=8.79701e+09 ; lpips=0.233491 ; repa=0.543476 ; sum_loss=8797.35]]
[T_total=30:50:19 | T_train=30:15:00 | T_epoch=02:57:24] Epoch 9, batch 6501 / 6666 (step 66494) loss=8784.13 (avg=343.4) [[all losses: diffusion=0.0885695 ; kl=8.78379e+09 ; lpips=0.233609 ; repa=0.543517 ; sum_loss=8784.13]]
[T_total=30:53:03 | T_train=30:17:44 | T_epoch=03:00:08] Epoch 9, batch 6601 / 6666 (step 66594) loss=8770.94 (avg=338.2) [[all losses: diffusion=0.0885536 ; kl=8.7706e+09 ; lpips=0.233536 ; repa=0.543436 ; sum_loss=8770.94]]
[2025-10-26 11:06:10,742][main][INFO] - [T_total=30:54:49 | T_train=30:19:31 | T_epoch=03:01:54] End of epoch 9 (66660 steps) train loss 334.899[2025-10-26 11:06:10,744][main][INFO] - [Epoch 9] All losses: [[diffusion=0.0824992 ; kl=3.34592e+08 ; lpips=0.198936 ; repa=0.500318]]Reconstructing from test set:   0%|          | 0/261 [00:00<?, ?it/s]
Reconstructing from test set:   0%|          | 1/261 [00:01<06:27,  1.49s/it]
Reconstructing from test set:   1%|          | 2/261 [00:02<04:38,  1.07s/it]
Reconstructing from test set:   1%|          | 3/261 [00:03<04:03,  1.06it/s]
Reconstructing from test set:   2%|▏         | 4/261 [00:03<03:45,  1.14it/s]
Reconstructing from test set:   2%|▏         | 5/261 [00:04<03:36,  1.18it/s]
Reconstructing from test set:   2%|▏         | 6/261 [00:05<03:30,  1.21it/s]
Reconstructing from test set:   3%|▎         | 7/261 [00:06<03:26,  1.23it/s]
Reconstructing from test set:   3%|▎         | 8/261 [00:06<03:22,  1.25it/s]
Reconstructing from test set:   3%|▎         | 9/261 [00:07<03:20,  1.26it/s]
Reconstructing from test set:   4%|▍         | 10/261 [00:08<03:18,  1.26it/s]
Reconstructing from test set:   4%|▍         | 11/261 [00:09<03:17,  1.27it/s]
Reconstructing from test set:   5%|▍         | 12/261 [00:10<03:16,  1.27it/s]
Reconstructing from test set:   5%|▍         | 13/261 [00:10<03:14,  1.27it/s]
Reconstructing from test set:   5%|▌         | 14/261 [00:11<03:13,  1.27it/s]
Reconstructing from test set:   6%|▌         | 15/261 [00:12<03:13,  1.27it/s]
Reconstructing from test set:   6%|▌         | 16/261 [00:13<03:12,  1.27it/s]
Reconstructing from test set:   7%|▋         | 17/261 [00:14<03:11,  1.27it/s]
Reconstructing from test set:   7%|▋         | 18/261 [00:14<03:10,  1.28it/s]
Reconstructing from test set:   7%|▋         | 19/261 [00:15<03:09,  1.28it/s]
Reconstructing from test set:   8%|▊         | 20/261 [00:16<03:08,  1.28it/s]
Reconstructing from test set:   8%|▊         | 21/261 [00:17<03:07,  1.28it/s]
Reconstructing from test set:   8%|▊         | 22/261 [00:17<03:07,  1.28it/s]
Reconstructing from test set:   9%|▉         | 23/261 [00:18<03:06,  1.28it/s]
Reconstructing from test set:   9%|▉         | 24/261 [00:19<03:05,  1.28it/s]
Reconstructing from test set:  10%|▉         | 25/261 [00:20<03:04,  1.28it/s]
Reconstructing from test set:  10%|▉         | 26/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  10%|█         | 27/261 [00:21<03:03,  1.28it/s]
Reconstructing from test set:  11%|█         | 28/261 [00:22<03:02,  1.28it/s]
Reconstructing from test set:  11%|█         | 29/261 [00:23<03:01,  1.28it/s]
Reconstructing from test set:  11%|█▏        | 30/261 [00:24<03:00,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 31/261 [00:24<02:59,  1.28it/s]
Reconstructing from test set:  12%|█▏        | 32/261 [00:25<02:58,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 33/261 [00:26<02:57,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 34/261 [00:27<02:57,  1.28it/s]
Reconstructing from test set:  13%|█▎        | 35/261 [00:28<02:56,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 36/261 [00:28<02:55,  1.28it/s]
Reconstructing from test set:  14%|█▍        | 37/261 [00:29<02:54,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 38/261 [00:30<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▍        | 39/261 [00:31<02:53,  1.28it/s]
Reconstructing from test set:  15%|█▌        | 40/261 [00:31<02:52,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 41/261 [00:32<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▌        | 42/261 [00:33<02:51,  1.28it/s]
Reconstructing from test set:  16%|█▋        | 43/261 [00:34<02:50,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 44/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  17%|█▋        | 45/261 [00:35<02:49,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 46/261 [00:36<02:48,  1.28it/s]
Reconstructing from test set:  18%|█▊        | 47/261 [00:37<02:47,  1.27it/s]
Reconstructing from test set:  18%|█▊        | 48/261 [00:38<02:47,  1.27it/s]
Reconstructing from test set:  19%|█▉        | 49/261 [00:39<02:46,  1.27it/s]
Reconstructing from test set:  19%|█▉        | 50/261 [00:39<02:45,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 51/261 [00:40<02:44,  1.28it/s]
Reconstructing from test set:  20%|█▉        | 52/261 [00:41<02:43,  1.28it/s]
Reconstructing from test set:  20%|██        | 53/261 [00:42<02:42,  1.28it/s]
Reconstructing from test set:  21%|██        | 54/261 [00:42<02:42,  1.28it/s]
Reconstructing from test set:  21%|██        | 55/261 [00:43<02:41,  1.28it/s]
Reconstructing from test set:  21%|██▏       | 56/261 [00:44<02:40,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 57/261 [00:45<02:39,  1.28it/s]
Reconstructing from test set:  22%|██▏       | 58/261 [00:46<02:39,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 59/261 [00:46<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 60/261 [00:47<02:37,  1.28it/s]
Reconstructing from test set:  23%|██▎       | 61/261 [00:48<02:36,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 62/261 [00:49<02:35,  1.28it/s]
Reconstructing from test set:  24%|██▍       | 63/261 [00:50<02:35,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 64/261 [00:50<02:34,  1.28it/s]
Reconstructing from test set:  25%|██▍       | 65/261 [00:51<02:33,  1.28it/s]
Reconstructing from test set:  25%|██▌       | 66/261 [00:52<02:32,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 67/261 [00:53<02:31,  1.28it/s]
Reconstructing from test set:  26%|██▌       | 68/261 [00:53<02:30,  1.28it/s]
Reconstructing from test set:  26%|██▋       | 69/261 [00:54<02:31,  1.27it/s]
Reconstructing from test set:  27%|██▋       | 70/261 [00:55<02:30,  1.27it/s]
Reconstructing from test set:  27%|██▋       | 71/261 [00:56<02:29,  1.27it/s]
Reconstructing from test set:  28%|██▊       | 72/261 [00:57<02:28,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 73/261 [00:57<02:27,  1.28it/s]
Reconstructing from test set:  28%|██▊       | 74/261 [00:58<02:26,  1.28it/s]
Reconstructing from test set:  29%|██▊       | 75/261 [00:59<02:25,  1.28it/s]
Reconstructing from test set:  29%|██▉       | 76/261 [01:00<02:24,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 77/261 [01:00<02:23,  1.28it/s]
Reconstructing from test set:  30%|██▉       | 78/261 [01:01<02:22,  1.28it/s]
Reconstructing from test set:  30%|███       | 79/261 [01:02<02:22,  1.28it/s]
Reconstructing from test set:  31%|███       | 80/261 [01:03<02:21,  1.28it/s]
Reconstructing from test set:  31%|███       | 81/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  31%|███▏      | 82/261 [01:04<02:20,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 83/261 [01:05<02:19,  1.28it/s]
Reconstructing from test set:  32%|███▏      | 84/261 [01:06<02:18,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 85/261 [01:07<02:18,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 86/261 [01:08<02:17,  1.28it/s]
Reconstructing from test set:  33%|███▎      | 87/261 [01:08<02:16,  1.28it/s]
Reconstructing from test set:  34%|███▎      | 88/261 [01:09<02:15,  1.28it/s]
Reconstructing from test set:  34%|███▍      | 89/261 [01:10<02:15,  1.27it/s]
Reconstructing from test set:  34%|███▍      | 90/261 [01:11<02:14,  1.27it/s]
Reconstructing from test set:  35%|███▍      | 91/261 [01:11<02:13,  1.28it/s]
Reconstructing from test set:  35%|███▌      | 92/261 [01:12<02:12,  1.27it/s]
Reconstructing from test set:  36%|███▌      | 93/261 [01:13<02:11,  1.28it/s]
Reconstructing from test set:  36%|███▌      | 94/261 [01:14<02:10,  1.28it/s]
Reconstructing from test set:  36%|███▋      | 95/261 [01:15<02:10,  1.28it/s]
Reconstructing from test set:  37%|███▋      | 96/261 [01:15<02:09,  1.27it/s]
Reconstructing from test set:  37%|███▋      | 97/261 [01:16<02:08,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 98/261 [01:17<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 99/261 [01:18<02:07,  1.28it/s]
Reconstructing from test set:  38%|███▊      | 100/261 [01:18<02:06,  1.28it/s]
Reconstructing from test set:  39%|███▊      | 101/261 [01:19<02:05,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 102/261 [01:20<02:04,  1.28it/s]
Reconstructing from test set:  39%|███▉      | 103/261 [01:21<02:03,  1.28it/s]
Reconstructing from test set:  40%|███▉      | 104/261 [01:22<02:02,  1.28it/s]
Reconstructing from test set:  40%|████      | 105/261 [01:22<02:02,  1.28it/s]
Reconstructing from test set:  41%|████      | 106/261 [01:23<02:01,  1.28it/s]
Reconstructing from test set:  41%|████      | 107/261 [01:24<02:00,  1.28it/s]
Reconstructing from test set:  41%|████▏     | 108/261 [01:25<02:00,  1.27it/s]
Reconstructing from test set:  42%|████▏     | 109/261 [01:26<01:59,  1.27it/s]
Reconstructing from test set:  42%|████▏     | 110/261 [01:26<01:58,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 111/261 [01:27<01:57,  1.28it/s]
Reconstructing from test set:  43%|████▎     | 112/261 [01:28<01:56,  1.27it/s]
Reconstructing from test set:  43%|████▎     | 113/261 [01:29<01:56,  1.27it/s]
Reconstructing from test set:  44%|████▎     | 114/261 [01:29<01:55,  1.27it/s]
Reconstructing from test set:  44%|████▍     | 115/261 [01:30<01:54,  1.27it/s]
Reconstructing from test set:  44%|████▍     | 116/261 [01:31<01:53,  1.27it/s]
Reconstructing from test set:  45%|████▍     | 117/261 [01:32<01:53,  1.27it/s]
Reconstructing from test set:  45%|████▌     | 118/261 [01:33<01:52,  1.27it/s]
Reconstructing from test set:  46%|████▌     | 119/261 [01:33<01:51,  1.27it/s]
Reconstructing from test set:  46%|████▌     | 120/261 [01:34<01:50,  1.28it/s]
Reconstructing from test set:  46%|████▋     | 121/261 [01:35<01:49,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 122/261 [01:36<01:48,  1.28it/s]
Reconstructing from test set:  47%|████▋     | 123/261 [01:37<01:48,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 124/261 [01:37<01:47,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 125/261 [01:38<01:46,  1.28it/s]
Reconstructing from test set:  48%|████▊     | 126/261 [01:39<01:45,  1.28it/s]
Reconstructing from test set:  49%|████▊     | 127/261 [01:40<01:44,  1.28it/s]
Reconstructing from test set:  49%|████▉     | 128/261 [01:40<01:44,  1.27it/s]
Reconstructing from test set:  49%|████▉     | 129/261 [01:41<01:43,  1.27it/s]
Reconstructing from test set:  50%|████▉     | 130/261 [01:42<01:43,  1.27it/s]
Reconstructing from test set:  50%|█████     | 131/261 [01:43<01:42,  1.27it/s]
Reconstructing from test set:  51%|█████     | 132/261 [01:44<01:41,  1.27it/s]
Reconstructing from test set:  51%|█████     | 133/261 [01:44<01:40,  1.27it/s]
Reconstructing from test set:  51%|█████▏    | 134/261 [01:45<01:39,  1.27it/s]
Reconstructing from test set:  52%|█████▏    | 135/261 [01:46<01:38,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 136/261 [01:47<01:37,  1.28it/s]
Reconstructing from test set:  52%|█████▏    | 137/261 [01:48<01:37,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 138/261 [01:48<01:36,  1.28it/s]
Reconstructing from test set:  53%|█████▎    | 139/261 [01:49<01:35,  1.28it/s]
Reconstructing from test set:  54%|█████▎    | 140/261 [01:50<01:34,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 141/261 [01:51<01:33,  1.28it/s]
Reconstructing from test set:  54%|█████▍    | 142/261 [01:51<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▍    | 143/261 [01:52<01:32,  1.28it/s]
Reconstructing from test set:  55%|█████▌    | 144/261 [01:53<01:31,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 145/261 [01:54<01:30,  1.28it/s]
Reconstructing from test set:  56%|█████▌    | 146/261 [01:55<01:29,  1.28it/s]
Reconstructing from test set:  56%|█████▋    | 147/261 [01:55<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 148/261 [01:56<01:28,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 149/261 [01:57<01:27,  1.28it/s]
Reconstructing from test set:  57%|█████▋    | 150/261 [01:58<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 151/261 [01:58<01:26,  1.28it/s]
Reconstructing from test set:  58%|█████▊    | 152/261 [01:59<01:25,  1.27it/s]
Reconstructing from test set:  59%|█████▊    | 153/261 [02:00<01:24,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 154/261 [02:01<01:23,  1.28it/s]
Reconstructing from test set:  59%|█████▉    | 155/261 [02:02<01:23,  1.28it/s]
Reconstructing from test set:  60%|█████▉    | 156/261 [02:02<01:22,  1.28it/s]
Reconstructing from test set:  60%|██████    | 157/261 [02:03<01:21,  1.28it/s]
Reconstructing from test set:  61%|██████    | 158/261 [02:04<01:20,  1.28it/s]
Reconstructing from test set:  61%|██████    | 159/261 [02:05<01:19,  1.28it/s]
Reconstructing from test set:  61%|██████▏   | 160/261 [02:05<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 161/261 [02:06<01:18,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 162/261 [02:07<01:17,  1.28it/s]
Reconstructing from test set:  62%|██████▏   | 163/261 [02:08<01:16,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 164/261 [02:09<01:15,  1.28it/s]
Reconstructing from test set:  63%|██████▎   | 165/261 [02:09<01:15,  1.28it/s]
Reconstructing from test set:  64%|██████▎   | 166/261 [02:10<01:14,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 167/261 [02:11<01:13,  1.28it/s]
Reconstructing from test set:  64%|██████▍   | 168/261 [02:12<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▍   | 169/261 [02:13<01:12,  1.28it/s]
Reconstructing from test set:  65%|██████▌   | 170/261 [02:13<01:11,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 171/261 [02:14<01:10,  1.28it/s]
Reconstructing from test set:  66%|██████▌   | 172/261 [02:15<01:09,  1.28it/s]
Reconstructing from test set:  66%|██████▋   | 173/261 [02:16<01:09,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 174/261 [02:16<01:08,  1.27it/s]
Reconstructing from test set:  67%|██████▋   | 175/261 [02:17<01:07,  1.28it/s]
Reconstructing from test set:  67%|██████▋   | 176/261 [02:18<01:06,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 177/261 [02:19<01:05,  1.28it/s]
Reconstructing from test set:  68%|██████▊   | 178/261 [02:20<01:05,  1.28it/s]
Reconstructing from test set:  69%|██████▊   | 179/261 [02:20<01:04,  1.27it/s]
Reconstructing from test set:  69%|██████▉   | 180/261 [02:21<01:03,  1.27it/s]
Reconstructing from test set:  69%|██████▉   | 181/261 [02:22<01:02,  1.27it/s]
Reconstructing from test set:  70%|██████▉   | 182/261 [02:23<01:02,  1.27it/s]
Reconstructing from test set:  70%|███████   | 183/261 [02:23<01:01,  1.28it/s]
Reconstructing from test set:  70%|███████   | 184/261 [02:24<01:00,  1.28it/s]
Reconstructing from test set:  71%|███████   | 185/261 [02:25<00:59,  1.27it/s]
Reconstructing from test set:  71%|███████▏  | 186/261 [02:26<00:58,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 187/261 [02:27<00:58,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 188/261 [02:27<00:57,  1.27it/s]
Reconstructing from test set:  72%|███████▏  | 189/261 [02:28<00:56,  1.27it/s]
Reconstructing from test set:  73%|███████▎  | 190/261 [02:29<00:55,  1.28it/s]
Reconstructing from test set:  73%|███████▎  | 191/261 [02:30<00:54,  1.28it/s]
Reconstructing from test set:  74%|███████▎  | 192/261 [02:31<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 193/261 [02:31<00:53,  1.28it/s]
Reconstructing from test set:  74%|███████▍  | 194/261 [02:32<00:52,  1.28it/s]
Reconstructing from test set:  75%|███████▍  | 195/261 [02:33<00:51,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 196/261 [02:34<00:50,  1.28it/s]
Reconstructing from test set:  75%|███████▌  | 197/261 [02:34<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 198/261 [02:35<00:49,  1.28it/s]
Reconstructing from test set:  76%|███████▌  | 199/261 [02:36<00:48,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 200/261 [02:37<00:47,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 201/261 [02:38<00:46,  1.28it/s]
Reconstructing from test set:  77%|███████▋  | 202/261 [02:38<00:46,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 203/261 [02:39<00:45,  1.28it/s]
Reconstructing from test set:  78%|███████▊  | 204/261 [02:40<00:44,  1.28it/s]
Reconstructing from test set:  79%|███████▊  | 205/261 [02:41<00:43,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 206/261 [02:41<00:42,  1.28it/s]
Reconstructing from test set:  79%|███████▉  | 207/261 [02:42<00:42,  1.28it/s]
Reconstructing from test set:  80%|███████▉  | 208/261 [02:43<00:41,  1.28it/s]
Reconstructing from test set:  80%|████████  | 209/261 [02:44<00:40,  1.28it/s]
Reconstructing from test set:  80%|████████  | 210/261 [02:45<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 211/261 [02:45<00:39,  1.28it/s]
Reconstructing from test set:  81%|████████  | 212/261 [02:46<00:38,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 213/261 [02:47<00:37,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 214/261 [02:48<00:36,  1.28it/s]
Reconstructing from test set:  82%|████████▏ | 215/261 [02:49<00:35,  1.28it/s]
Reconstructing from test set:  83%|████████▎ | 216/261 [02:49<00:35,  1.29it/s]
Reconstructing from test set:  83%|████████▎ | 217/261 [02:50<00:34,  1.28it/s]
Reconstructing from test set:  84%|████████▎ | 218/261 [02:51<00:33,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 219/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  84%|████████▍ | 220/261 [02:52<00:32,  1.28it/s]
Reconstructing from test set:  85%|████████▍ | 221/261 [02:53<00:31,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 222/261 [02:54<00:30,  1.28it/s]
Reconstructing from test set:  85%|████████▌ | 223/261 [02:55<00:29,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 224/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  86%|████████▌ | 225/261 [02:56<00:28,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 226/261 [02:57<00:27,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 227/261 [02:58<00:26,  1.28it/s]
Reconstructing from test set:  87%|████████▋ | 228/261 [02:59<00:25,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 229/261 [02:59<00:24,  1.28it/s]
Reconstructing from test set:  88%|████████▊ | 230/261 [03:00<00:24,  1.28it/s]
Reconstructing from test set:  89%|████████▊ | 231/261 [03:01<00:23,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 232/261 [03:02<00:22,  1.28it/s]
Reconstructing from test set:  89%|████████▉ | 233/261 [03:03<00:21,  1.28it/s]
Reconstructing from test set:  90%|████████▉ | 234/261 [03:03<00:21,  1.27it/s]
Reconstructing from test set:  90%|█████████ | 235/261 [03:04<00:20,  1.28it/s]
Reconstructing from test set:  90%|█████████ | 236/261 [03:05<00:19,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 237/261 [03:06<00:18,  1.28it/s]
Reconstructing from test set:  91%|█████████ | 238/261 [03:06<00:18,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 239/261 [03:07<00:17,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 240/261 [03:08<00:16,  1.28it/s]
Reconstructing from test set:  92%|█████████▏| 241/261 [03:09<00:15,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 242/261 [03:10<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 243/261 [03:10<00:14,  1.28it/s]
Reconstructing from test set:  93%|█████████▎| 244/261 [03:11<00:13,  1.28it/s]
Reconstructing from test set:  94%|█████████▍| 245/261 [03:12<00:12,  1.28it/s]
Reconstructing from test set:  94%|█████████▍| 246/261 [03:13<00:11,  1.28it/s]
Reconstructing from test set:  95%|█████████▍| 247/261 [03:14<00:10,  1.28it/s]
Reconstructing from test set:  95%|█████████▌| 248/261 [03:14<00:10,  1.29it/s]
Reconstructing from test set:  95%|█████████▌| 249/261 [03:15<00:09,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 250/261 [03:16<00:08,  1.28it/s]
Reconstructing from test set:  96%|█████████▌| 251/261 [03:17<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 252/261 [03:17<00:07,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 253/261 [03:18<00:06,  1.28it/s]
Reconstructing from test set:  97%|█████████▋| 254/261 [03:19<00:05,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 255/261 [03:20<00:04,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 256/261 [03:21<00:03,  1.28it/s]
Reconstructing from test set:  98%|█████████▊| 257/261 [03:21<00:03,  1.28it/s]
Reconstructing from test set:  99%|█████████▉| 258/261 [03:22<00:02,  1.28it/s]
Reconstructing from test set:  99%|█████████▉| 259/261 [03:23<00:01,  1.28it/s]
Reconstructing from test set: 100%|█████████▉| 260/261 [03:24<00:00,  1.28it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.28it/s]
Reconstructing from test set: 100%|██████████| 261/261 [03:24<00:00,  1.27it/s]
[2025-10-26 11:09:38,812][main][INFO] - [Epoch 10] Test metrics: [[MSE=24.11 | MAE=0.1086 | LPIPS=0.1724 | PSNR=16.18 | SSIM=0.3956 | dreamsim=0.255 | FID=23.81]][2025-10-26 11:09:38,814][main][INFO] - [Epoch 10] Best metrics: [[min_MSE=24.11 | min_MAE=0.1086 | min_LPIPS=0.1724 | max_PSNR=16.18 | max_SSIM=0.3956 | min_dreamsim=0.255 | min_FID=23.81]][2025-10-26 11:09:38,815][main][DEBUG] - Writing images to disk...[2025-10-26 11:09:39,906][main][DEBUG] - Image(s) saved on disk[2025-10-26 11:09:40,115][main][INFO] - End of epoch timers: [T_train=30:19:31 | T_epoch=03:01:54 | T_eval=00:34:53 | T_total=30:58:18][2025-10-26 11:09:40,117][main][INFO] - Storing model checkpoint inside /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/last[2025-10-26 11:09:51,824][main][INFO] - Best FID so far, storing a copy of the model checkpoint to /workspace/DC_SSDAE/runs/jobs/train_enc_dc_f32c32_FM/checkpoints/best[2025-10-26 11:10:02,555][main][INFO] - ---

[2025-10-26 11:10:02,556][main][INFO] - [T_total=30:58:41 | T_train=30:19:31] Start epoch 10[T_total=30:58:43 | T_train=30:19:33 | T_epoch=00:00:02] Epoch 10, batch 1 / 6666 (step 66660) loss=8762.25 (avg=0.2989) [[all losses: diffusion=0.0885445 ; kl=8.76191e+09 ; lpips=0.233487 ; repa=0.543382 ; sum_loss=8762.25]]
W1026 11:11:34.207000 58142 site-packages/torch/distributed/elastic/agent/server/api.py:725] Received 15 death signal, shutting down workers
W1026 11:11:34.216000 58142 site-packages/torch/distributed/elastic/multiprocessing/api.py:908] Sending process 58214 closing signal SIGTERM
W1026 11:11:34.221000 58142 site-packages/torch/distributed/elastic/multiprocessing/api.py:908] Sending process 58215 closing signal SIGTERM
W1026 11:11:34.226000 58142 site-packages/torch/distributed/elastic/multiprocessing/api.py:908] Sending process 58216 closing signal SIGTERM
W1026 11:11:34.230000 58142 site-packages/torch/distributed/elastic/multiprocessing/api.py:908] Sending process 58217 closing signal SIGTERM
W1026 11:11:34.235000 58142 site-packages/torch/distributed/elastic/multiprocessing/api.py:908] Sending process 58218 closing signal SIGTERM
W1026 11:11:34.239000 58142 site-packages/torch/distributed/elastic/multiprocessing/api.py:908] Sending process 58219 closing signal SIGTERM
W1026 11:11:34.244000 58142 site-packages/torch/distributed/elastic/multiprocessing/api.py:908] Sending process 58220 closing signal SIGTERM
W1026 11:11:34.252000 58142 site-packages/torch/distributed/elastic/multiprocessing/api.py:908] Sending process 58221 closing signal SIGTERM
Traceback (most recent call last):
  File "/workspace/miniconda3/envs/DC_SSDAE/bin/accelerate", line 7, in <module>
    sys.exit(main())
             ^^^^^^
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/accelerate/commands/accelerate_cli.py", line 50, in main
    args.func(args)
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/accelerate/commands/launch.py", line 1226, in launch_command
    multi_gpu_launcher(args)
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/accelerate/commands/launch.py", line 853, in multi_gpu_launcher
    distrib_run.run(args)
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/distributed/run.py", line 927, in run
    elastic_launch(
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/distributed/launcher/api.py", line 156, in __call__
    return launch_agent(self._config, self._entrypoint, list(args))
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/distributed/launcher/api.py", line 284, in launch_agent
    result = agent.run()
             ^^^^^^^^^^^
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/distributed/elastic/metrics/api.py", line 138, in wrapper
    result = f(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/distributed/elastic/agent/server/api.py", line 717, in run
    result = self._invoke_run(role)
             ^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/distributed/elastic/agent/server/api.py", line 881, in _invoke_run
    time.sleep(monitor_interval)
  File "/workspace/miniconda3/envs/DC_SSDAE/lib/python3.12/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 85, in _terminate_process_handler
    raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval)
torch.distributed.elastic.multiprocessing.api.SignalException: Process 58142 got signal: 15