training_logs/h100_maximized_training.log

nohup: ignoring input
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:70: FutureWarning: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
  self.scaler = GradScaler()
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:116: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.embeddings = torch.load(combined_path, map_location=self.device)
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:180: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.compressor.load_state_dict(torch.load('final_compressor_model.pth', map_location=self.device))
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:181: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.decompressor.load_state_dict(torch.load('final_decompressor_model.pth', map_location=self.device))
/data2/edwardsun/flow_home/cfg_dataset.py:253: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.embeddings = torch.load(combined_path, map_location='cpu')
Starting optimized training with batch_size=512, epochs=1500
Using GPU 0 for optimized H100 training
Mixed precision: True
Batch size: 512
Target epochs: 1500
Learning rate: 0.0016 -> 0.0008
✓ Mixed precision training enabled (BF16)
Loading ALL AMP embeddings from /data2/edwardsun/flow_project/peptide_embeddings/...
Loading combined embeddings from /data2/edwardsun/flow_project/peptide_embeddings/all_peptide_embeddings.pt...
✓ Loaded ALL embeddings: torch.Size([17968, 50, 1280])
Computing preprocessing statistics...
✓ Statistics computed and saved:
  Total embeddings: 17,968
  Mean: -0.0005 ± 0.0897
  Std: 0.0869 ± 0.1168
  Range: [-9.1738, 3.2894]
Initializing models...
✓ Model compiled with torch.compile for speedup
✓ Models initialized:
  Compressor parameters: 78,817,360
  Decompressor parameters: 39,458,720
  Flow model parameters: 50,779,584
Initializing datasets with FULL data...
Loading AMP embeddings from /data2/edwardsun/flow_project/peptide_embeddings/...
Loading combined embeddings from /data2/edwardsun/flow_project/peptide_embeddings/all_peptide_embeddings.pt (FULL DATA)...
✓ Loaded ALL embeddings: torch.Size([17968, 50, 1280])
Loading CFG data from FASTA: /home/edwardsun/flow/combined_final.fasta...
Parsing FASTA file: /home/edwardsun/flow/combined_final.fasta
Label assignment: >AP = AMP (0), >sp = Non-AMP (1)
✓ Parsed 6983 valid sequences from FASTA
  AMP sequences: 3306
  Non-AMP sequences: 3677
  Masked for CFG: 698
Loaded 6983 CFG sequences
Label distribution: [3306 3677]
Masked 698 labels for CFG training
Aligning AMP embeddings with CFG data...
Aligned 6983 samples
CFG Flow Dataset initialized:
  AMP embeddings: torch.Size([17968, 50, 1280])
  CFG labels: 6983
  Aligned samples: 6983
✓ Dataset initialized with FULL data:
  Total samples: 6,983
  Batch size: 512
  Batches per epoch: 14
  Total training steps: 21,000
  Validation every: 5,000 steps
Initializing optimizer and scheduler...
✓ Optimizer initialized:
  Base LR: 0.0016
  Min LR: 0.0008
  Warmup steps: 3000
  Weight decay: 0.01
  Gradient clip norm: 1.0
✓ Optimized Single GPU training setup complete with FULL DATA!
🚀 Starting Optimized Single GPU Flow Matching Training with FULL DATA
GPU: 0
Total iterations: 1500
Batch size: 512
Total samples: 6,983
Mixed precision: True
Estimated time: ~8-10 hours (overnight training with ALL data)
============================================================

Training Flow Model:   0%|          | 0/1500 [00:00<?, ?it/s]/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):

Training Flow Model:   0%|          | 1/1500 [00:47<19:58:38, 47.98s/it]Epoch    0 | Step      1/ 21000 | Loss: 2.348453 | LR: 1.60e-04 | Speed: 0.0 steps/s | ETA: 198.3h
Epoch    0 | Avg Loss: 1.154665 | LR: 1.67e-04 | Time: 48.0s | Samples: 6,983
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):

Training Flow Model:   0%|          | 2/1500 [00:53<9:39:12, 23.20s/it] Epoch    1 | Step     15/ 21000 | Loss: 1.012731 | LR: 1.67e-04 | Speed: 0.3 steps/s | ETA: 19.5h
Epoch    1 | Avg Loss: 1.011483 | LR: 1.73e-04 | Time: 5.9s | Samples: 6,983
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):

Training Flow Model:   0%|          | 3/1500 [00:57<5:51:46, 14.10s/it]Epoch    2 | Step     29/ 21000 | Loss: 1.011287 | LR: 1.74e-04 | Speed: 0.5 steps/s | ETA: 11.2h
Epoch    2 | Avg Loss: 1.008224 | LR: 1.80e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   0%|          | 4/1500 [01:00<4:05:10,  9.83s/it]Epoch    3 | Step     43/ 21000 | Loss: 1.003925 | LR: 1.81e-04 | Speed: 0.7 steps/s | ETA: 8.0h
Epoch    3 | Avg Loss: 1.101865 | LR: 1.87e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   0%|          | 5/1500 [01:03<3:06:38,  7.49s/it]Epoch    4 | Step     57/ 21000 | Loss: 1.003482 | LR: 1.87e-04 | Speed: 0.9 steps/s | ETA: 6.4h
Epoch    4 | Avg Loss: 1.002411 | LR: 1.94e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   0%|          | 6/1500 [01:07<2:31:52,  6.10s/it]Epoch    5 | Step     71/ 21000 | Loss: 0.996259 | LR: 1.94e-04 | Speed: 1.1 steps/s | ETA: 5.4h
Epoch    5 | Avg Loss: 1.002186 | LR: 2.00e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   0%|          | 7/1500 [01:10<2:10:05,  5.23s/it]Epoch    6 | Step     85/ 21000 | Loss: 0.976243 | LR: 2.01e-04 | Speed: 1.2 steps/s | ETA: 4.7h
Epoch    6 | Avg Loss: 0.928040 | LR: 2.07e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   1%|          | 8/1500 [01:14<1:56:10,  4.67s/it]Epoch    7 | Step     99/ 21000 | Loss: 0.834150 | LR: 2.08e-04 | Speed: 1.4 steps/s | ETA: 4.3h
Epoch    7 | Avg Loss: 0.764061 | LR: 2.14e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   1%|          | 9/1500 [01:17<1:46:09,  4.27s/it]Epoch    8 | Step    113/ 21000 | Loss: 0.666364 | LR: 2.14e-04 | Speed: 1.5 steps/s | ETA: 3.9h
Epoch    8 | Avg Loss: 0.599929 | LR: 2.20e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   1%|          | 10/1500 [01:20<1:38:41,  3.97s/it]Epoch    9 | Step    127/ 21000 | Loss: 0.514230 | LR: 2.21e-04 | Speed: 1.6 steps/s | ETA: 3.6h
Epoch    9 | Avg Loss: 0.464759 | LR: 2.27e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   1%|          | 11/1500 [01:24<1:33:37,  3.77s/it]Epoch   10 | Step    141/ 21000 | Loss: 0.395490 | LR: 2.28e-04 | Speed: 1.7 steps/s | ETA: 3.4h
Epoch   10 | Avg Loss: 0.357536 | LR: 2.34e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   1%|          | 12/1500 [01:27<1:31:07,  3.67s/it]Epoch   11 | Step    155/ 21000 | Loss: 0.300763 | LR: 2.34e-04 | Speed: 1.8 steps/s | ETA: 3.2h
Epoch   11 | Avg Loss: 0.273293 | LR: 2.41e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   1%|          | 13/1500 [01:30<1:28:33,  3.57s/it]Epoch   12 | Step    169/ 21000 | Loss: 0.225660 | LR: 2.41e-04 | Speed: 1.9 steps/s | ETA: 3.1h
Epoch   12 | Avg Loss: 0.213443 | LR: 2.47e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   1%|          | 14/1500 [01:34<1:27:28,  3.53s/it]Epoch   13 | Step    183/ 21000 | Loss: 0.195754 | LR: 2.48e-04 | Speed: 2.0 steps/s | ETA: 2.9h
Epoch   13 | Avg Loss: 0.170437 | LR: 2.54e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   1%|          | 15/1500 [01:37<1:25:31,  3.46s/it]Epoch   14 | Step    197/ 21000 | Loss: 0.187381 | LR: 2.55e-04 | Speed: 2.0 steps/s | ETA: 2.8h
Epoch   14 | Avg Loss: 0.144519 | LR: 2.61e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   1%|          | 16/1500 [01:40<1:24:28,  3.42s/it]Epoch   15 | Step    211/ 21000 | Loss: 0.138201 | LR: 2.61e-04 | Speed: 2.1 steps/s | ETA: 2.7h
Epoch   15 | Avg Loss: 0.123737 | LR: 2.68e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   1%|          | 17/1500 [01:44<1:24:32,  3.42s/it]Epoch   16 | Step    225/ 21000 | Loss: 0.111913 | LR: 2.68e-04 | Speed: 2.2 steps/s | ETA: 2.6h
Epoch   16 | Avg Loss: 0.116430 | LR: 2.74e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   1%|          | 18/1500 [01:47<1:24:34,  3.42s/it]Epoch   17 | Step    239/ 21000 | Loss: 0.113527 | LR: 2.75e-04 | Speed: 2.2 steps/s | ETA: 2.6h
Epoch   17 | Avg Loss: 0.110281 | LR: 2.81e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   1%|▏         | 19/1500 [01:51<1:23:57,  3.40s/it]Epoch   18 | Step    253/ 21000 | Loss: 0.089264 | LR: 2.81e-04 | Speed: 2.3 steps/s | ETA: 2.5h
Epoch   18 | Avg Loss: 0.097118 | LR: 2.88e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   1%|▏         | 20/1500 [01:54<1:22:43,  3.35s/it]Epoch   19 | Step    267/ 21000 | Loss: 0.101127 | LR: 2.88e-04 | Speed: 2.4 steps/s | ETA: 2.4h
Epoch   19 | Avg Loss: 0.098588 | LR: 2.94e-04 | Time: 3.2s | Samples: 6,983

Training Flow Model:   1%|▏         | 21/1500 [01:57<1:21:58,  3.33s/it]Epoch   20 | Step    281/ 21000 | Loss: 0.093898 | LR: 2.95e-04 | Speed: 2.4 steps/s | ETA: 2.4h
Epoch   20 | Avg Loss: 0.097397 | LR: 3.01e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   1%|▏         | 22/1500 [02:00<1:21:45,  3.32s/it]Epoch   21 | Step    295/ 21000 | Loss: 0.099576 | LR: 3.02e-04 | Speed: 2.5 steps/s | ETA: 2.3h
Epoch   21 | Avg Loss: 0.094745 | LR: 3.08e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   2%|▏         | 23/1500 [02:04<1:23:02,  3.37s/it]Epoch   22 | Step    309/ 21000 | Loss: 0.093006 | LR: 3.08e-04 | Speed: 2.5 steps/s | ETA: 2.3h
Epoch   22 | Avg Loss: 0.089745 | LR: 3.15e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   2%|▏         | 24/1500 [02:07<1:22:00,  3.33s/it]Epoch   23 | Step    323/ 21000 | Loss: 0.081982 | LR: 3.15e-04 | Speed: 2.6 steps/s | ETA: 2.2h
Epoch   23 | Avg Loss: 0.082143 | LR: 3.21e-04 | Time: 3.2s | Samples: 6,983

Training Flow Model:   2%|▏         | 25/1500 [02:11<1:22:49,  3.37s/it]Epoch   24 | Step    337/ 21000 | Loss: 0.080047 | LR: 3.22e-04 | Speed: 2.6 steps/s | ETA: 2.2h
Epoch   24 | Avg Loss: 0.087128 | LR: 3.28e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   2%|▏         | 26/1500 [02:14<1:22:35,  3.36s/it]Epoch   25 | Step    351/ 21000 | Loss: 0.088592 | LR: 3.28e-04 | Speed: 2.6 steps/s | ETA: 2.2h
Epoch   25 | Avg Loss: 0.078387 | LR: 3.35e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   2%|▏         | 27/1500 [02:17<1:22:15,  3.35s/it]Epoch   26 | Step    365/ 21000 | Loss: 0.085473 | LR: 3.35e-04 | Speed: 2.7 steps/s | ETA: 2.1h
Epoch   26 | Avg Loss: 0.073784 | LR: 3.41e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   2%|▏         | 28/1500 [02:21<1:22:12,  3.35s/it]Epoch   27 | Step    379/ 21000 | Loss: 0.070519 | LR: 3.42e-04 | Speed: 2.7 steps/s | ETA: 2.1h
Epoch   27 | Avg Loss: 0.077967 | LR: 3.48e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   2%|▏         | 29/1500 [02:24<1:22:28,  3.36s/it]Epoch   28 | Step    393/ 21000 | Loss: 0.063307 | LR: 3.49e-04 | Speed: 2.7 steps/s | ETA: 2.1h
Epoch   28 | Avg Loss: 0.074035 | LR: 3.55e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   2%|▏         | 30/1500 [02:27<1:21:47,  3.34s/it]Epoch   29 | Step    407/ 21000 | Loss: 0.073874 | LR: 3.55e-04 | Speed: 2.8 steps/s | ETA: 2.1h
Epoch   29 | Avg Loss: 0.073863 | LR: 3.62e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   2%|▏         | 31/1500 [02:31<1:22:13,  3.36s/it]Epoch   30 | Step    421/ 21000 | Loss: 0.075808 | LR: 3.62e-04 | Speed: 2.8 steps/s | ETA: 2.0h
Epoch   30 | Avg Loss: 0.068544 | LR: 3.68e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   2%|▏         | 32/1500 [02:34<1:22:36,  3.38s/it]Epoch   31 | Step    435/ 21000 | Loss: 0.072362 | LR: 3.69e-04 | Speed: 2.8 steps/s | ETA: 2.0h
Epoch   31 | Avg Loss: 0.072260 | LR: 3.75e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   2%|▏         | 33/1500 [02:37<1:21:59,  3.35s/it]Epoch   32 | Step    449/ 21000 | Loss: 0.071200 | LR: 3.76e-04 | Speed: 2.9 steps/s | ETA: 2.0h
Epoch   32 | Avg Loss: 0.068997 | LR: 3.82e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   2%|▏         | 34/1500 [02:41<1:22:07,  3.36s/it]Epoch   33 | Step    463/ 21000 | Loss: 0.070137 | LR: 3.82e-04 | Speed: 2.9 steps/s | ETA: 2.0h
Epoch   33 | Avg Loss: 0.068473 | LR: 3.88e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   2%|▏         | 35/1500 [02:44<1:21:46,  3.35s/it]Epoch   34 | Step    477/ 21000 | Loss: 0.061482 | LR: 3.89e-04 | Speed: 2.9 steps/s | ETA: 2.0h
Epoch   34 | Avg Loss: 0.066994 | LR: 3.95e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   2%|▏         | 36/1500 [02:48<1:22:12,  3.37s/it]Epoch   35 | Step    491/ 21000 | Loss: 0.074752 | LR: 3.96e-04 | Speed: 2.9 steps/s | ETA: 1.9h
Epoch   35 | Avg Loss: 0.070879 | LR: 4.02e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   2%|▏         | 37/1500 [02:51<1:21:20,  3.34s/it]Epoch   36 | Step    505/ 21000 | Loss: 0.057749 | LR: 4.02e-04 | Speed: 3.0 steps/s | ETA: 1.9h
Epoch   36 | Avg Loss: 0.057021 | LR: 4.09e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   3%|▎         | 38/1500 [02:54<1:21:50,  3.36s/it]Epoch   37 | Step    519/ 21000 | Loss: 0.061772 | LR: 4.09e-04 | Speed: 3.0 steps/s | ETA: 1.9h
Epoch   37 | Avg Loss: 0.057016 | LR: 4.15e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   3%|▎         | 39/1500 [02:57<1:20:50,  3.32s/it]Epoch   38 | Step    533/ 21000 | Loss: 0.055348 | LR: 4.16e-04 | Speed: 3.0 steps/s | ETA: 1.9h
Epoch   38 | Avg Loss: 0.062512 | LR: 4.22e-04 | Time: 3.2s | Samples: 6,983

Training Flow Model:   3%|▎         | 40/1500 [03:01<1:21:38,  3.36s/it]Epoch   39 | Step    547/ 21000 | Loss: 0.062446 | LR: 4.23e-04 | Speed: 3.0 steps/s | ETA: 1.9h
Epoch   39 | Avg Loss: 0.055946 | LR: 4.29e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   3%|▎         | 41/1500 [03:04<1:21:52,  3.37s/it]Epoch   40 | Step    561/ 21000 | Loss: 0.055766 | LR: 4.29e-04 | Speed: 3.1 steps/s | ETA: 1.9h
Epoch   40 | Avg Loss: 0.052785 | LR: 4.36e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   3%|▎         | 42/1500 [03:08<1:21:38,  3.36s/it]Epoch   41 | Step    575/ 21000 | Loss: 0.066500 | LR: 4.36e-04 | Speed: 3.1 steps/s | ETA: 1.8h
Epoch   41 | Avg Loss: 0.058182 | LR: 4.42e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   3%|▎         | 43/1500 [03:11<1:21:14,  3.35s/it]Epoch   42 | Step    589/ 21000 | Loss: 0.058983 | LR: 4.43e-04 | Speed: 3.1 steps/s | ETA: 1.8h
Epoch   42 | Avg Loss: 0.054515 | LR: 4.49e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   3%|▎         | 44/1500 [03:14<1:21:09,  3.34s/it]Epoch   43 | Step    603/ 21000 | Loss: 0.050246 | LR: 4.49e-04 | Speed: 3.1 steps/s | ETA: 1.8h
Epoch   43 | Avg Loss: 0.055557 | LR: 4.56e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   3%|▎         | 45/1500 [03:18<1:21:28,  3.36s/it]Epoch   44 | Step    617/ 21000 | Loss: 0.058090 | LR: 4.56e-04 | Speed: 3.1 steps/s | ETA: 1.8h
Epoch   44 | Avg Loss: 0.053506 | LR: 4.62e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   3%|▎         | 46/1500 [03:21<1:21:37,  3.37s/it]Epoch   45 | Step    631/ 21000 | Loss: 0.050240 | LR: 4.63e-04 | Speed: 3.2 steps/s | ETA: 1.8h
Epoch   45 | Avg Loss: 0.053557 | LR: 4.69e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   3%|▎         | 47/1500 [03:24<1:21:55,  3.38s/it]Epoch   46 | Step    645/ 21000 | Loss: 0.056234 | LR: 4.70e-04 | Speed: 3.2 steps/s | ETA: 1.8h
Epoch   46 | Avg Loss: 0.050458 | LR: 4.76e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   3%|▎         | 48/1500 [03:28<1:21:08,  3.35s/it]Epoch   47 | Step    659/ 21000 | Loss: 0.050834 | LR: 4.76e-04 | Speed: 3.2 steps/s | ETA: 1.8h
Epoch   47 | Avg Loss: 0.053890 | LR: 4.83e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   3%|▎         | 49/1500 [03:31<1:21:41,  3.38s/it]Epoch   48 | Step    673/ 21000 | Loss: 0.043349 | LR: 4.83e-04 | Speed: 3.2 steps/s | ETA: 1.8h
Epoch   48 | Avg Loss: 0.049642 | LR: 4.89e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   3%|▎         | 50/1500 [03:34<1:20:26,  3.33s/it]Epoch   49 | Step    687/ 21000 | Loss: 0.055819 | LR: 4.90e-04 | Speed: 3.2 steps/s | ETA: 1.8h
Epoch   49 | Avg Loss: 0.051639 | LR: 4.96e-04 | Time: 3.2s | Samples: 6,983

Training Flow Model:   3%|▎         | 51/1500 [03:38<1:19:57,  3.31s/it]Epoch   50 | Step    701/ 21000 | Loss: 0.048318 | LR: 4.96e-04 | Speed: 3.2 steps/s | ETA: 1.7h
Epoch   50 | Avg Loss: 0.051892 | LR: 5.03e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   3%|▎         | 52/1500 [03:41<1:19:38,  3.30s/it]Epoch   51 | Step    715/ 21000 | Loss: 0.037495 | LR: 5.03e-04 | Speed: 3.2 steps/s | ETA: 1.7h
Epoch   51 | Avg Loss: 0.043957 | LR: 5.09e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   4%|▎         | 53/1500 [03:44<1:19:25,  3.29s/it]Epoch   52 | Step    729/ 21000 | Loss: 0.041565 | LR: 5.10e-04 | Speed: 3.3 steps/s | ETA: 1.7h
Epoch   52 | Avg Loss: 0.049808 | LR: 5.16e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   4%|▎         | 54/1500 [03:47<1:19:11,  3.29s/it]Epoch   53 | Step    743/ 21000 | Loss: 0.051807 | LR: 5.17e-04 | Speed: 3.3 steps/s | ETA: 1.7h
Epoch   53 | Avg Loss: 0.059701 | LR: 5.23e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   4%|▎         | 55/1500 [03:51<1:20:01,  3.32s/it]Epoch   54 | Step    757/ 21000 | Loss: 0.050800 | LR: 5.23e-04 | Speed: 3.3 steps/s | ETA: 1.7h
Epoch   54 | Avg Loss: 0.050593 | LR: 5.30e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   4%|▎         | 56/1500 [03:54<1:21:36,  3.39s/it]Epoch   55 | Step    771/ 21000 | Loss: 0.057050 | LR: 5.30e-04 | Speed: 3.3 steps/s | ETA: 1.7h
Epoch   55 | Avg Loss: 0.051511 | LR: 5.36e-04 | Time: 3.6s | Samples: 6,983

Training Flow Model:   4%|▍         | 57/1500 [03:58<1:21:33,  3.39s/it]Epoch   56 | Step    785/ 21000 | Loss: 0.056579 | LR: 5.37e-04 | Speed: 3.3 steps/s | ETA: 1.7h
Epoch   56 | Avg Loss: 0.047250 | LR: 5.43e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   4%|▍         | 58/1500 [04:01<1:20:45,  3.36s/it]Epoch   57 | Step    799/ 21000 | Loss: 0.050855 | LR: 5.44e-04 | Speed: 3.3 steps/s | ETA: 1.7h
Epoch   57 | Avg Loss: 0.053542 | LR: 5.50e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   4%|▍         | 59/1500 [04:05<1:20:58,  3.37s/it]Epoch   58 | Step    813/ 21000 | Loss: 0.049122 | LR: 5.50e-04 | Speed: 3.3 steps/s | ETA: 1.7h
Epoch   58 | Avg Loss: 0.048603 | LR: 5.56e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   4%|▍         | 60/1500 [04:08<1:20:57,  3.37s/it]Epoch   59 | Step    827/ 21000 | Loss: 0.051872 | LR: 5.57e-04 | Speed: 3.3 steps/s | ETA: 1.7h
Epoch   59 | Avg Loss: 0.044602 | LR: 5.63e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   4%|▍         | 61/1500 [04:11<1:21:36,  3.40s/it]Epoch   60 | Step    841/ 21000 | Loss: 0.044589 | LR: 5.64e-04 | Speed: 3.4 steps/s | ETA: 1.7h
Epoch   60 | Avg Loss: 0.046526 | LR: 5.70e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   4%|▍         | 62/1500 [04:15<1:21:03,  3.38s/it]Epoch   61 | Step    855/ 21000 | Loss: 0.041265 | LR: 5.70e-04 | Speed: 3.4 steps/s | ETA: 1.7h
Epoch   61 | Avg Loss: 0.051947 | LR: 5.77e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   4%|▍         | 63/1500 [04:18<1:19:54,  3.34s/it]Epoch   62 | Step    869/ 21000 | Loss: 0.046667 | LR: 5.77e-04 | Speed: 3.4 steps/s | ETA: 1.7h
Epoch   62 | Avg Loss: 0.060930 | LR: 5.83e-04 | Time: 3.2s | Samples: 6,983

Training Flow Model:   4%|▍         | 64/1500 [04:21<1:20:41,  3.37s/it]Epoch   63 | Step    883/ 21000 | Loss: 0.052620 | LR: 5.84e-04 | Speed: 3.4 steps/s | ETA: 1.6h
Epoch   63 | Avg Loss: 0.050808 | LR: 5.90e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   4%|▍         | 65/1500 [04:25<1:20:27,  3.36s/it]Epoch   64 | Step    897/ 21000 | Loss: 0.057084 | LR: 5.91e-04 | Speed: 3.4 steps/s | ETA: 1.6h
Epoch   64 | Avg Loss: 0.056893 | LR: 5.97e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   4%|▍         | 66/1500 [04:28<1:20:06,  3.35s/it]Epoch   65 | Step    911/ 21000 | Loss: 0.043265 | LR: 5.97e-04 | Speed: 3.4 steps/s | ETA: 1.6h
Epoch   65 | Avg Loss: 0.046618 | LR: 6.04e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   4%|▍         | 67/1500 [04:31<1:20:12,  3.36s/it]Epoch   66 | Step    925/ 21000 | Loss: 0.051255 | LR: 6.04e-04 | Speed: 3.4 steps/s | ETA: 1.6h
Epoch   66 | Avg Loss: 0.047629 | LR: 6.10e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   5%|▍         | 68/1500 [04:35<1:20:10,  3.36s/it]Epoch   67 | Step    939/ 21000 | Loss: 0.059373 | LR: 6.11e-04 | Speed: 3.4 steps/s | ETA: 1.6h
Epoch   67 | Avg Loss: 0.049539 | LR: 6.17e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   5%|▍         | 69/1500 [04:38<1:19:39,  3.34s/it]Epoch   68 | Step    953/ 21000 | Loss: 0.052057 | LR: 6.17e-04 | Speed: 3.4 steps/s | ETA: 1.6h
Epoch   68 | Avg Loss: 0.045869 | LR: 6.24e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▍         | 70/1500 [04:41<1:19:07,  3.32s/it]Epoch   69 | Step    967/ 21000 | Loss: 0.042184 | LR: 6.24e-04 | Speed: 3.4 steps/s | ETA: 1.6h
Epoch   69 | Avg Loss: 0.046375 | LR: 6.30e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▍         | 71/1500 [04:45<1:18:48,  3.31s/it]Epoch   70 | Step    981/ 21000 | Loss: 0.037335 | LR: 6.31e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   70 | Avg Loss: 0.044112 | LR: 6.37e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▍         | 72/1500 [04:48<1:19:08,  3.33s/it]Epoch   71 | Step    995/ 21000 | Loss: 0.040643 | LR: 6.38e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   71 | Avg Loss: 0.040547 | LR: 6.44e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   5%|▍         | 73/1500 [04:51<1:19:13,  3.33s/it]Epoch   72 | Step   1009/ 21000 | Loss: 0.040631 | LR: 6.44e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   72 | Avg Loss: 0.043072 | LR: 6.51e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▍         | 74/1500 [04:55<1:18:22,  3.30s/it]Epoch   73 | Step   1023/ 21000 | Loss: 0.052481 | LR: 6.51e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   73 | Avg Loss: 0.054466 | LR: 6.57e-04 | Time: 3.2s | Samples: 6,983

Training Flow Model:   5%|▌         | 75/1500 [04:58<1:18:17,  3.30s/it]Epoch   74 | Step   1037/ 21000 | Loss: 0.059234 | LR: 6.58e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   74 | Avg Loss: 0.056152 | LR: 6.64e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▌         | 76/1500 [05:01<1:18:00,  3.29s/it]Epoch   75 | Step   1051/ 21000 | Loss: 0.049168 | LR: 6.64e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   75 | Avg Loss: 0.043867 | LR: 6.71e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▌         | 77/1500 [05:04<1:18:14,  3.30s/it]Epoch   76 | Step   1065/ 21000 | Loss: 0.044122 | LR: 6.71e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   76 | Avg Loss: 0.046304 | LR: 6.77e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▌         | 78/1500 [05:08<1:18:42,  3.32s/it]Epoch   77 | Step   1079/ 21000 | Loss: 0.043278 | LR: 6.78e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   77 | Avg Loss: 0.044190 | LR: 6.84e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   5%|▌         | 79/1500 [05:11<1:18:26,  3.31s/it]Epoch   78 | Step   1093/ 21000 | Loss: 0.045077 | LR: 6.85e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   78 | Avg Loss: 0.039687 | LR: 6.91e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▌         | 80/1500 [05:14<1:18:05,  3.30s/it]Epoch   79 | Step   1107/ 21000 | Loss: 0.043229 | LR: 6.91e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   79 | Avg Loss: 0.045789 | LR: 6.98e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▌         | 81/1500 [05:18<1:17:51,  3.29s/it]Epoch   80 | Step   1121/ 21000 | Loss: 0.045537 | LR: 6.98e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   80 | Avg Loss: 0.048807 | LR: 7.04e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   5%|▌         | 82/1500 [05:21<1:18:22,  3.32s/it]Epoch   81 | Step   1135/ 21000 | Loss: 0.047845 | LR: 7.05e-04 | Speed: 3.5 steps/s | ETA: 1.6h
Epoch   81 | Avg Loss: 0.046174 | LR: 7.11e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   6%|▌         | 83/1500 [05:24<1:18:18,  3.32s/it]Epoch   82 | Step   1149/ 21000 | Loss: 0.042759 | LR: 7.12e-04 | Speed: 3.6 steps/s | ETA: 1.6h
Epoch   82 | Avg Loss: 0.040676 | LR: 7.18e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   6%|▌         | 84/1500 [05:28<1:18:42,  3.34s/it]Epoch   83 | Step   1163/ 21000 | Loss: 0.031483 | LR: 7.18e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   83 | Avg Loss: 0.039434 | LR: 7.24e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   6%|▌         | 85/1500 [05:31<1:19:31,  3.37s/it]Epoch   84 | Step   1177/ 21000 | Loss: 0.043087 | LR: 7.25e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   84 | Avg Loss: 0.053384 | LR: 7.31e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   6%|▌         | 86/1500 [05:35<1:19:46,  3.38s/it]Epoch   85 | Step   1191/ 21000 | Loss: 0.045207 | LR: 7.32e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   85 | Avg Loss: 0.047736 | LR: 7.38e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   6%|▌         | 87/1500 [05:38<1:20:30,  3.42s/it]Epoch   86 | Step   1205/ 21000 | Loss: 0.038893 | LR: 7.38e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   86 | Avg Loss: 0.044388 | LR: 7.45e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   6%|▌         | 88/1500 [05:41<1:19:23,  3.37s/it]Epoch   87 | Step   1219/ 21000 | Loss: 0.036651 | LR: 7.45e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   87 | Avg Loss: 0.043202 | LR: 7.51e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   6%|▌         | 89/1500 [05:45<1:18:59,  3.36s/it]Epoch   88 | Step   1233/ 21000 | Loss: 0.049969 | LR: 7.52e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   88 | Avg Loss: 0.039903 | LR: 7.58e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   6%|▌         | 90/1500 [05:48<1:19:05,  3.37s/it]Epoch   89 | Step   1247/ 21000 | Loss: 0.059831 | LR: 7.59e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   89 | Avg Loss: 0.049894 | LR: 7.65e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   6%|▌         | 91/1500 [05:52<1:19:42,  3.39s/it]Epoch   90 | Step   1261/ 21000 | Loss: 0.057150 | LR: 7.65e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   90 | Avg Loss: 0.054641 | LR: 7.72e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   6%|▌         | 92/1500 [05:55<1:19:05,  3.37s/it]Epoch   91 | Step   1275/ 21000 | Loss: 0.056272 | LR: 7.72e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   91 | Avg Loss: 0.060308 | LR: 7.78e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   6%|▌         | 93/1500 [05:58<1:19:07,  3.37s/it]Epoch   92 | Step   1289/ 21000 | Loss: 0.054234 | LR: 7.79e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   92 | Avg Loss: 0.051279 | LR: 7.85e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   6%|▋         | 94/1500 [06:02<1:18:50,  3.36s/it]Epoch   93 | Step   1303/ 21000 | Loss: 0.038766 | LR: 7.85e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   93 | Avg Loss: 0.040706 | LR: 7.92e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   6%|▋         | 95/1500 [06:05<1:18:41,  3.36s/it]Epoch   94 | Step   1317/ 21000 | Loss: 0.034445 | LR: 7.92e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   94 | Avg Loss: 0.037526 | LR: 7.98e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   6%|▋         | 96/1500 [06:08<1:19:09,  3.38s/it]Epoch   95 | Step   1331/ 21000 | Loss: 0.042873 | LR: 7.99e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   95 | Avg Loss: 0.040381 | LR: 8.05e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   6%|▋         | 97/1500 [06:12<1:18:55,  3.38s/it]Epoch   96 | Step   1345/ 21000 | Loss: 0.033452 | LR: 8.06e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   96 | Avg Loss: 0.044442 | LR: 8.12e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   7%|▋         | 98/1500 [06:15<1:18:20,  3.35s/it]Epoch   97 | Step   1359/ 21000 | Loss: 0.049492 | LR: 8.12e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   97 | Avg Loss: 0.049140 | LR: 8.19e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   7%|▋         | 99/1500 [06:18<1:18:08,  3.35s/it]Epoch   98 | Step   1373/ 21000 | Loss: 0.036935 | LR: 8.19e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   98 | Avg Loss: 0.039066 | LR: 8.25e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   7%|▋         | 100/1500 [06:22<1:17:55,  3.34s/it]Epoch   99 | Step   1387/ 21000 | Loss: 0.034987 | LR: 8.26e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch   99 | Avg Loss: 0.035230 | LR: 8.32e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   7%|▋         | 101/1500 [06:25<1:17:34,  3.33s/it]Epoch  100 | Step   1401/ 21000 | Loss: 0.037965 | LR: 8.32e-04 | Speed: 3.6 steps/s | ETA: 1.5h
Epoch  100 | Avg Loss: 0.039429 | LR: 8.39e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   7%|▋         | 102/1500 [06:28<1:17:25,  3.32s/it]Epoch  101 | Step   1415/ 21000 | Loss: 0.044940 | LR: 8.39e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  101 | Avg Loss: 0.043091 | LR: 8.45e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   7%|▋         | 103/1500 [06:32<1:17:17,  3.32s/it]Epoch  102 | Step   1429/ 21000 | Loss: 0.046929 | LR: 8.46e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  102 | Avg Loss: 0.045039 | LR: 8.52e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   7%|▋         | 104/1500 [06:35<1:18:00,  3.35s/it]Epoch  103 | Step   1443/ 21000 | Loss: 0.040008 | LR: 8.53e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  103 | Avg Loss: 0.035983 | LR: 8.59e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   7%|▋         | 105/1500 [06:39<1:19:01,  3.40s/it]Epoch  104 | Step   1457/ 21000 | Loss: 0.037500 | LR: 8.59e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  104 | Avg Loss: 0.035712 | LR: 8.66e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   7%|▋         | 106/1500 [06:42<1:18:31,  3.38s/it]Epoch  105 | Step   1471/ 21000 | Loss: 0.038647 | LR: 8.66e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  105 | Avg Loss: 0.037083 | LR: 8.72e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   7%|▋         | 107/1500 [06:45<1:18:31,  3.38s/it]Epoch  106 | Step   1485/ 21000 | Loss: 0.032988 | LR: 8.73e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  106 | Avg Loss: 0.034224 | LR: 8.79e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   7%|▋         | 108/1500 [06:49<1:18:06,  3.37s/it]Epoch  107 | Step   1499/ 21000 | Loss: 0.041273 | LR: 8.80e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  107 | Avg Loss: 0.054691 | LR: 8.86e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   7%|▋         | 109/1500 [06:52<1:18:04,  3.37s/it]Epoch  108 | Step   1513/ 21000 | Loss: 0.066792 | LR: 8.86e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  108 | Avg Loss: 0.056016 | LR: 8.92e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   7%|▋         | 110/1500 [06:55<1:17:38,  3.35s/it]Epoch  109 | Step   1527/ 21000 | Loss: 0.056582 | LR: 8.93e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  109 | Avg Loss: 0.047941 | LR: 8.99e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   7%|▋         | 111/1500 [06:59<1:16:48,  3.32s/it]Epoch  110 | Step   1541/ 21000 | Loss: 0.038037 | LR: 9.00e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  110 | Avg Loss: 0.039082 | LR: 9.06e-04 | Time: 3.2s | Samples: 6,983

Training Flow Model:   7%|▋         | 112/1500 [07:02<1:17:21,  3.34s/it]Epoch  111 | Step   1555/ 21000 | Loss: 0.036922 | LR: 9.06e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  111 | Avg Loss: 0.037414 | LR: 9.13e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   8%|▊         | 113/1500 [07:05<1:16:59,  3.33s/it]Epoch  112 | Step   1569/ 21000 | Loss: 0.050282 | LR: 9.13e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  112 | Avg Loss: 0.052520 | LR: 9.19e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   8%|▊         | 114/1500 [07:09<1:18:00,  3.38s/it]Epoch  113 | Step   1583/ 21000 | Loss: 0.064816 | LR: 9.20e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  113 | Avg Loss: 0.060289 | LR: 9.26e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   8%|▊         | 115/1500 [07:12<1:18:39,  3.41s/it]Epoch  114 | Step   1597/ 21000 | Loss: 0.041934 | LR: 9.27e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  114 | Avg Loss: 0.044027 | LR: 9.33e-04 | Time: 3.5s | Samples: 6,983

Training Flow Model:   8%|▊         | 116/1500 [07:16<1:18:17,  3.39s/it]Epoch  115 | Step   1611/ 21000 | Loss: 0.034175 | LR: 9.33e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  115 | Avg Loss: 0.039415 | LR: 9.40e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   8%|▊         | 117/1500 [07:19<1:17:52,  3.38s/it]Epoch  116 | Step   1625/ 21000 | Loss: 0.042169 | LR: 9.40e-04 | Speed: 3.7 steps/s | ETA: 1.5h
Epoch  116 | Avg Loss: 0.037331 | LR: 9.46e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   8%|▊         | 118/1500 [07:22<1:18:13,  3.40s/it]Epoch  117 | Step   1639/ 21000 | Loss: 0.035896 | LR: 9.47e-04 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  117 | Avg Loss: 0.038203 | LR: 9.53e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   8%|▊         | 119/1500 [07:26<1:17:14,  3.36s/it]Epoch  118 | Step   1653/ 21000 | Loss: 0.044916 | LR: 9.53e-04 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  118 | Avg Loss: 0.039857 | LR: 9.60e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   8%|▊         | 120/1500 [07:29<1:16:45,  3.34s/it]Epoch  119 | Step   1667/ 21000 | Loss: 0.036022 | LR: 9.60e-04 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  119 | Avg Loss: 0.038705 | LR: 9.66e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   8%|▊         | 121/1500 [07:32<1:17:09,  3.36s/it]Epoch  120 | Step   1681/ 21000 | Loss: 0.053714 | LR: 9.67e-04 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  120 | Avg Loss: 0.058985 | LR: 9.73e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   8%|▊         | 122/1500 [07:36<1:17:01,  3.35s/it]Epoch  121 | Step   1695/ 21000 | Loss: 0.053556 | LR: 9.74e-04 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  121 | Avg Loss: 0.041034 | LR: 9.80e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   8%|▊         | 123/1500 [07:39<1:16:34,  3.34s/it]Epoch  122 | Step   1709/ 21000 | Loss: 0.031987 | LR: 9.80e-04 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  122 | Avg Loss: 0.034538 | LR: 9.87e-04 | Time: 3.3s | Samples: 6,983

Training Flow Model:   8%|▊         | 124/1500 [07:42<1:16:48,  3.35s/it]Epoch  123 | Step   1723/ 21000 | Loss: 0.028273 | LR: 9.87e-04 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  123 | Avg Loss: 0.040622 | LR: 9.93e-04 | Time: 3.4s | Samples: 6,983

Training Flow Model:   8%|▊         | 125/1500 [07:46<1:16:48,  3.35s/it]Epoch  124 | Step   1737/ 21000 | Loss: 0.065265 | LR: 9.94e-04 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  124 | Avg Loss: 0.081324 | LR: 1.00e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:   8%|▊         | 126/1500 [07:49<1:17:06,  3.37s/it]Epoch  125 | Step   1751/ 21000 | Loss: 0.099073 | LR: 1.00e-03 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  125 | Avg Loss: 0.103266 | LR: 1.01e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:   8%|▊         | 127/1500 [07:52<1:17:15,  3.38s/it]Epoch  126 | Step   1765/ 21000 | Loss: 0.116206 | LR: 1.01e-03 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  126 | Avg Loss: 0.114660 | LR: 1.01e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:   9%|▊         | 128/1500 [07:56<1:16:20,  3.34s/it]Epoch  127 | Step   1779/ 21000 | Loss: 0.109866 | LR: 1.01e-03 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  127 | Avg Loss: 0.094334 | LR: 1.02e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▊         | 129/1500 [07:59<1:16:04,  3.33s/it]Epoch  128 | Step   1793/ 21000 | Loss: 0.082153 | LR: 1.02e-03 | Speed: 3.7 steps/s | ETA: 1.4h
Epoch  128 | Avg Loss: 0.070192 | LR: 1.03e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▊         | 130/1500 [08:02<1:15:52,  3.32s/it]Epoch  129 | Step   1807/ 21000 | Loss: 0.045357 | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  129 | Avg Loss: 0.047373 | LR: 1.03e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▊         | 131/1500 [08:06<1:15:16,  3.30s/it]Epoch  130 | Step   1821/ 21000 | Loss: 0.046692 | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  130 | Avg Loss: 0.043258 | LR: 1.04e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:   9%|▉         | 132/1500 [08:09<1:15:25,  3.31s/it]Epoch  131 | Step   1835/ 21000 | Loss: 0.032795 | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  131 | Avg Loss: 0.051812 | LR: 1.05e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▉         | 133/1500 [08:12<1:15:32,  3.32s/it]Epoch  132 | Step   1849/ 21000 | Loss: 0.056749 | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  132 | Avg Loss: 0.048600 | LR: 1.05e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▉         | 134/1500 [08:16<1:15:27,  3.31s/it]Epoch  133 | Step   1863/ 21000 | Loss: 0.042399 | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  133 | Avg Loss: 0.042586 | LR: 1.06e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▉         | 135/1500 [08:19<1:15:14,  3.31s/it]Epoch  134 | Step   1877/ 21000 | Loss: 0.044964 | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  134 | Avg Loss: 0.048532 | LR: 1.07e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▉         | 136/1500 [08:22<1:15:20,  3.31s/it]Epoch  135 | Step   1891/ 21000 | Loss: 0.057734 | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  135 | Avg Loss: 0.048324 | LR: 1.07e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▉         | 137/1500 [08:25<1:15:08,  3.31s/it]Epoch  136 | Step   1905/ 21000 | Loss: 0.043287 | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  136 | Avg Loss: 0.038560 | LR: 1.08e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▉         | 138/1500 [08:29<1:15:21,  3.32s/it]Epoch  137 | Step   1919/ 21000 | Loss: 0.037605 | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  137 | Avg Loss: 0.043462 | LR: 1.09e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▉         | 139/1500 [08:32<1:15:11,  3.31s/it]Epoch  138 | Step   1933/ 21000 | Loss: 0.043478 | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  138 | Avg Loss: 0.042978 | LR: 1.09e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▉         | 140/1500 [08:36<1:15:58,  3.35s/it]Epoch  139 | Step   1947/ 21000 | Loss: 0.031243 | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  139 | Avg Loss: 0.041040 | LR: 1.10e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:   9%|▉         | 141/1500 [08:39<1:15:50,  3.35s/it]Epoch  140 | Step   1961/ 21000 | Loss: 0.044837 | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  140 | Avg Loss: 0.039195 | LR: 1.11e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:   9%|▉         | 142/1500 [08:42<1:16:55,  3.40s/it]Epoch  141 | Step   1975/ 21000 | Loss: 0.039025 | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  141 | Avg Loss: 0.038292 | LR: 1.11e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  10%|▉         | 143/1500 [08:46<1:15:56,  3.36s/it]Epoch  142 | Step   1989/ 21000 | Loss: 0.032414 | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  142 | Avg Loss: 0.033116 | LR: 1.12e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  10%|▉         | 144/1500 [08:49<1:15:17,  3.33s/it]Epoch  143 | Step   2003/ 21000 | Loss: 0.034082 | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  143 | Avg Loss: 0.041510 | LR: 1.13e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  10%|▉         | 145/1500 [08:52<1:15:48,  3.36s/it]Epoch  144 | Step   2017/ 21000 | Loss: 0.052641 | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  144 | Avg Loss: 0.055978 | LR: 1.13e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  10%|▉         | 146/1500 [08:56<1:15:24,  3.34s/it]Epoch  145 | Step   2031/ 21000 | Loss: 0.064799 | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  145 | Avg Loss: 0.064812 | LR: 1.14e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  10%|▉         | 147/1500 [08:59<1:16:12,  3.38s/it]Epoch  146 | Step   2045/ 21000 | Loss: 0.046493 | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  146 | Avg Loss: 0.043557 | LR: 1.15e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  10%|▉         | 148/1500 [09:03<1:16:25,  3.39s/it]Epoch  147 | Step   2059/ 21000 | Loss: 0.040831 | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  147 | Avg Loss: 0.047060 | LR: 1.15e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  10%|▉         | 149/1500 [09:06<1:17:16,  3.43s/it]Epoch  148 | Step   2073/ 21000 | Loss: 0.059485 | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  148 | Avg Loss: 0.055089 | LR: 1.16e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  10%|█         | 150/1500 [09:09<1:16:53,  3.42s/it]Epoch  149 | Step   2087/ 21000 | Loss: 0.051692 | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  149 | Avg Loss: 0.043558 | LR: 1.17e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  10%|█         | 151/1500 [09:13<1:16:00,  3.38s/it]Epoch  150 | Step   2101/ 21000 | Loss: 0.036517 | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  150 | Avg Loss: 0.031243 | LR: 1.17e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  10%|█         | 152/1500 [09:16<1:15:44,  3.37s/it]Epoch  151 | Step   2115/ 21000 | Loss: 0.034116 | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  151 | Avg Loss: 0.036845 | LR: 1.18e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  10%|█         | 153/1500 [09:20<1:16:03,  3.39s/it]Epoch  152 | Step   2129/ 21000 | Loss: 0.053794 | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  152 | Avg Loss: 0.048498 | LR: 1.19e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  10%|█         | 154/1500 [09:23<1:15:33,  3.37s/it]Epoch  153 | Step   2143/ 21000 | Loss: 0.051553 | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  153 | Avg Loss: 0.072978 | LR: 1.19e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  10%|█         | 155/1500 [09:26<1:15:20,  3.36s/it]Epoch  154 | Step   2157/ 21000 | Loss: 0.104989 | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  154 | Avg Loss: 0.123019 | LR: 1.20e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  10%|█         | 156/1500 [09:30<1:15:30,  3.37s/it]Epoch  155 | Step   2171/ 21000 | Loss: 0.128032 | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  155 | Avg Loss: 0.139193 | LR: 1.21e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  10%|█         | 157/1500 [09:33<1:16:11,  3.40s/it]Epoch  156 | Step   2185/ 21000 | Loss: 0.136658 | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  156 | Avg Loss: 0.101100 | LR: 1.22e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  11%|█         | 158/1500 [09:36<1:15:12,  3.36s/it]Epoch  157 | Step   2199/ 21000 | Loss: 0.076101 | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  157 | Avg Loss: 0.064460 | LR: 1.22e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  11%|█         | 159/1500 [09:40<1:15:15,  3.37s/it]Epoch  158 | Step   2213/ 21000 | Loss: 0.048741 | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  158 | Avg Loss: 0.048600 | LR: 1.23e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  11%|█         | 160/1500 [09:43<1:14:39,  3.34s/it]Epoch  159 | Step   2227/ 21000 | Loss: 0.042711 | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  159 | Avg Loss: 0.044990 | LR: 1.24e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  11%|█         | 161/1500 [09:46<1:14:06,  3.32s/it]Epoch  160 | Step   2241/ 21000 | Loss: 0.047004 | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  160 | Avg Loss: 0.041976 | LR: 1.24e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  11%|█         | 162/1500 [09:50<1:15:12,  3.37s/it]Epoch  161 | Step   2255/ 21000 | Loss: 0.046487 | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  161 | Avg Loss: 0.054036 | LR: 1.25e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  11%|█         | 163/1500 [09:53<1:16:14,  3.42s/it]Epoch  162 | Step   2269/ 21000 | Loss: 0.058590 | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  162 | Avg Loss: 0.059503 | LR: 1.26e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  11%|█         | 164/1500 [09:57<1:14:52,  3.36s/it]Epoch  163 | Step   2283/ 21000 | Loss: 0.054144 | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  163 | Avg Loss: 0.046268 | LR: 1.26e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  11%|█         | 165/1500 [10:00<1:14:05,  3.33s/it]Epoch  164 | Step   2297/ 21000 | Loss: 0.044274 | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  164 | Avg Loss: 0.044183 | LR: 1.27e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  11%|█         | 166/1500 [10:03<1:13:54,  3.32s/it]Epoch  165 | Step   2311/ 21000 | Loss: 0.057007 | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  165 | Avg Loss: 0.052928 | LR: 1.28e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  11%|█         | 167/1500 [10:07<1:14:52,  3.37s/it]Epoch  166 | Step   2325/ 21000 | Loss: 0.046256 | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 1.4h
Epoch  166 | Avg Loss: 0.045207 | LR: 1.28e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  11%|█         | 168/1500 [10:10<1:14:07,  3.34s/it]Epoch  167 | Step   2339/ 21000 | Loss: 0.042863 | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  167 | Avg Loss: 0.042001 | LR: 1.29e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  11%|█▏        | 169/1500 [10:13<1:14:26,  3.36s/it]Epoch  168 | Step   2353/ 21000 | Loss: 0.041435 | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  168 | Avg Loss: 0.043884 | LR: 1.30e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  11%|█▏        | 170/1500 [10:17<1:15:11,  3.39s/it]Epoch  169 | Step   2367/ 21000 | Loss: 0.040525 | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  169 | Avg Loss: 0.037584 | LR: 1.30e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  11%|█▏        | 171/1500 [10:20<1:14:37,  3.37s/it]Epoch  170 | Step   2381/ 21000 | Loss: 0.045585 | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  170 | Avg Loss: 0.038397 | LR: 1.31e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  11%|█▏        | 172/1500 [10:23<1:14:34,  3.37s/it]Epoch  171 | Step   2395/ 21000 | Loss: 0.044551 | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  171 | Avg Loss: 0.039978 | LR: 1.32e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  12%|█▏        | 173/1500 [10:27<1:19:16,  3.58s/it]Epoch  172 | Step   2409/ 21000 | Loss: 0.056261 | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  172 | Avg Loss: 0.054898 | LR: 1.32e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  12%|█▏        | 174/1500 [10:32<1:22:53,  3.75s/it]Epoch  173 | Step   2423/ 21000 | Loss: 0.055114 | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  173 | Avg Loss: 0.053797 | LR: 1.33e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  12%|█▏        | 175/1500 [10:36<1:25:43,  3.88s/it]Epoch  174 | Step   2437/ 21000 | Loss: 0.052196 | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  174 | Avg Loss: 0.051997 | LR: 1.34e-03 | Time: 4.2s | Samples: 6,983

Training Flow Model:  12%|█▏        | 176/1500 [10:40<1:27:04,  3.95s/it]Epoch  175 | Step   2451/ 21000 | Loss: 0.043481 | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  175 | Avg Loss: 0.041886 | LR: 1.34e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  12%|█▏        | 177/1500 [10:44<1:28:34,  4.02s/it]Epoch  176 | Step   2465/ 21000 | Loss: 0.042922 | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  176 | Avg Loss: 0.046460 | LR: 1.35e-03 | Time: 4.2s | Samples: 6,983

Training Flow Model:  12%|█▏        | 178/1500 [10:48<1:28:13,  4.00s/it]Epoch  177 | Step   2479/ 21000 | Loss: 0.058198 | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  177 | Avg Loss: 0.047011 | LR: 1.36e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  12%|█▏        | 179/1500 [10:52<1:28:48,  4.03s/it]Epoch  178 | Step   2493/ 21000 | Loss: 0.071548 | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  178 | Avg Loss: 0.051371 | LR: 1.36e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  12%|█▏        | 180/1500 [10:56<1:28:53,  4.04s/it]Epoch  179 | Step   2507/ 21000 | Loss: 0.050783 | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  179 | Avg Loss: 0.051891 | LR: 1.37e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  12%|█▏        | 181/1500 [11:01<1:30:35,  4.12s/it]Epoch  180 | Step   2521/ 21000 | Loss: 0.048897 | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  180 | Avg Loss: 0.045531 | LR: 1.38e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  12%|█▏        | 182/1500 [11:05<1:29:45,  4.09s/it]Epoch  181 | Step   2535/ 21000 | Loss: 0.034484 | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  181 | Avg Loss: 0.072573 | LR: 1.38e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  12%|█▏        | 183/1500 [11:09<1:30:32,  4.13s/it]Epoch  182 | Step   2549/ 21000 | Loss: 0.119103 | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  182 | Avg Loss: 0.162198 | LR: 1.39e-03 | Time: 4.2s | Samples: 6,983

Training Flow Model:  12%|█▏        | 184/1500 [11:13<1:29:48,  4.09s/it]Epoch  183 | Step   2563/ 21000 | Loss: 0.277480 | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  183 | Avg Loss: 0.859323 | LR: 1.40e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  12%|█▏        | 185/1500 [11:17<1:29:41,  4.09s/it]Epoch  184 | Step   2577/ 21000 | Loss: 1.219188 | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  184 | Avg Loss: 1.177793 | LR: 1.40e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  12%|█▏        | 186/1500 [11:21<1:29:13,  4.07s/it]Epoch  185 | Step   2591/ 21000 | Loss: 1.151801 | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  185 | Avg Loss: 1.152081 | LR: 1.41e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  12%|█▏        | 187/1500 [11:25<1:27:55,  4.02s/it]Epoch  186 | Step   2605/ 21000 | Loss: 1.148472 | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  186 | Avg Loss: 1.151613 | LR: 1.42e-03 | Time: 3.9s | Samples: 6,983

Training Flow Model:  13%|█▎        | 188/1500 [11:29<1:30:20,  4.13s/it]Epoch  187 | Step   2619/ 21000 | Loss: 1.150249 | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  187 | Avg Loss: 1.151310 | LR: 1.42e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  13%|█▎        | 189/1500 [11:33<1:29:18,  4.09s/it]Epoch  188 | Step   2633/ 21000 | Loss: 1.151358 | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  188 | Avg Loss: 1.151897 | LR: 1.43e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  13%|█▎        | 190/1500 [11:37<1:28:50,  4.07s/it]Epoch  189 | Step   2647/ 21000 | Loss: 1.152040 | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  189 | Avg Loss: 1.152013 | LR: 1.44e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  13%|█▎        | 191/1500 [11:41<1:27:32,  4.01s/it]Epoch  190 | Step   2661/ 21000 | Loss: 1.151124 | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  190 | Avg Loss: 1.152279 | LR: 1.44e-03 | Time: 3.9s | Samples: 6,983

Training Flow Model:  13%|█▎        | 192/1500 [11:45<1:27:50,  4.03s/it]Epoch  191 | Step   2675/ 21000 | Loss: 1.152866 | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  191 | Avg Loss: 1.151694 | LR: 1.45e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  13%|█▎        | 193/1500 [11:49<1:27:23,  4.01s/it]Epoch  192 | Step   2689/ 21000 | Loss: 1.154455 | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  192 | Avg Loss: 1.152716 | LR: 1.46e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  13%|█▎        | 194/1500 [11:53<1:27:32,  4.02s/it]Epoch  193 | Step   2703/ 21000 | Loss: 1.152693 | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  193 | Avg Loss: 1.152033 | LR: 1.46e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  13%|█▎        | 195/1500 [11:57<1:28:06,  4.05s/it]Epoch  194 | Step   2717/ 21000 | Loss: 1.152691 | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  194 | Avg Loss: 1.152434 | LR: 1.47e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  13%|█▎        | 196/1500 [12:01<1:27:22,  4.02s/it]Epoch  195 | Step   2731/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  195 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.9s | Samples: 6,983

Training Flow Model:  13%|█▎        | 197/1500 [12:05<1:28:55,  4.10s/it]Epoch  196 | Step   2745/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  196 | Avg Loss: nan | LR: 1.48e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  13%|█▎        | 198/1500 [12:10<1:28:32,  4.08s/it]Epoch  197 | Step   2759/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  197 | Avg Loss: nan | LR: 1.49e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  13%|█▎        | 199/1500 [12:14<1:28:06,  4.06s/it]Epoch  198 | Step   2773/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  198 | Avg Loss: nan | LR: 1.50e-03 | Time: 4.0s | Samples: 6,983

Training Flow Model:  13%|█▎        | 200/1500 [12:18<1:28:09,  4.07s/it]Epoch  199 | Step   2787/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  199 | Avg Loss: nan | LR: 1.50e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  13%|█▎        | 201/1500 [12:21<1:23:52,  3.87s/it]Epoch  200 | Step   2801/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  200 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  13%|█▎        | 202/1500 [12:25<1:21:15,  3.76s/it]Epoch  201 | Step   2815/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  201 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  14%|█▎        | 203/1500 [12:28<1:19:18,  3.67s/it]Epoch  202 | Step   2829/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  202 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  14%|█▎        | 204/1500 [12:31<1:17:54,  3.61s/it]Epoch  203 | Step   2843/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  203 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  14%|█▎        | 205/1500 [12:35<1:15:48,  3.51s/it]Epoch  204 | Step   2857/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  204 | Avg Loss: nan | LR: 1.54e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  14%|█▎        | 206/1500 [12:39<1:20:30,  3.73s/it]Epoch  205 | Step   2871/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  205 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.2s | Samples: 6,983

Training Flow Model:  14%|█▍        | 207/1500 [12:43<1:24:25,  3.92s/it]Epoch  206 | Step   2885/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  206 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  14%|█▍        | 208/1500 [12:48<1:26:55,  4.04s/it]Epoch  207 | Step   2899/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  207 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  14%|█▍        | 209/1500 [12:52<1:29:08,  4.14s/it]Epoch  208 | Step   2913/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  208 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  14%|█▍        | 210/1500 [12:56<1:29:38,  4.17s/it]Epoch  209 | Step   2927/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  209 | Avg Loss: nan | LR: 1.57e-03 | Time: 4.2s | Samples: 6,983

Training Flow Model:  14%|█▍        | 211/1500 [13:01<1:30:55,  4.23s/it]Epoch  210 | Step   2941/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  210 | Avg Loss: nan | LR: 1.58e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  14%|█▍        | 212/1500 [13:05<1:32:27,  4.31s/it]Epoch  211 | Step   2955/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  211 | Avg Loss: nan | LR: 1.58e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  14%|█▍        | 213/1500 [13:10<1:32:42,  4.32s/it]Epoch  212 | Step   2969/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  212 | Avg Loss: nan | LR: 1.59e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  14%|█▍        | 214/1500 [13:14<1:33:04,  4.34s/it]Epoch  213 | Step   2983/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  213 | Avg Loss: nan | LR: 1.60e-03 | Time: 4.4s | Samples: 6,983
/home/edwardsun/miniconda3/envs/flow/lib/python3.9/site-packages/torch/optim/lr_scheduler.py:240: UserWarning: The epoch parameter in `scheduler.step()` was not necessary and is being deprecated where possible. Please use `scheduler.step()` to step the scheduler. During the deprecation, if epoch is different from None, the closed form is used instead of the new chainable form, where available. Please open an issue if you are unable to replicate your use case: https://github.com/pytorch/pytorch/issues/new/choose.
  warnings.warn(EPOCH_DEPRECATION_WARNING, UserWarning)

Training Flow Model:  14%|█▍        | 215/1500 [13:18<1:32:08,  4.30s/it]Epoch  214 | Step   2997/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  214 | Avg Loss: nan | LR: 1.60e-03 | Time: 4.2s | Samples: 6,983

Training Flow Model:  14%|█▍        | 216/1500 [13:22<1:31:31,  4.28s/it]Epoch  215 | Step   3011/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  215 | Avg Loss: nan | LR: 1.60e-03 | Time: 4.2s | Samples: 6,983

Training Flow Model:  14%|█▍        | 217/1500 [13:27<1:31:12,  4.27s/it]Epoch  216 | Step   3025/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  216 | Avg Loss: nan | LR: 1.60e-03 | Time: 4.2s | Samples: 6,983

Training Flow Model:  15%|█▍        | 218/1500 [13:31<1:30:35,  4.24s/it]Epoch  217 | Step   3039/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  217 | Avg Loss: nan | LR: 1.60e-03 | Time: 4.2s | Samples: 6,983

Training Flow Model:  15%|█▍        | 219/1500 [13:35<1:30:45,  4.25s/it]Epoch  218 | Step   3053/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  218 | Avg Loss: nan | LR: 1.60e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  15%|█▍        | 220/1500 [13:39<1:31:51,  4.31s/it]Epoch  219 | Step   3067/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  219 | Avg Loss: nan | LR: 1.60e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  15%|█▍        | 221/1500 [13:44<1:31:51,  4.31s/it]Epoch  220 | Step   3081/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.7 steps/s | ETA: 1.3h
Epoch  220 | Avg Loss: nan | LR: 1.60e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  15%|█▍        | 222/1500 [13:47<1:25:09,  4.00s/it]Epoch  221 | Step   3095/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.7 steps/s | ETA: 1.3h
Epoch  221 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  15%|█▍        | 223/1500 [13:50<1:19:59,  3.76s/it]Epoch  222 | Step   3109/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.7 steps/s | ETA: 1.3h
Epoch  222 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  15%|█▍        | 224/1500 [13:53<1:16:06,  3.58s/it]Epoch  223 | Step   3123/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  223 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  15%|█▌        | 225/1500 [13:57<1:15:24,  3.55s/it]Epoch  224 | Step   3137/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  224 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  15%|█▌        | 226/1500 [14:00<1:14:16,  3.50s/it]Epoch  225 | Step   3151/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  225 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  15%|█▌        | 227/1500 [14:04<1:13:14,  3.45s/it]Epoch  226 | Step   3165/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  226 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  15%|█▌        | 228/1500 [14:07<1:11:49,  3.39s/it]Epoch  227 | Step   3179/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  227 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  15%|█▌        | 229/1500 [14:10<1:11:02,  3.35s/it]Epoch  228 | Step   3193/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  228 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  15%|█▌        | 230/1500 [14:13<1:10:05,  3.31s/it]Epoch  229 | Step   3207/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  229 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  15%|█▌        | 231/1500 [14:17<1:09:49,  3.30s/it]Epoch  230 | Step   3221/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  230 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  15%|█▌        | 232/1500 [14:20<1:09:41,  3.30s/it]Epoch  231 | Step   3235/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  231 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  16%|█▌        | 233/1500 [14:23<1:10:26,  3.34s/it]Epoch  232 | Step   3249/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  232 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  16%|█▌        | 234/1500 [14:27<1:11:28,  3.39s/it]Epoch  233 | Step   3263/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  233 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  16%|█▌        | 235/1500 [14:30<1:11:04,  3.37s/it]Epoch  234 | Step   3277/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  234 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  16%|█▌        | 236/1500 [14:33<1:09:53,  3.32s/it]Epoch  235 | Step   3291/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  235 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  16%|█▌        | 237/1500 [14:37<1:08:54,  3.27s/it]Epoch  236 | Step   3305/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  236 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  16%|█▌        | 238/1500 [14:40<1:08:58,  3.28s/it]Epoch  237 | Step   3319/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  237 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  16%|█▌        | 239/1500 [14:43<1:08:32,  3.26s/it]Epoch  238 | Step   3333/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  238 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  16%|█▌        | 240/1500 [14:46<1:08:31,  3.26s/it]Epoch  239 | Step   3347/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  239 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  16%|█▌        | 241/1500 [14:50<1:08:47,  3.28s/it]Epoch  240 | Step   3361/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  240 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  16%|█▌        | 242/1500 [14:53<1:09:10,  3.30s/it]Epoch  241 | Step   3375/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  241 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  16%|█▌        | 243/1500 [14:56<1:09:42,  3.33s/it]Epoch  242 | Step   3389/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  242 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  16%|█▋        | 244/1500 [15:00<1:09:08,  3.30s/it]Epoch  243 | Step   3403/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  243 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  16%|█▋        | 245/1500 [15:03<1:09:57,  3.34s/it]Epoch  244 | Step   3417/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  244 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  16%|█▋        | 246/1500 [15:06<1:10:04,  3.35s/it]Epoch  245 | Step   3431/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  245 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  16%|█▋        | 247/1500 [15:10<1:10:18,  3.37s/it]Epoch  246 | Step   3445/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  246 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  17%|█▋        | 248/1500 [15:13<1:09:19,  3.32s/it]Epoch  247 | Step   3459/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  247 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  17%|█▋        | 249/1500 [15:16<1:08:55,  3.31s/it]Epoch  248 | Step   3473/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  248 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  17%|█▋        | 250/1500 [15:20<1:08:39,  3.30s/it]Epoch  249 | Step   3487/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  249 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  17%|█▋        | 251/1500 [15:23<1:08:18,  3.28s/it]Epoch  250 | Step   3501/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  250 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  17%|█▋        | 252/1500 [15:26<1:07:24,  3.24s/it]Epoch  251 | Step   3515/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  251 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.1s | Samples: 6,983

Training Flow Model:  17%|█▋        | 253/1500 [15:29<1:07:45,  3.26s/it]Epoch  252 | Step   3529/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  252 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  17%|█▋        | 254/1500 [15:33<1:07:40,  3.26s/it]Epoch  253 | Step   3543/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  253 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  17%|█▋        | 255/1500 [15:36<1:07:59,  3.28s/it]Epoch  254 | Step   3557/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  254 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  17%|█▋        | 256/1500 [15:39<1:07:49,  3.27s/it]Epoch  255 | Step   3571/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  255 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  17%|█▋        | 257/1500 [15:42<1:07:51,  3.28s/it]Epoch  256 | Step   3585/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  256 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  17%|█▋        | 258/1500 [15:46<1:07:23,  3.26s/it]Epoch  257 | Step   3599/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  257 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  17%|█▋        | 259/1500 [15:49<1:07:27,  3.26s/it]Epoch  258 | Step   3613/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  258 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  17%|█▋        | 260/1500 [15:52<1:06:55,  3.24s/it]Epoch  259 | Step   3627/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  259 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  17%|█▋        | 261/1500 [15:55<1:07:35,  3.27s/it]Epoch  260 | Step   3641/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  260 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  17%|█▋        | 262/1500 [15:59<1:08:40,  3.33s/it]Epoch  261 | Step   3655/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  261 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  18%|█▊        | 263/1500 [16:02<1:07:24,  3.27s/it]Epoch  262 | Step   3669/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  262 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.1s | Samples: 6,983

Training Flow Model:  18%|█▊        | 264/1500 [16:05<1:06:51,  3.25s/it]Epoch  263 | Step   3683/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  263 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  18%|█▊        | 265/1500 [16:09<1:07:08,  3.26s/it]Epoch  264 | Step   3697/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  264 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  18%|█▊        | 266/1500 [16:12<1:06:59,  3.26s/it]Epoch  265 | Step   3711/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  265 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  18%|█▊        | 267/1500 [16:15<1:07:23,  3.28s/it]Epoch  266 | Step   3725/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  266 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  18%|█▊        | 268/1500 [16:18<1:06:25,  3.24s/it]Epoch  267 | Step   3739/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  267 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.1s | Samples: 6,983

Training Flow Model:  18%|█▊        | 269/1500 [16:22<1:07:28,  3.29s/it]Epoch  268 | Step   3753/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  268 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  18%|█▊        | 270/1500 [16:25<1:06:35,  3.25s/it]Epoch  269 | Step   3767/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.3h
Epoch  269 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  18%|█▊        | 271/1500 [16:28<1:06:16,  3.24s/it]Epoch  270 | Step   3781/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  270 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  18%|█▊        | 272/1500 [16:31<1:06:13,  3.24s/it]Epoch  271 | Step   3795/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  271 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  18%|█▊        | 273/1500 [16:35<1:06:40,  3.26s/it]Epoch  272 | Step   3809/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  272 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  18%|█▊        | 274/1500 [16:38<1:06:30,  3.25s/it]Epoch  273 | Step   3823/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  273 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  18%|█▊        | 275/1500 [16:41<1:06:35,  3.26s/it]Epoch  274 | Step   3837/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  274 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  18%|█▊        | 276/1500 [16:44<1:06:20,  3.25s/it]Epoch  275 | Step   3851/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  275 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  18%|█▊        | 277/1500 [16:48<1:06:22,  3.26s/it]Epoch  276 | Step   3865/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  276 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  19%|█▊        | 278/1500 [16:51<1:06:54,  3.29s/it]Epoch  277 | Step   3879/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  277 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  19%|█▊        | 279/1500 [16:54<1:06:53,  3.29s/it]Epoch  278 | Step   3893/ 21000 | Loss: nan | LR: 1.60e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  278 | Avg Loss: nan | LR: 1.60e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  19%|█▊        | 280/1500 [16:57<1:06:32,  3.27s/it]Epoch  279 | Step   3907/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  279 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  19%|█▊        | 281/1500 [17:01<1:06:54,  3.29s/it]Epoch  280 | Step   3921/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  280 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  19%|█▉        | 282/1500 [17:04<1:07:29,  3.32s/it]Epoch  281 | Step   3935/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  281 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  19%|█▉        | 283/1500 [17:08<1:07:32,  3.33s/it]Epoch  282 | Step   3949/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  282 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  19%|█▉        | 284/1500 [17:11<1:07:17,  3.32s/it]Epoch  283 | Step   3963/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  283 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  19%|█▉        | 285/1500 [17:14<1:06:40,  3.29s/it]Epoch  284 | Step   3977/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.8 steps/s | ETA: 1.2h
Epoch  284 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  19%|█▉        | 286/1500 [17:17<1:07:13,  3.32s/it]Epoch  285 | Step   3991/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  285 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  19%|█▉        | 287/1500 [17:21<1:06:45,  3.30s/it]Epoch  286 | Step   4005/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  286 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  19%|█▉        | 288/1500 [17:24<1:06:42,  3.30s/it]Epoch  287 | Step   4019/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  287 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  19%|█▉        | 289/1500 [17:27<1:06:58,  3.32s/it]Epoch  288 | Step   4033/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  288 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  19%|█▉        | 290/1500 [17:31<1:07:18,  3.34s/it]Epoch  289 | Step   4047/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  289 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  19%|█▉        | 291/1500 [17:34<1:06:24,  3.30s/it]Epoch  290 | Step   4061/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  290 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  19%|█▉        | 292/1500 [17:37<1:07:10,  3.34s/it]Epoch  291 | Step   4075/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  291 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  20%|█▉        | 293/1500 [17:41<1:06:51,  3.32s/it]Epoch  292 | Step   4089/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  292 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  20%|█▉        | 294/1500 [17:44<1:06:02,  3.29s/it]Epoch  293 | Step   4103/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  293 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  20%|█▉        | 295/1500 [17:47<1:06:31,  3.31s/it]Epoch  294 | Step   4117/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  294 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  20%|█▉        | 296/1500 [17:51<1:06:56,  3.34s/it]Epoch  295 | Step   4131/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  295 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  20%|█▉        | 297/1500 [17:54<1:06:52,  3.34s/it]Epoch  296 | Step   4145/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  296 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  20%|█▉        | 298/1500 [17:57<1:06:53,  3.34s/it]Epoch  297 | Step   4159/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  297 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  20%|█▉        | 299/1500 [18:01<1:06:52,  3.34s/it]Epoch  298 | Step   4173/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  298 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  20%|██        | 300/1500 [18:05<1:12:12,  3.61s/it]Epoch  299 | Step   4187/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  299 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.5s | Samples: 6,983
✓ Checkpoint saved: /data2/edwardsun/flow_checkpoints/amp_flow_model_final_optimized.pth (loss: nan, step: 4200)

Training Flow Model:  20%|██        | 301/1500 [18:08<1:09:48,  3.49s/it]Epoch  300 | Step   4201/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  300 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  20%|██        | 302/1500 [18:11<1:08:08,  3.41s/it]Epoch  301 | Step   4215/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  301 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  20%|██        | 303/1500 [18:14<1:06:38,  3.34s/it]Epoch  302 | Step   4229/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  302 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  20%|██        | 304/1500 [18:18<1:06:10,  3.32s/it]Epoch  303 | Step   4243/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  303 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  20%|██        | 305/1500 [18:21<1:05:16,  3.28s/it]Epoch  304 | Step   4257/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  304 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  20%|██        | 306/1500 [18:24<1:05:16,  3.28s/it]Epoch  305 | Step   4271/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  305 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  20%|██        | 307/1500 [18:27<1:04:37,  3.25s/it]Epoch  306 | Step   4285/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  306 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  21%|██        | 308/1500 [18:31<1:05:45,  3.31s/it]Epoch  307 | Step   4299/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  307 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  21%|██        | 309/1500 [18:34<1:05:12,  3.28s/it]Epoch  308 | Step   4313/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  308 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  21%|██        | 310/1500 [18:37<1:04:50,  3.27s/it]Epoch  309 | Step   4327/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  309 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  21%|██        | 311/1500 [18:41<1:05:19,  3.30s/it]Epoch  310 | Step   4341/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  310 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  21%|██        | 312/1500 [18:44<1:04:41,  3.27s/it]Epoch  311 | Step   4355/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  311 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  21%|██        | 313/1500 [18:47<1:04:17,  3.25s/it]Epoch  312 | Step   4369/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  312 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  21%|██        | 314/1500 [18:50<1:04:30,  3.26s/it]Epoch  313 | Step   4383/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  313 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  21%|██        | 315/1500 [18:54<1:04:30,  3.27s/it]Epoch  314 | Step   4397/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  314 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  21%|██        | 316/1500 [18:57<1:04:35,  3.27s/it]Epoch  315 | Step   4411/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  315 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  21%|██        | 317/1500 [19:00<1:04:53,  3.29s/it]Epoch  316 | Step   4425/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  316 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  21%|██        | 318/1500 [19:04<1:04:38,  3.28s/it]Epoch  317 | Step   4439/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  317 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  21%|██▏       | 319/1500 [19:07<1:04:35,  3.28s/it]Epoch  318 | Step   4453/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  318 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  21%|██▏       | 320/1500 [19:10<1:05:18,  3.32s/it]Epoch  319 | Step   4467/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  319 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  21%|██▏       | 321/1500 [19:13<1:04:40,  3.29s/it]Epoch  320 | Step   4481/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  320 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  21%|██▏       | 322/1500 [19:17<1:04:30,  3.29s/it]Epoch  321 | Step   4495/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  321 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 323/1500 [19:20<1:04:12,  3.27s/it]Epoch  322 | Step   4509/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  322 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  22%|██▏       | 324/1500 [19:23<1:04:23,  3.29s/it]Epoch  323 | Step   4523/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  323 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 325/1500 [19:27<1:05:06,  3.32s/it]Epoch  324 | Step   4537/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  324 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  22%|██▏       | 326/1500 [19:30<1:04:59,  3.32s/it]Epoch  325 | Step   4551/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  325 | Avg Loss: nan | LR: 1.59e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 327/1500 [19:33<1:04:33,  3.30s/it]Epoch  326 | Step   4565/ 21000 | Loss: nan | LR: 1.59e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  326 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 328/1500 [19:37<1:04:16,  3.29s/it]Epoch  327 | Step   4579/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  327 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 329/1500 [19:40<1:04:08,  3.29s/it]Epoch  328 | Step   4593/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  328 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 330/1500 [19:43<1:04:19,  3.30s/it]Epoch  329 | Step   4607/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  329 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 331/1500 [19:46<1:04:01,  3.29s/it]Epoch  330 | Step   4621/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  330 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 332/1500 [19:50<1:03:07,  3.24s/it]Epoch  331 | Step   4635/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  331 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.1s | Samples: 6,983

Training Flow Model:  22%|██▏       | 333/1500 [19:53<1:02:55,  3.24s/it]Epoch  332 | Step   4649/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  332 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  22%|██▏       | 334/1500 [19:56<1:03:32,  3.27s/it]Epoch  333 | Step   4663/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  333 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 335/1500 [19:59<1:03:29,  3.27s/it]Epoch  334 | Step   4677/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  334 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  22%|██▏       | 336/1500 [20:03<1:04:37,  3.33s/it]Epoch  335 | Step   4691/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  335 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  22%|██▏       | 337/1500 [20:06<1:04:13,  3.31s/it]Epoch  336 | Step   4705/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  336 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 338/1500 [20:09<1:04:13,  3.32s/it]Epoch  337 | Step   4719/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  337 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 339/1500 [20:13<1:04:10,  3.32s/it]Epoch  338 | Step   4733/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  338 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 340/1500 [20:16<1:04:12,  3.32s/it]Epoch  339 | Step   4747/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  339 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 341/1500 [20:19<1:03:41,  3.30s/it]Epoch  340 | Step   4761/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  340 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  23%|██▎       | 342/1500 [20:23<1:03:28,  3.29s/it]Epoch  341 | Step   4775/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  341 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 343/1500 [20:26<1:03:35,  3.30s/it]Epoch  342 | Step   4789/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  342 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 344/1500 [20:29<1:03:27,  3.29s/it]Epoch  343 | Step   4803/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.2h
Epoch  343 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 345/1500 [20:32<1:03:05,  3.28s/it]Epoch  344 | Step   4817/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  344 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  23%|██▎       | 346/1500 [20:36<1:02:46,  3.26s/it]Epoch  345 | Step   4831/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  345 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  23%|██▎       | 347/1500 [20:39<1:02:51,  3.27s/it]Epoch  346 | Step   4845/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  346 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 348/1500 [20:42<1:02:47,  3.27s/it]Epoch  347 | Step   4859/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  347 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 349/1500 [20:45<1:02:23,  3.25s/it]Epoch  348 | Step   4873/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  348 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  23%|██▎       | 350/1500 [20:49<1:02:23,  3.25s/it]Epoch  349 | Step   4887/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  349 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  23%|██▎       | 351/1500 [20:52<1:03:03,  3.29s/it]Epoch  350 | Step   4901/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  350 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  23%|██▎       | 352/1500 [20:55<1:03:01,  3.29s/it]Epoch  351 | Step   4915/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  351 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  24%|██▎       | 353/1500 [20:59<1:02:28,  3.27s/it]Epoch  352 | Step   4929/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  352 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  24%|██▎       | 354/1500 [21:02<1:02:57,  3.30s/it]Epoch  353 | Step   4943/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  353 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  24%|██▎       | 355/1500 [21:05<1:03:38,  3.33s/it]Epoch  354 | Step   4957/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  354 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  24%|██▎       | 356/1500 [21:09<1:02:43,  3.29s/it]Epoch  355 | Step   4971/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  355 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.2s | Samples: 6,983

Training Flow Model:  24%|██▍       | 357/1500 [21:12<1:03:37,  3.34s/it]Epoch  356 | Step   4985/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  356 | Avg Loss: nan | LR: 1.58e-03 | Time: 3.5s | Samples: 6,983
/data2/edwardsun/flow_home/cfg_dataset.py:360: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  'index': torch.tensor(idx, dtype=torch.long)
/data2/edwardsun/flow_home/cfg_dataset.py:360: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  'index': torch.tensor(idx, dtype=torch.long)
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):

Training Flow Model:  24%|██▍       | 358/1500 [21:23<1:47:36,  5.65s/it]Epoch  357 | Step   4999/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Validation at step 5000: Loss = nan
Epoch  357 | Avg Loss: nan | LR: 1.58e-03 | Time: 11.1s | Samples: 6,983

Training Flow Model:  24%|██▍       | 359/1500 [21:28<1:41:54,  5.36s/it]Epoch  358 | Step   5013/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  358 | Avg Loss: nan | LR: 1.58e-03 | Time: 4.7s | Samples: 6,983
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):

Training Flow Model:  24%|██▍       | 360/1500 [21:31<1:31:23,  4.81s/it]Epoch  359 | Step   5027/ 21000 | Loss: nan | LR: 1.58e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  359 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  24%|██▍       | 361/1500 [21:35<1:24:26,  4.45s/it]Epoch  360 | Step   5041/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  360 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  24%|██▍       | 362/1500 [21:38<1:18:56,  4.16s/it]Epoch  361 | Step   5055/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  361 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  24%|██▍       | 363/1500 [21:42<1:15:13,  3.97s/it]Epoch  362 | Step   5069/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  362 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  24%|██▍       | 364/1500 [21:46<1:13:18,  3.87s/it]Epoch  363 | Step   5083/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  363 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  24%|██▍       | 365/1500 [21:49<1:11:53,  3.80s/it]Epoch  364 | Step   5097/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  364 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  24%|██▍       | 366/1500 [21:53<1:09:30,  3.68s/it]Epoch  365 | Step   5111/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  365 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  24%|██▍       | 367/1500 [21:56<1:08:54,  3.65s/it]Epoch  366 | Step   5125/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  366 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  25%|██▍       | 368/1500 [22:00<1:08:09,  3.61s/it]Epoch  367 | Step   5139/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  367 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  25%|██▍       | 369/1500 [22:03<1:07:28,  3.58s/it]Epoch  368 | Step   5153/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  368 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  25%|██▍       | 370/1500 [22:07<1:06:21,  3.52s/it]Epoch  369 | Step   5167/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  369 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  25%|██▍       | 371/1500 [22:10<1:05:47,  3.50s/it]Epoch  370 | Step   5181/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  370 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  25%|██▍       | 372/1500 [22:14<1:05:51,  3.50s/it]Epoch  371 | Step   5195/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  371 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  25%|██▍       | 373/1500 [22:17<1:06:52,  3.56s/it]Epoch  372 | Step   5209/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  372 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  25%|██▍       | 374/1500 [22:21<1:06:45,  3.56s/it]Epoch  373 | Step   5223/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  373 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  25%|██▌       | 375/1500 [22:24<1:05:47,  3.51s/it]Epoch  374 | Step   5237/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  374 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  25%|██▌       | 376/1500 [22:28<1:05:40,  3.51s/it]Epoch  375 | Step   5251/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  375 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  25%|██▌       | 377/1500 [22:31<1:05:45,  3.51s/it]Epoch  376 | Step   5265/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  376 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  25%|██▌       | 378/1500 [22:35<1:05:47,  3.52s/it]Epoch  377 | Step   5279/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  377 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  25%|██▌       | 379/1500 [22:38<1:05:39,  3.51s/it]Epoch  378 | Step   5293/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  378 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  25%|██▌       | 380/1500 [22:42<1:05:59,  3.54s/it]Epoch  379 | Step   5307/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  379 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  25%|██▌       | 381/1500 [22:45<1:05:11,  3.50s/it]Epoch  380 | Step   5321/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  380 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  25%|██▌       | 382/1500 [22:49<1:05:44,  3.53s/it]Epoch  381 | Step   5335/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  381 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  26%|██▌       | 383/1500 [22:52<1:05:15,  3.51s/it]Epoch  382 | Step   5349/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  382 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  26%|██▌       | 384/1500 [22:56<1:05:31,  3.52s/it]Epoch  383 | Step   5363/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  383 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  26%|██▌       | 385/1500 [22:59<1:05:54,  3.55s/it]Epoch  384 | Step   5377/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  384 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  26%|██▌       | 386/1500 [23:03<1:05:27,  3.53s/it]Epoch  385 | Step   5391/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  385 | Avg Loss: nan | LR: 1.57e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  26%|██▌       | 387/1500 [23:07<1:05:58,  3.56s/it]Epoch  386 | Step   5405/ 21000 | Loss: nan | LR: 1.57e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  386 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  26%|██▌       | 388/1500 [23:10<1:05:05,  3.51s/it]Epoch  387 | Step   5419/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  387 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  26%|██▌       | 389/1500 [23:14<1:05:30,  3.54s/it]Epoch  388 | Step   5433/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  388 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  26%|██▌       | 390/1500 [23:17<1:05:29,  3.54s/it]Epoch  389 | Step   5447/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  389 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  26%|██▌       | 391/1500 [23:21<1:04:59,  3.52s/it]Epoch  390 | Step   5461/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  390 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  26%|██▌       | 392/1500 [23:24<1:05:49,  3.56s/it]Epoch  391 | Step   5475/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  391 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  26%|██▌       | 393/1500 [23:28<1:04:30,  3.50s/it]Epoch  392 | Step   5489/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  392 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  26%|██▋       | 394/1500 [23:31<1:04:20,  3.49s/it]Epoch  393 | Step   5503/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  393 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  26%|██▋       | 395/1500 [23:35<1:05:23,  3.55s/it]Epoch  394 | Step   5517/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  394 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  26%|██▋       | 396/1500 [23:38<1:04:55,  3.53s/it]Epoch  395 | Step   5531/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  395 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  26%|██▋       | 397/1500 [23:42<1:03:54,  3.48s/it]Epoch  396 | Step   5545/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  396 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  27%|██▋       | 398/1500 [23:45<1:03:39,  3.47s/it]Epoch  397 | Step   5559/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  397 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  27%|██▋       | 399/1500 [23:49<1:04:17,  3.50s/it]Epoch  398 | Step   5573/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  398 | Avg Loss: nan | LR: 1.56e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  27%|██▋       | 400/1500 [23:53<1:09:11,  3.77s/it]Epoch  399 | Step   5587/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  399 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  27%|██▋       | 401/1500 [23:57<1:11:54,  3.93s/it]Epoch  400 | Step   5601/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  400 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  27%|██▋       | 402/1500 [24:02<1:14:30,  4.07s/it]Epoch  401 | Step   5615/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  401 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  27%|██▋       | 403/1500 [24:06<1:17:32,  4.24s/it]Epoch  402 | Step   5629/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  402 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  27%|██▋       | 404/1500 [24:11<1:19:07,  4.33s/it]Epoch  403 | Step   5643/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  403 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  27%|██▋       | 405/1500 [24:15<1:19:33,  4.36s/it]Epoch  404 | Step   5657/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  404 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  27%|██▋       | 406/1500 [24:20<1:19:02,  4.34s/it]Epoch  405 | Step   5671/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  405 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  27%|██▋       | 407/1500 [24:24<1:19:31,  4.37s/it]Epoch  406 | Step   5685/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  406 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  27%|██▋       | 408/1500 [24:29<1:21:16,  4.47s/it]Epoch  407 | Step   5699/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  407 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  27%|██▋       | 409/1500 [24:33<1:21:00,  4.46s/it]Epoch  408 | Step   5713/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  408 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  27%|██▋       | 410/1500 [24:38<1:20:49,  4.45s/it]Epoch  409 | Step   5727/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  409 | Avg Loss: nan | LR: 1.56e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  27%|██▋       | 411/1500 [24:42<1:20:13,  4.42s/it]Epoch  410 | Step   5741/ 21000 | Loss: nan | LR: 1.56e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  410 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  27%|██▋       | 412/1500 [24:46<1:20:40,  4.45s/it]Epoch  411 | Step   5755/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  411 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  28%|██▊       | 413/1500 [24:51<1:22:04,  4.53s/it]Epoch  412 | Step   5769/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  412 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  28%|██▊       | 414/1500 [24:56<1:21:20,  4.49s/it]Epoch  413 | Step   5783/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  413 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  28%|██▊       | 415/1500 [25:00<1:21:11,  4.49s/it]Epoch  414 | Step   5797/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  414 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  28%|██▊       | 416/1500 [25:05<1:20:52,  4.48s/it]Epoch  415 | Step   5811/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  415 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  28%|██▊       | 417/1500 [25:09<1:20:19,  4.45s/it]Epoch  416 | Step   5825/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  416 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  28%|██▊       | 418/1500 [25:14<1:21:22,  4.51s/it]Epoch  417 | Step   5839/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  417 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  28%|██▊       | 419/1500 [25:18<1:20:32,  4.47s/it]Epoch  418 | Step   5853/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  418 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  28%|██▊       | 420/1500 [25:22<1:20:12,  4.46s/it]Epoch  419 | Step   5867/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  419 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  28%|██▊       | 421/1500 [25:27<1:19:15,  4.41s/it]Epoch  420 | Step   5881/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  420 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  28%|██▊       | 422/1500 [25:31<1:18:44,  4.38s/it]Epoch  421 | Step   5895/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  421 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  28%|██▊       | 423/1500 [25:36<1:21:13,  4.52s/it]Epoch  422 | Step   5909/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  422 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.9s | Samples: 6,983

Training Flow Model:  28%|██▊       | 424/1500 [25:40<1:20:07,  4.47s/it]Epoch  423 | Step   5923/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.9 steps/s | ETA: 1.1h
Epoch  423 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  28%|██▊       | 425/1500 [25:45<1:19:32,  4.44s/it]Epoch  424 | Step   5937/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  424 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  28%|██▊       | 426/1500 [25:49<1:19:20,  4.43s/it]Epoch  425 | Step   5951/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  425 | Avg Loss: nan | LR: 1.55e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  28%|██▊       | 427/1500 [25:53<1:15:10,  4.20s/it]Epoch  426 | Step   5965/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  426 | Avg Loss: nan | LR: 1.55e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  29%|██▊       | 428/1500 [25:56<1:11:44,  4.02s/it]Epoch  427 | Step   5979/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  427 | Avg Loss: nan | LR: 1.55e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  29%|██▊       | 429/1500 [26:00<1:09:48,  3.91s/it]Epoch  428 | Step   5993/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  428 | Avg Loss: nan | LR: 1.55e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  29%|██▊       | 430/1500 [26:04<1:08:33,  3.84s/it]Epoch  429 | Step   6007/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  429 | Avg Loss: nan | LR: 1.55e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  29%|██▊       | 431/1500 [26:07<1:08:36,  3.85s/it]Epoch  430 | Step   6021/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  430 | Avg Loss: nan | LR: 1.55e-03 | Time: 3.9s | Samples: 6,983

Training Flow Model:  29%|██▉       | 432/1500 [26:12<1:13:09,  4.11s/it]Epoch  431 | Step   6035/ 21000 | Loss: nan | LR: 1.55e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  431 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  29%|██▉       | 433/1500 [26:17<1:16:17,  4.29s/it]Epoch  432 | Step   6049/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  432 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  29%|██▉       | 434/1500 [26:22<1:18:12,  4.40s/it]Epoch  433 | Step   6063/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  433 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  29%|██▉       | 435/1500 [26:26<1:18:36,  4.43s/it]Epoch  434 | Step   6077/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  434 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  29%|██▉       | 436/1500 [26:30<1:18:39,  4.44s/it]Epoch  435 | Step   6091/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  435 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  29%|██▉       | 437/1500 [26:35<1:18:55,  4.45s/it]Epoch  436 | Step   6105/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  436 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  29%|██▉       | 438/1500 [26:39<1:19:09,  4.47s/it]Epoch  437 | Step   6119/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  437 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  29%|██▉       | 439/1500 [26:44<1:19:40,  4.51s/it]Epoch  438 | Step   6133/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  438 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  29%|██▉       | 440/1500 [26:49<1:20:46,  4.57s/it]Epoch  439 | Step   6147/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  439 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  29%|██▉       | 441/1500 [26:54<1:22:05,  4.65s/it]Epoch  440 | Step   6161/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  440 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  29%|██▉       | 442/1500 [26:59<1:24:09,  4.77s/it]Epoch  441 | Step   6175/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  441 | Avg Loss: nan | LR: 1.54e-03 | Time: 5.1s | Samples: 6,983

Training Flow Model:  30%|██▉       | 443/1500 [27:03<1:23:08,  4.72s/it]Epoch  442 | Step   6189/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  442 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  30%|██▉       | 444/1500 [27:08<1:22:16,  4.67s/it]Epoch  443 | Step   6203/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  443 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  30%|██▉       | 445/1500 [27:12<1:20:44,  4.59s/it]Epoch  444 | Step   6217/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  444 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  30%|██▉       | 446/1500 [27:17<1:19:25,  4.52s/it]Epoch  445 | Step   6231/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  445 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  30%|██▉       | 447/1500 [27:21<1:19:01,  4.50s/it]Epoch  446 | Step   6245/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  446 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  30%|██▉       | 448/1500 [27:26<1:19:19,  4.52s/it]Epoch  447 | Step   6259/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  447 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  30%|██▉       | 449/1500 [27:30<1:20:49,  4.61s/it]Epoch  448 | Step   6273/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  448 | Avg Loss: nan | LR: 1.54e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  30%|███       | 450/1500 [27:34<1:14:41,  4.27s/it]Epoch  449 | Step   6287/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  449 | Avg Loss: nan | LR: 1.54e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  30%|███       | 451/1500 [27:37<1:10:43,  4.05s/it]Epoch  450 | Step   6301/ 21000 | Loss: nan | LR: 1.54e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  450 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  30%|███       | 452/1500 [27:41<1:08:11,  3.90s/it]Epoch  451 | Step   6315/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  451 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  30%|███       | 453/1500 [27:45<1:06:23,  3.80s/it]Epoch  452 | Step   6329/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  452 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  30%|███       | 454/1500 [27:48<1:04:45,  3.71s/it]Epoch  453 | Step   6343/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  453 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  30%|███       | 455/1500 [27:52<1:03:32,  3.65s/it]Epoch  454 | Step   6357/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  454 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  30%|███       | 456/1500 [27:55<1:02:47,  3.61s/it]Epoch  455 | Step   6371/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  455 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  30%|███       | 457/1500 [27:59<1:02:30,  3.60s/it]Epoch  456 | Step   6385/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  456 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  31%|███       | 458/1500 [28:02<1:02:44,  3.61s/it]Epoch  457 | Step   6399/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  457 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  31%|███       | 459/1500 [28:06<1:02:36,  3.61s/it]Epoch  458 | Step   6413/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  458 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  31%|███       | 460/1500 [28:09<1:01:45,  3.56s/it]Epoch  459 | Step   6427/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  459 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  31%|███       | 461/1500 [28:13<1:01:30,  3.55s/it]Epoch  460 | Step   6441/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  460 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  31%|███       | 462/1500 [28:16<1:01:26,  3.55s/it]Epoch  461 | Step   6455/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  461 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  31%|███       | 463/1500 [28:20<1:01:05,  3.53s/it]Epoch  462 | Step   6469/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  462 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  31%|███       | 464/1500 [28:23<1:00:24,  3.50s/it]Epoch  463 | Step   6483/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  463 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  31%|███       | 465/1500 [28:27<1:00:00,  3.48s/it]Epoch  464 | Step   6497/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  464 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  31%|███       | 466/1500 [28:30<59:39,  3.46s/it]  Epoch  465 | Step   6511/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  465 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  31%|███       | 467/1500 [28:34<1:00:00,  3.49s/it]Epoch  466 | Step   6525/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  466 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  31%|███       | 468/1500 [28:37<1:00:00,  3.49s/it]Epoch  467 | Step   6539/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  467 | Avg Loss: nan | LR: 1.53e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  31%|███▏      | 469/1500 [28:41<1:00:21,  3.51s/it]Epoch  468 | Step   6553/ 21000 | Loss: nan | LR: 1.53e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  468 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  31%|███▏      | 470/1500 [28:44<1:00:19,  3.51s/it]Epoch  469 | Step   6567/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  469 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  31%|███▏      | 471/1500 [28:48<1:00:05,  3.50s/it]Epoch  470 | Step   6581/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.1h
Epoch  470 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  31%|███▏      | 472/1500 [28:51<1:00:27,  3.53s/it]Epoch  471 | Step   6595/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  471 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  32%|███▏      | 473/1500 [28:55<1:00:55,  3.56s/it]Epoch  472 | Step   6609/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  472 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  32%|███▏      | 474/1500 [28:59<1:00:30,  3.54s/it]Epoch  473 | Step   6623/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  473 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  32%|███▏      | 475/1500 [29:02<59:48,  3.50s/it]  Epoch  474 | Step   6637/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  474 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  32%|███▏      | 476/1500 [29:05<59:26,  3.48s/it]Epoch  475 | Step   6651/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  475 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  32%|███▏      | 477/1500 [29:09<59:15,  3.48s/it]Epoch  476 | Step   6665/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  476 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  32%|███▏      | 478/1500 [29:12<59:45,  3.51s/it]Epoch  477 | Step   6679/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  477 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  32%|███▏      | 479/1500 [29:16<1:00:07,  3.53s/it]Epoch  478 | Step   6693/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  478 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  32%|███▏      | 480/1500 [29:20<1:00:19,  3.55s/it]Epoch  479 | Step   6707/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  479 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  32%|███▏      | 481/1500 [29:23<1:00:05,  3.54s/it]Epoch  480 | Step   6721/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  480 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  32%|███▏      | 482/1500 [29:27<59:34,  3.51s/it]  Epoch  481 | Step   6735/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  481 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  32%|███▏      | 483/1500 [29:30<59:02,  3.48s/it]Epoch  482 | Step   6749/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  482 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  32%|███▏      | 484/1500 [29:33<58:19,  3.44s/it]Epoch  483 | Step   6763/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  483 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  32%|███▏      | 485/1500 [29:37<58:13,  3.44s/it]Epoch  484 | Step   6777/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  484 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  32%|███▏      | 486/1500 [29:40<58:59,  3.49s/it]Epoch  485 | Step   6791/ 21000 | Loss: nan | LR: 1.52e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  485 | Avg Loss: nan | LR: 1.52e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  32%|███▏      | 487/1500 [29:44<59:23,  3.52s/it]Epoch  486 | Step   6805/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  486 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  33%|███▎      | 488/1500 [29:47<59:31,  3.53s/it]Epoch  487 | Step   6819/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  487 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  33%|███▎      | 489/1500 [29:51<59:47,  3.55s/it]Epoch  488 | Step   6833/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  488 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  33%|███▎      | 490/1500 [29:55<1:00:01,  3.57s/it]Epoch  489 | Step   6847/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  489 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  33%|███▎      | 491/1500 [29:58<59:00,  3.51s/it]  Epoch  490 | Step   6861/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  490 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  33%|███▎      | 492/1500 [30:02<59:17,  3.53s/it]Epoch  491 | Step   6875/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  491 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  33%|███▎      | 493/1500 [30:05<58:39,  3.50s/it]Epoch  492 | Step   6889/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  492 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  33%|███▎      | 494/1500 [30:09<59:00,  3.52s/it]Epoch  493 | Step   6903/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  493 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  33%|███▎      | 495/1500 [30:12<59:06,  3.53s/it]Epoch  494 | Step   6917/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  494 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  33%|███▎      | 496/1500 [30:16<59:30,  3.56s/it]Epoch  495 | Step   6931/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  495 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  33%|███▎      | 497/1500 [30:19<58:51,  3.52s/it]Epoch  496 | Step   6945/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  496 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  33%|███▎      | 498/1500 [30:23<58:52,  3.53s/it]Epoch  497 | Step   6959/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  497 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  33%|███▎      | 499/1500 [30:26<58:19,  3.50s/it]Epoch  498 | Step   6973/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  498 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  33%|███▎      | 500/1500 [30:30<58:08,  3.49s/it]Epoch  499 | Step   6987/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  499 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  33%|███▎      | 501/1500 [30:33<59:39,  3.58s/it]Epoch  500 | Step   7001/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  500 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.8s | Samples: 6,983

Training Flow Model:  33%|███▎      | 502/1500 [30:37<58:41,  3.53s/it]Epoch  501 | Step   7015/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  501 | Avg Loss: nan | LR: 1.51e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  34%|███▎      | 503/1500 [30:41<59:13,  3.56s/it]Epoch  502 | Step   7029/ 21000 | Loss: nan | LR: 1.51e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  502 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  34%|███▎      | 504/1500 [30:44<58:02,  3.50s/it]Epoch  503 | Step   7043/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  503 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  34%|███▎      | 505/1500 [30:47<57:49,  3.49s/it]Epoch  504 | Step   7057/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  504 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  34%|███▎      | 506/1500 [30:51<57:38,  3.48s/it]Epoch  505 | Step   7071/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  505 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  34%|███▍      | 507/1500 [30:54<57:36,  3.48s/it]Epoch  506 | Step   7085/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  506 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  34%|███▍      | 508/1500 [30:58<57:41,  3.49s/it]Epoch  507 | Step   7099/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  507 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  34%|███▍      | 509/1500 [31:01<57:36,  3.49s/it]Epoch  508 | Step   7113/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  508 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  34%|███▍      | 510/1500 [31:05<58:04,  3.52s/it]Epoch  509 | Step   7127/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  509 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  34%|███▍      | 511/1500 [31:08<57:59,  3.52s/it]Epoch  510 | Step   7141/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  510 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  34%|███▍      | 512/1500 [31:12<57:54,  3.52s/it]Epoch  511 | Step   7155/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  511 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  34%|███▍      | 513/1500 [31:16<58:16,  3.54s/it]Epoch  512 | Step   7169/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  512 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  34%|███▍      | 514/1500 [31:19<58:21,  3.55s/it]Epoch  513 | Step   7183/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  513 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  34%|███▍      | 515/1500 [31:23<57:42,  3.52s/it]Epoch  514 | Step   7197/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  514 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  34%|███▍      | 516/1500 [31:26<58:04,  3.54s/it]Epoch  515 | Step   7211/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  515 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  34%|███▍      | 517/1500 [31:30<57:49,  3.53s/it]Epoch  516 | Step   7225/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  516 | Avg Loss: nan | LR: 1.50e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  35%|███▍      | 518/1500 [31:33<58:03,  3.55s/it]Epoch  517 | Step   7239/ 21000 | Loss: nan | LR: 1.50e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  517 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  35%|███▍      | 519/1500 [31:37<58:06,  3.55s/it]Epoch  518 | Step   7253/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  518 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  35%|███▍      | 520/1500 [31:40<57:57,  3.55s/it]Epoch  519 | Step   7267/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  519 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  35%|███▍      | 521/1500 [31:44<57:56,  3.55s/it]Epoch  520 | Step   7281/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  520 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  35%|███▍      | 522/1500 [31:48<58:21,  3.58s/it]Epoch  521 | Step   7295/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  521 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  35%|███▍      | 523/1500 [31:51<58:30,  3.59s/it]Epoch  522 | Step   7309/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  522 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  35%|███▍      | 524/1500 [31:55<57:58,  3.56s/it]Epoch  523 | Step   7323/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  523 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  35%|███▌      | 525/1500 [31:58<57:44,  3.55s/it]Epoch  524 | Step   7337/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  524 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  35%|███▌      | 526/1500 [32:02<57:13,  3.53s/it]Epoch  525 | Step   7351/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  525 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  35%|███▌      | 527/1500 [32:05<56:43,  3.50s/it]Epoch  526 | Step   7365/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  526 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  35%|███▌      | 528/1500 [32:08<56:03,  3.46s/it]Epoch  527 | Step   7379/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  527 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  35%|███▌      | 529/1500 [32:12<55:45,  3.45s/it]Epoch  528 | Step   7393/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  528 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  35%|███▌      | 530/1500 [32:15<55:28,  3.43s/it]Epoch  529 | Step   7407/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  529 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  35%|███▌      | 531/1500 [32:19<55:12,  3.42s/it]Epoch  530 | Step   7421/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  530 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  35%|███▌      | 532/1500 [32:22<56:04,  3.48s/it]Epoch  531 | Step   7435/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  531 | Avg Loss: nan | LR: 1.49e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  36%|███▌      | 533/1500 [32:26<57:14,  3.55s/it]Epoch  532 | Step   7449/ 21000 | Loss: nan | LR: 1.49e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  532 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  36%|███▌      | 534/1500 [32:30<57:27,  3.57s/it]Epoch  533 | Step   7463/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  533 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  36%|███▌      | 535/1500 [32:33<57:37,  3.58s/it]Epoch  534 | Step   7477/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  534 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  36%|███▌      | 536/1500 [32:37<57:43,  3.59s/it]Epoch  535 | Step   7491/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  535 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  36%|███▌      | 537/1500 [32:40<57:28,  3.58s/it]Epoch  536 | Step   7505/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  536 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  36%|███▌      | 538/1500 [32:44<56:52,  3.55s/it]Epoch  537 | Step   7519/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  537 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  36%|███▌      | 539/1500 [32:47<56:24,  3.52s/it]Epoch  538 | Step   7533/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  538 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  36%|███▌      | 540/1500 [32:51<56:09,  3.51s/it]Epoch  539 | Step   7547/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  539 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  36%|███▌      | 541/1500 [32:54<56:32,  3.54s/it]Epoch  540 | Step   7561/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  540 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  36%|███▌      | 542/1500 [32:58<56:23,  3.53s/it]Epoch  541 | Step   7575/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  541 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  36%|███▌      | 543/1500 [33:01<56:12,  3.52s/it]Epoch  542 | Step   7589/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  542 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  36%|███▋      | 544/1500 [33:05<56:04,  3.52s/it]Epoch  543 | Step   7603/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  543 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  36%|███▋      | 545/1500 [33:08<55:42,  3.50s/it]Epoch  544 | Step   7617/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  544 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  36%|███▋      | 546/1500 [33:12<55:43,  3.51s/it]Epoch  545 | Step   7631/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  545 | Avg Loss: nan | LR: 1.48e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  36%|███▋      | 547/1500 [33:15<55:04,  3.47s/it]Epoch  546 | Step   7645/ 21000 | Loss: nan | LR: 1.48e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  546 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  37%|███▋      | 548/1500 [33:19<55:41,  3.51s/it]Epoch  547 | Step   7659/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  547 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  37%|███▋      | 549/1500 [33:22<55:23,  3.50s/it]Epoch  548 | Step   7673/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  548 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  37%|███▋      | 550/1500 [33:26<54:46,  3.46s/it]Epoch  549 | Step   7687/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  549 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  37%|███▋      | 551/1500 [33:29<55:10,  3.49s/it]Epoch  550 | Step   7701/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  550 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  37%|███▋      | 552/1500 [33:33<54:26,  3.45s/it]Epoch  551 | Step   7715/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  551 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  37%|███▋      | 553/1500 [33:36<56:02,  3.55s/it]Epoch  552 | Step   7729/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  552 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.8s | Samples: 6,983

Training Flow Model:  37%|███▋      | 554/1500 [33:40<56:17,  3.57s/it]Epoch  553 | Step   7743/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  553 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  37%|███▋      | 555/1500 [33:43<55:37,  3.53s/it]Epoch  554 | Step   7757/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  554 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  37%|███▋      | 556/1500 [33:47<55:15,  3.51s/it]Epoch  555 | Step   7771/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  555 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  37%|███▋      | 557/1500 [33:50<55:04,  3.50s/it]Epoch  556 | Step   7785/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  556 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  37%|███▋      | 558/1500 [33:54<55:09,  3.51s/it]Epoch  557 | Step   7799/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  557 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  37%|███▋      | 559/1500 [33:57<54:56,  3.50s/it]Epoch  558 | Step   7813/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  558 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  37%|███▋      | 560/1500 [34:01<55:04,  3.52s/it]Epoch  559 | Step   7827/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  559 | Avg Loss: nan | LR: 1.47e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  37%|███▋      | 561/1500 [34:04<54:44,  3.50s/it]Epoch  560 | Step   7841/ 21000 | Loss: nan | LR: 1.47e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  560 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  37%|███▋      | 562/1500 [34:08<54:37,  3.49s/it]Epoch  561 | Step   7855/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  561 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  38%|███▊      | 563/1500 [34:11<54:34,  3.49s/it]Epoch  562 | Step   7869/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 1.0h
Epoch  562 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  38%|███▊      | 564/1500 [34:15<54:15,  3.48s/it]Epoch  563 | Step   7883/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  563 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  38%|███▊      | 565/1500 [34:18<54:46,  3.51s/it]Epoch  564 | Step   7897/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  564 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  38%|███▊      | 566/1500 [34:22<55:00,  3.53s/it]Epoch  565 | Step   7911/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  565 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  38%|███▊      | 567/1500 [34:26<55:07,  3.54s/it]Epoch  566 | Step   7925/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  566 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  38%|███▊      | 568/1500 [34:29<56:01,  3.61s/it]Epoch  567 | Step   7939/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  567 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  38%|███▊      | 569/1500 [34:33<56:01,  3.61s/it]Epoch  568 | Step   7953/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  568 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  38%|███▊      | 570/1500 [34:37<55:39,  3.59s/it]Epoch  569 | Step   7967/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  569 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  38%|███▊      | 571/1500 [34:40<55:09,  3.56s/it]Epoch  570 | Step   7981/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  570 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  38%|███▊      | 572/1500 [34:43<54:02,  3.49s/it]Epoch  571 | Step   7995/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  571 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  38%|███▊      | 573/1500 [34:47<53:25,  3.46s/it]Epoch  572 | Step   8009/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  572 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  38%|███▊      | 574/1500 [34:50<54:30,  3.53s/it]Epoch  573 | Step   8023/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  573 | Avg Loss: nan | LR: 1.46e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  38%|███▊      | 575/1500 [34:54<53:44,  3.49s/it]Epoch  574 | Step   8037/ 21000 | Loss: nan | LR: 1.46e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  574 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  38%|███▊      | 576/1500 [34:57<54:19,  3.53s/it]Epoch  575 | Step   8051/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  575 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  38%|███▊      | 577/1500 [35:01<54:23,  3.54s/it]Epoch  576 | Step   8065/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  576 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  39%|███▊      | 578/1500 [35:04<54:08,  3.52s/it]Epoch  577 | Step   8079/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  577 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  39%|███▊      | 579/1500 [35:08<54:16,  3.54s/it]Epoch  578 | Step   8093/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  578 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  39%|███▊      | 580/1500 [35:11<53:27,  3.49s/it]Epoch  579 | Step   8107/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  579 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  39%|███▊      | 581/1500 [35:15<53:39,  3.50s/it]Epoch  580 | Step   8121/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  580 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  39%|███▉      | 582/1500 [35:18<53:09,  3.47s/it]Epoch  581 | Step   8135/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  581 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  39%|███▉      | 583/1500 [35:22<54:02,  3.54s/it]Epoch  582 | Step   8149/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  582 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  39%|███▉      | 584/1500 [35:26<54:09,  3.55s/it]Epoch  583 | Step   8163/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  583 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  39%|███▉      | 585/1500 [35:29<53:34,  3.51s/it]Epoch  584 | Step   8177/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  584 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  39%|███▉      | 586/1500 [35:33<53:54,  3.54s/it]Epoch  585 | Step   8191/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  585 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  39%|███▉      | 587/1500 [35:36<53:21,  3.51s/it]Epoch  586 | Step   8205/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  586 | Avg Loss: nan | LR: 1.45e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  39%|███▉      | 588/1500 [35:40<53:30,  3.52s/it]Epoch  587 | Step   8219/ 21000 | Loss: nan | LR: 1.45e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  587 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  39%|███▉      | 589/1500 [35:43<53:41,  3.54s/it]Epoch  588 | Step   8233/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  588 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  39%|███▉      | 590/1500 [35:47<53:57,  3.56s/it]Epoch  589 | Step   8247/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  589 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  39%|███▉      | 591/1500 [35:50<52:51,  3.49s/it]Epoch  590 | Step   8261/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  590 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  39%|███▉      | 592/1500 [35:54<52:58,  3.50s/it]Epoch  591 | Step   8275/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  591 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|███▉      | 593/1500 [35:57<53:04,  3.51s/it]Epoch  592 | Step   8289/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  592 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|███▉      | 594/1500 [36:01<53:07,  3.52s/it]Epoch  593 | Step   8303/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  593 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|███▉      | 595/1500 [36:04<52:58,  3.51s/it]Epoch  594 | Step   8317/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  594 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|███▉      | 596/1500 [36:08<53:02,  3.52s/it]Epoch  595 | Step   8331/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  595 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|███▉      | 597/1500 [36:11<52:29,  3.49s/it]Epoch  596 | Step   8345/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  596 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  40%|███▉      | 598/1500 [36:15<52:00,  3.46s/it]Epoch  597 | Step   8359/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  597 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  40%|███▉      | 599/1500 [36:18<51:57,  3.46s/it]Epoch  598 | Step   8373/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  598 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|████      | 600/1500 [36:25<1:07:13,  4.48s/it]Epoch  599 | Step   8387/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  599 | Avg Loss: nan | LR: 1.44e-03 | Time: 3.6s | Samples: 6,983
✓ Checkpoint saved: /data2/edwardsun/flow_checkpoints/amp_flow_model_final_optimized.pth (loss: nan, step: 8400)

Training Flow Model:  40%|████      | 601/1500 [36:28<1:02:32,  4.17s/it]Epoch  600 | Step   8401/ 21000 | Loss: nan | LR: 1.44e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  600 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|████      | 602/1500 [36:32<59:24,  3.97s/it]  Epoch  601 | Step   8415/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  601 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|████      | 603/1500 [36:35<57:11,  3.83s/it]Epoch  602 | Step   8429/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  602 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|████      | 604/1500 [36:39<55:32,  3.72s/it]Epoch  603 | Step   8443/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  603 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|████      | 605/1500 [36:42<54:27,  3.65s/it]Epoch  604 | Step   8457/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  604 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  40%|████      | 606/1500 [36:46<54:09,  3.63s/it]Epoch  605 | Step   8471/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  605 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  40%|████      | 607/1500 [36:50<54:46,  3.68s/it]Epoch  606 | Step   8485/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  606 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.8s | Samples: 6,983

Training Flow Model:  41%|████      | 608/1500 [36:53<54:07,  3.64s/it]Epoch  607 | Step   8499/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  607 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  41%|████      | 609/1500 [36:57<53:20,  3.59s/it]Epoch  608 | Step   8513/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  608 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  41%|████      | 610/1500 [37:00<53:27,  3.60s/it]Epoch  609 | Step   8527/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  609 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  41%|████      | 611/1500 [37:04<53:05,  3.58s/it]Epoch  610 | Step   8541/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  610 | Avg Loss: nan | LR: 1.43e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  41%|████      | 612/1500 [37:08<56:38,  3.83s/it]Epoch  611 | Step   8555/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  611 | Avg Loss: nan | LR: 1.43e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  41%|████      | 613/1500 [37:13<59:28,  4.02s/it]Epoch  612 | Step   8569/ 21000 | Loss: nan | LR: 1.43e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  612 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  41%|████      | 614/1500 [37:17<1:01:01,  4.13s/it]Epoch  613 | Step   8583/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  613 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  41%|████      | 615/1500 [37:22<1:02:24,  4.23s/it]Epoch  614 | Step   8597/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  614 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  41%|████      | 616/1500 [37:26<1:02:57,  4.27s/it]Epoch  615 | Step   8611/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  615 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  41%|████      | 617/1500 [37:30<1:03:08,  4.29s/it]Epoch  616 | Step   8625/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  616 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  41%|████      | 618/1500 [37:35<1:03:58,  4.35s/it]Epoch  617 | Step   8639/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  617 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  41%|████▏     | 619/1500 [37:39<1:04:24,  4.39s/it]Epoch  618 | Step   8653/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  618 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  41%|████▏     | 620/1500 [37:44<1:05:15,  4.45s/it]Epoch  619 | Step   8667/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  619 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  41%|████▏     | 621/1500 [37:48<1:05:11,  4.45s/it]Epoch  620 | Step   8681/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  620 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  41%|████▏     | 622/1500 [37:53<1:05:25,  4.47s/it]Epoch  621 | Step   8695/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  621 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  42%|████▏     | 623/1500 [37:57<1:04:56,  4.44s/it]Epoch  622 | Step   8709/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  622 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  42%|████▏     | 624/1500 [38:02<1:04:40,  4.43s/it]Epoch  623 | Step   8723/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  623 | Avg Loss: nan | LR: 1.42e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  42%|████▏     | 625/1500 [38:06<1:04:19,  4.41s/it]Epoch  624 | Step   8737/ 21000 | Loss: nan | LR: 1.42e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  624 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  42%|████▏     | 626/1500 [38:10<1:04:37,  4.44s/it]Epoch  625 | Step   8751/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  625 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  42%|████▏     | 627/1500 [38:15<1:05:12,  4.48s/it]Epoch  626 | Step   8765/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  626 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  42%|████▏     | 628/1500 [38:19<1:04:21,  4.43s/it]Epoch  627 | Step   8779/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  627 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  42%|████▏     | 629/1500 [38:24<1:04:31,  4.45s/it]Epoch  628 | Step   8793/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  628 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  42%|████▏     | 630/1500 [38:28<1:04:52,  4.47s/it]Epoch  629 | Step   8807/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  629 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  42%|████▏     | 631/1500 [38:33<1:04:59,  4.49s/it]Epoch  630 | Step   8821/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  630 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  42%|████▏     | 632/1500 [38:37<1:05:19,  4.52s/it]Epoch  631 | Step   8835/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  631 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  42%|████▏     | 633/1500 [38:42<1:04:23,  4.46s/it]Epoch  632 | Step   8849/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  632 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  42%|████▏     | 634/1500 [38:46<1:04:06,  4.44s/it]Epoch  633 | Step   8863/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  633 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  42%|████▏     | 635/1500 [38:50<1:03:14,  4.39s/it]Epoch  634 | Step   8877/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  634 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  42%|████▏     | 636/1500 [38:55<1:03:29,  4.41s/it]Epoch  635 | Step   8891/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  635 | Avg Loss: nan | LR: 1.41e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  42%|████▏     | 637/1500 [39:00<1:04:20,  4.47s/it]Epoch  636 | Step   8905/ 21000 | Loss: nan | LR: 1.41e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  636 | Avg Loss: nan | LR: 1.40e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  43%|████▎     | 638/1500 [39:04<1:03:32,  4.42s/it]Epoch  637 | Step   8919/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  637 | Avg Loss: nan | LR: 1.40e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  43%|████▎     | 639/1500 [39:07<59:44,  4.16s/it]  Epoch  638 | Step   8933/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  638 | Avg Loss: nan | LR: 1.40e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  43%|████▎     | 640/1500 [39:11<56:40,  3.95s/it]Epoch  639 | Step   8947/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  639 | Avg Loss: nan | LR: 1.40e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  43%|████▎     | 641/1500 [39:15<55:33,  3.88s/it]Epoch  640 | Step   8961/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  640 | Avg Loss: nan | LR: 1.40e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  43%|████▎     | 642/1500 [39:18<53:59,  3.78s/it]Epoch  641 | Step   8975/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  641 | Avg Loss: nan | LR: 1.40e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  43%|████▎     | 643/1500 [39:22<53:41,  3.76s/it]Epoch  642 | Step   8989/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  642 | Avg Loss: nan | LR: 1.40e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  43%|████▎     | 644/1500 [39:27<59:37,  4.18s/it]Epoch  643 | Step   9003/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  643 | Avg Loss: nan | LR: 1.40e-03 | Time: 5.2s | Samples: 6,983

Training Flow Model:  43%|████▎     | 645/1500 [39:32<1:01:19,  4.30s/it]Epoch  644 | Step   9017/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  644 | Avg Loss: nan | LR: 1.40e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  43%|████▎     | 646/1500 [39:36<1:01:49,  4.34s/it]Epoch  645 | Step   9031/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  645 | Avg Loss: nan | LR: 1.40e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  43%|████▎     | 647/1500 [39:40<1:01:49,  4.35s/it]Epoch  646 | Step   9045/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  646 | Avg Loss: nan | LR: 1.40e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  43%|████▎     | 648/1500 [39:45<1:01:52,  4.36s/it]Epoch  647 | Step   9059/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  647 | Avg Loss: nan | LR: 1.40e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  43%|████▎     | 649/1500 [39:49<1:02:29,  4.41s/it]Epoch  648 | Step   9073/ 21000 | Loss: nan | LR: 1.40e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  648 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  43%|████▎     | 650/1500 [39:54<1:03:51,  4.51s/it]Epoch  649 | Step   9087/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  649 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  43%|████▎     | 651/1500 [39:59<1:05:11,  4.61s/it]Epoch  650 | Step   9101/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  650 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  43%|████▎     | 652/1500 [40:04<1:06:31,  4.71s/it]Epoch  651 | Step   9115/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  651 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.9s | Samples: 6,983

Training Flow Model:  44%|████▎     | 653/1500 [40:08<1:05:53,  4.67s/it]Epoch  652 | Step   9129/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  652 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  44%|████▎     | 654/1500 [40:13<1:05:13,  4.63s/it]Epoch  653 | Step   9143/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  653 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  44%|████▎     | 655/1500 [40:17<1:04:27,  4.58s/it]Epoch  654 | Step   9157/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  654 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  44%|████▎     | 656/1500 [40:22<1:03:41,  4.53s/it]Epoch  655 | Step   9171/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  655 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  44%|████▍     | 657/1500 [40:27<1:05:07,  4.64s/it]Epoch  656 | Step   9185/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  656 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.9s | Samples: 6,983

Training Flow Model:  44%|████▍     | 658/1500 [40:31<1:05:42,  4.68s/it]Epoch  657 | Step   9199/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  657 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  44%|████▍     | 659/1500 [40:36<1:05:42,  4.69s/it]Epoch  658 | Step   9213/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  658 | Avg Loss: nan | LR: 1.39e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  44%|████▍     | 660/1500 [40:42<1:09:45,  4.98s/it]Epoch  659 | Step   9227/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  659 | Avg Loss: nan | LR: 1.39e-03 | Time: 5.7s | Samples: 6,983

Training Flow Model:  44%|████▍     | 661/1500 [40:47<1:08:35,  4.91s/it]Epoch  660 | Step   9241/ 21000 | Loss: nan | LR: 1.39e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  660 | Avg Loss: nan | LR: 1.38e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  44%|████▍     | 662/1500 [40:51<1:05:48,  4.71s/it]Epoch  661 | Step   9255/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  661 | Avg Loss: nan | LR: 1.38e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  44%|████▍     | 663/1500 [40:54<1:01:04,  4.38s/it]Epoch  662 | Step   9269/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  662 | Avg Loss: nan | LR: 1.38e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  44%|████▍     | 664/1500 [40:58<58:14,  4.18s/it]  Epoch  663 | Step   9283/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  663 | Avg Loss: nan | LR: 1.38e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  44%|████▍     | 665/1500 [41:02<55:04,  3.96s/it]Epoch  664 | Step   9297/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  664 | Avg Loss: nan | LR: 1.38e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  44%|████▍     | 666/1500 [41:05<53:42,  3.86s/it]Epoch  665 | Step   9311/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  665 | Avg Loss: nan | LR: 1.38e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  44%|████▍     | 667/1500 [41:09<52:35,  3.79s/it]Epoch  666 | Step   9325/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  666 | Avg Loss: nan | LR: 1.38e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  45%|████▍     | 668/1500 [41:12<51:19,  3.70s/it]Epoch  667 | Step   9339/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  667 | Avg Loss: nan | LR: 1.38e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  45%|████▍     | 669/1500 [41:16<50:48,  3.67s/it]Epoch  668 | Step   9353/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  668 | Avg Loss: nan | LR: 1.38e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  45%|████▍     | 670/1500 [41:19<50:17,  3.64s/it]Epoch  669 | Step   9367/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  669 | Avg Loss: nan | LR: 1.38e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  45%|████▍     | 671/1500 [41:23<49:28,  3.58s/it]Epoch  670 | Step   9381/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  670 | Avg Loss: nan | LR: 1.38e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  45%|████▍     | 672/1500 [41:26<48:40,  3.53s/it]Epoch  671 | Step   9395/ 21000 | Loss: nan | LR: 1.38e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  671 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  45%|████▍     | 673/1500 [41:30<48:13,  3.50s/it]Epoch  672 | Step   9409/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  672 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  45%|████▍     | 674/1500 [41:33<47:44,  3.47s/it]Epoch  673 | Step   9423/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.9h
Epoch  673 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  45%|████▌     | 675/1500 [41:37<48:38,  3.54s/it]Epoch  674 | Step   9437/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  674 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  45%|████▌     | 676/1500 [41:40<47:53,  3.49s/it]Epoch  675 | Step   9451/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  675 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  45%|████▌     | 677/1500 [41:44<48:05,  3.51s/it]Epoch  676 | Step   9465/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  676 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  45%|████▌     | 678/1500 [41:47<48:27,  3.54s/it]Epoch  677 | Step   9479/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  677 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  45%|████▌     | 679/1500 [41:51<48:50,  3.57s/it]Epoch  678 | Step   9493/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  678 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  45%|████▌     | 680/1500 [41:54<48:07,  3.52s/it]Epoch  679 | Step   9507/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  679 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  45%|████▌     | 681/1500 [41:58<47:50,  3.50s/it]Epoch  680 | Step   9521/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  680 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  45%|████▌     | 682/1500 [42:01<47:31,  3.49s/it]Epoch  681 | Step   9535/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  681 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  46%|████▌     | 683/1500 [42:05<47:45,  3.51s/it]Epoch  682 | Step   9549/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  682 | Avg Loss: nan | LR: 1.37e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  46%|████▌     | 684/1500 [42:08<47:37,  3.50s/it]Epoch  683 | Step   9563/ 21000 | Loss: nan | LR: 1.37e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  683 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  46%|████▌     | 685/1500 [42:12<47:43,  3.51s/it]Epoch  684 | Step   9577/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  684 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  46%|████▌     | 686/1500 [42:16<48:07,  3.55s/it]Epoch  685 | Step   9591/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  685 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  46%|████▌     | 687/1500 [42:19<47:51,  3.53s/it]Epoch  686 | Step   9605/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  686 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  46%|████▌     | 688/1500 [42:23<47:50,  3.54s/it]Epoch  687 | Step   9619/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  687 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  46%|████▌     | 689/1500 [42:26<48:21,  3.58s/it]Epoch  688 | Step   9633/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  688 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  46%|████▌     | 690/1500 [42:30<47:56,  3.55s/it]Epoch  689 | Step   9647/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  689 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  46%|████▌     | 691/1500 [42:33<47:17,  3.51s/it]Epoch  690 | Step   9661/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  690 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  46%|████▌     | 692/1500 [42:37<46:52,  3.48s/it]Epoch  691 | Step   9675/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  691 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  46%|████▌     | 693/1500 [42:40<46:58,  3.49s/it]Epoch  692 | Step   9689/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  692 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  46%|████▋     | 694/1500 [42:44<46:31,  3.46s/it]Epoch  693 | Step   9703/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  693 | Avg Loss: nan | LR: 1.36e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  46%|████▋     | 695/1500 [42:47<46:29,  3.47s/it]Epoch  694 | Step   9717/ 21000 | Loss: nan | LR: 1.36e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  694 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  46%|████▋     | 696/1500 [42:51<46:58,  3.51s/it]Epoch  695 | Step   9731/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  695 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  46%|████▋     | 697/1500 [42:54<46:28,  3.47s/it]Epoch  696 | Step   9745/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  696 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  47%|████▋     | 698/1500 [42:57<46:14,  3.46s/it]Epoch  697 | Step   9759/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  697 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  47%|████▋     | 699/1500 [43:01<46:23,  3.48s/it]Epoch  698 | Step   9773/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  698 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  47%|████▋     | 700/1500 [43:04<46:09,  3.46s/it]Epoch  699 | Step   9787/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  699 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  47%|████▋     | 701/1500 [43:08<46:46,  3.51s/it]Epoch  700 | Step   9801/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  700 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  47%|████▋     | 702/1500 [43:12<46:54,  3.53s/it]Epoch  701 | Step   9815/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  701 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  47%|████▋     | 703/1500 [43:15<47:00,  3.54s/it]Epoch  702 | Step   9829/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  702 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  47%|████▋     | 704/1500 [43:19<46:50,  3.53s/it]Epoch  703 | Step   9843/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  703 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  47%|████▋     | 705/1500 [43:22<47:36,  3.59s/it]Epoch  704 | Step   9857/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  704 | Avg Loss: nan | LR: 1.35e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  47%|████▋     | 706/1500 [43:26<47:10,  3.56s/it]Epoch  705 | Step   9871/ 21000 | Loss: nan | LR: 1.35e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  705 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  47%|████▋     | 707/1500 [43:29<46:24,  3.51s/it]Epoch  706 | Step   9885/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  706 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  47%|████▋     | 708/1500 [43:33<46:10,  3.50s/it]Epoch  707 | Step   9899/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  707 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  47%|████▋     | 709/1500 [43:36<45:23,  3.44s/it]Epoch  708 | Step   9913/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  708 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  47%|████▋     | 710/1500 [43:40<46:04,  3.50s/it]Epoch  709 | Step   9927/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  709 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  47%|████▋     | 711/1500 [43:43<46:12,  3.51s/it]Epoch  710 | Step   9941/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  710 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  47%|████▋     | 712/1500 [43:47<45:46,  3.48s/it]Epoch  711 | Step   9955/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  711 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  48%|████▊     | 713/1500 [43:50<46:07,  3.52s/it]Epoch  712 | Step   9969/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  712 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  48%|████▊     | 714/1500 [43:54<46:31,  3.55s/it]Epoch  713 | Step   9983/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  713 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.6s | Samples: 6,983
/data2/edwardsun/flow_home/cfg_dataset.py:360: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  'index': torch.tensor(idx, dtype=torch.long)
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):

Training Flow Model:  48%|████▊     | 715/1500 [43:59<50:45,  3.88s/it]Epoch  714 | Step   9997/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Validation at step 10000: Loss = nan
Epoch  714 | Avg Loss: nan | LR: 1.34e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  48%|████▊     | 716/1500 [44:02<48:40,  3.73s/it]Epoch  715 | Step  10011/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  715 | Avg Loss: nan | LR: 1.34e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  48%|████▊     | 717/1500 [44:05<47:30,  3.64s/it]Epoch  716 | Step  10025/ 21000 | Loss: nan | LR: 1.34e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  716 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  48%|████▊     | 718/1500 [44:09<46:48,  3.59s/it]Epoch  717 | Step  10039/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  717 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  48%|████▊     | 719/1500 [44:12<46:39,  3.58s/it]Epoch  718 | Step  10053/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  718 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  48%|████▊     | 720/1500 [44:16<45:58,  3.54s/it]Epoch  719 | Step  10067/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  719 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  48%|████▊     | 721/1500 [44:19<46:01,  3.54s/it]Epoch  720 | Step  10081/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  720 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  48%|████▊     | 722/1500 [44:23<45:33,  3.51s/it]Epoch  721 | Step  10095/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  721 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  48%|████▊     | 723/1500 [44:26<45:18,  3.50s/it]Epoch  722 | Step  10109/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  722 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  48%|████▊     | 724/1500 [44:30<45:33,  3.52s/it]Epoch  723 | Step  10123/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  723 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  48%|████▊     | 725/1500 [44:33<45:28,  3.52s/it]Epoch  724 | Step  10137/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  724 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  48%|████▊     | 726/1500 [44:37<45:17,  3.51s/it]Epoch  725 | Step  10151/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  725 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  48%|████▊     | 727/1500 [44:40<45:22,  3.52s/it]Epoch  726 | Step  10165/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  726 | Avg Loss: nan | LR: 1.33e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  49%|████▊     | 728/1500 [44:44<45:11,  3.51s/it]Epoch  727 | Step  10179/ 21000 | Loss: nan | LR: 1.33e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  727 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  49%|████▊     | 729/1500 [44:47<44:43,  3.48s/it]Epoch  728 | Step  10193/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  728 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  49%|████▊     | 730/1500 [44:51<44:51,  3.50s/it]Epoch  729 | Step  10207/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  729 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  49%|████▊     | 731/1500 [44:54<45:08,  3.52s/it]Epoch  730 | Step  10221/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  730 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  49%|████▉     | 732/1500 [44:58<45:22,  3.55s/it]Epoch  731 | Step  10235/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  731 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  49%|████▉     | 733/1500 [45:01<45:02,  3.52s/it]Epoch  732 | Step  10249/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  732 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  49%|████▉     | 734/1500 [45:05<44:57,  3.52s/it]Epoch  733 | Step  10263/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  733 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  49%|████▉     | 735/1500 [45:08<44:12,  3.47s/it]Epoch  734 | Step  10277/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  734 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  49%|████▉     | 736/1500 [45:12<44:04,  3.46s/it]Epoch  735 | Step  10291/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  735 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  49%|████▉     | 737/1500 [45:15<43:45,  3.44s/it]Epoch  736 | Step  10305/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  736 | Avg Loss: nan | LR: 1.32e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  49%|████▉     | 738/1500 [45:19<44:41,  3.52s/it]Epoch  737 | Step  10319/ 21000 | Loss: nan | LR: 1.32e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  737 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  49%|████▉     | 739/1500 [45:22<44:42,  3.53s/it]Epoch  738 | Step  10333/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  738 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  49%|████▉     | 740/1500 [45:26<44:03,  3.48s/it]Epoch  739 | Step  10347/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  739 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  49%|████▉     | 741/1500 [45:29<44:13,  3.50s/it]Epoch  740 | Step  10361/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  740 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  49%|████▉     | 742/1500 [45:33<43:57,  3.48s/it]Epoch  741 | Step  10375/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  741 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  50%|████▉     | 743/1500 [45:36<43:47,  3.47s/it]Epoch  742 | Step  10389/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  742 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  50%|████▉     | 744/1500 [45:40<44:02,  3.50s/it]Epoch  743 | Step  10403/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  743 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  50%|████▉     | 745/1500 [45:43<44:21,  3.52s/it]Epoch  744 | Step  10417/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  744 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  50%|████▉     | 746/1500 [45:47<44:45,  3.56s/it]Epoch  745 | Step  10431/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  745 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  50%|████▉     | 747/1500 [45:51<44:47,  3.57s/it]Epoch  746 | Step  10445/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  746 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  50%|████▉     | 748/1500 [45:54<44:13,  3.53s/it]Epoch  747 | Step  10459/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  747 | Avg Loss: nan | LR: 1.31e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  50%|████▉     | 749/1500 [45:57<43:43,  3.49s/it]Epoch  748 | Step  10473/ 21000 | Loss: nan | LR: 1.31e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  748 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  50%|█████     | 750/1500 [46:01<44:00,  3.52s/it]Epoch  749 | Step  10487/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  749 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  50%|█████     | 751/1500 [46:04<43:44,  3.50s/it]Epoch  750 | Step  10501/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  750 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  50%|█████     | 752/1500 [46:08<43:32,  3.49s/it]Epoch  751 | Step  10515/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  751 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  50%|█████     | 753/1500 [46:12<43:53,  3.53s/it]Epoch  752 | Step  10529/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  752 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  50%|█████     | 754/1500 [46:15<43:51,  3.53s/it]Epoch  753 | Step  10543/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  753 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  50%|█████     | 755/1500 [46:19<44:04,  3.55s/it]Epoch  754 | Step  10557/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  754 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  50%|█████     | 756/1500 [46:22<43:41,  3.52s/it]Epoch  755 | Step  10571/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  755 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  50%|█████     | 757/1500 [46:26<43:07,  3.48s/it]Epoch  756 | Step  10585/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  756 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  51%|█████     | 758/1500 [46:29<42:56,  3.47s/it]Epoch  757 | Step  10599/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  757 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  51%|█████     | 759/1500 [46:32<42:40,  3.45s/it]Epoch  758 | Step  10613/ 21000 | Loss: nan | LR: 1.30e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  758 | Avg Loss: nan | LR: 1.30e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  51%|█████     | 760/1500 [46:36<42:59,  3.49s/it]Epoch  759 | Step  10627/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  759 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  51%|█████     | 761/1500 [46:39<42:47,  3.47s/it]Epoch  760 | Step  10641/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  760 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  51%|█████     | 762/1500 [46:43<42:42,  3.47s/it]Epoch  761 | Step  10655/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  761 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  51%|█████     | 763/1500 [46:46<42:34,  3.47s/it]Epoch  762 | Step  10669/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  762 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  51%|█████     | 764/1500 [46:50<42:33,  3.47s/it]Epoch  763 | Step  10683/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  763 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  51%|█████     | 765/1500 [46:53<42:11,  3.44s/it]Epoch  764 | Step  10697/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  764 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  51%|█████     | 766/1500 [46:57<42:07,  3.44s/it]Epoch  765 | Step  10711/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  765 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  51%|█████     | 767/1500 [47:00<42:11,  3.45s/it]Epoch  766 | Step  10725/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.8h
Epoch  766 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  51%|█████     | 768/1500 [47:04<42:24,  3.48s/it]Epoch  767 | Step  10739/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  767 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  51%|█████▏    | 769/1500 [47:07<42:35,  3.50s/it]Epoch  768 | Step  10753/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  768 | Avg Loss: nan | LR: 1.29e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  51%|█████▏    | 770/1500 [47:11<42:24,  3.49s/it]Epoch  769 | Step  10767/ 21000 | Loss: nan | LR: 1.29e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  769 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  51%|█████▏    | 771/1500 [47:14<42:14,  3.48s/it]Epoch  770 | Step  10781/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  770 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  51%|█████▏    | 772/1500 [47:18<42:07,  3.47s/it]Epoch  771 | Step  10795/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  771 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 773/1500 [47:21<42:16,  3.49s/it]Epoch  772 | Step  10809/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  772 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 774/1500 [47:25<42:41,  3.53s/it]Epoch  773 | Step  10823/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  773 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 775/1500 [47:28<42:22,  3.51s/it]Epoch  774 | Step  10837/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  774 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 776/1500 [47:32<42:27,  3.52s/it]Epoch  775 | Step  10851/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  775 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 777/1500 [47:35<42:26,  3.52s/it]Epoch  776 | Step  10865/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  776 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 778/1500 [47:39<42:42,  3.55s/it]Epoch  777 | Step  10879/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  777 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 779/1500 [47:42<42:52,  3.57s/it]Epoch  778 | Step  10893/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  778 | Avg Loss: nan | LR: 1.28e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 780/1500 [47:46<43:07,  3.59s/it]Epoch  779 | Step  10907/ 21000 | Loss: nan | LR: 1.28e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  779 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 781/1500 [47:50<42:48,  3.57s/it]Epoch  780 | Step  10921/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  780 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 782/1500 [47:53<42:09,  3.52s/it]Epoch  781 | Step  10935/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  781 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 783/1500 [47:57<41:57,  3.51s/it]Epoch  782 | Step  10949/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  782 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 784/1500 [48:00<42:26,  3.56s/it]Epoch  783 | Step  10963/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  783 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 785/1500 [48:04<41:49,  3.51s/it]Epoch  784 | Step  10977/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  784 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 786/1500 [48:07<42:03,  3.53s/it]Epoch  785 | Step  10991/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  785 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  52%|█████▏    | 787/1500 [48:11<41:29,  3.49s/it]Epoch  786 | Step  11005/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  786 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 788/1500 [48:14<41:24,  3.49s/it]Epoch  787 | Step  11019/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  787 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 789/1500 [48:17<41:11,  3.48s/it]Epoch  788 | Step  11033/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  788 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 790/1500 [48:21<41:21,  3.49s/it]Epoch  789 | Step  11047/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  789 | Avg Loss: nan | LR: 1.27e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 791/1500 [48:24<40:57,  3.47s/it]Epoch  790 | Step  11061/ 21000 | Loss: nan | LR: 1.27e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  790 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 792/1500 [48:28<40:56,  3.47s/it]Epoch  791 | Step  11075/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  791 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 793/1500 [48:31<41:16,  3.50s/it]Epoch  792 | Step  11089/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  792 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 794/1500 [48:35<41:24,  3.52s/it]Epoch  793 | Step  11103/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  793 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 795/1500 [48:38<40:54,  3.48s/it]Epoch  794 | Step  11117/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  794 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 796/1500 [48:42<40:41,  3.47s/it]Epoch  795 | Step  11131/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  795 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 797/1500 [48:45<40:34,  3.46s/it]Epoch  796 | Step  11145/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  796 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 798/1500 [48:49<40:31,  3.46s/it]Epoch  797 | Step  11159/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  797 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 799/1500 [48:52<40:18,  3.45s/it]Epoch  798 | Step  11173/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  798 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 800/1500 [48:56<40:15,  3.45s/it]Epoch  799 | Step  11187/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  799 | Avg Loss: nan | LR: 1.26e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 801/1500 [48:59<40:18,  3.46s/it]Epoch  800 | Step  11201/ 21000 | Loss: nan | LR: 1.26e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  800 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  53%|█████▎    | 802/1500 [49:03<40:39,  3.49s/it]Epoch  801 | Step  11215/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  801 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  54%|█████▎    | 803/1500 [49:06<40:54,  3.52s/it]Epoch  802 | Step  11229/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  802 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  54%|█████▎    | 804/1500 [49:10<40:41,  3.51s/it]Epoch  803 | Step  11243/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  803 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  54%|█████▎    | 805/1500 [49:13<40:09,  3.47s/it]Epoch  804 | Step  11257/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  804 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  54%|█████▎    | 806/1500 [49:17<39:55,  3.45s/it]Epoch  805 | Step  11271/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  805 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 807/1500 [49:20<40:08,  3.48s/it]Epoch  806 | Step  11285/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  806 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 808/1500 [49:24<40:31,  3.51s/it]Epoch  807 | Step  11299/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  807 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 809/1500 [49:27<40:27,  3.51s/it]Epoch  808 | Step  11313/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  808 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 810/1500 [49:31<40:17,  3.50s/it]Epoch  809 | Step  11327/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  809 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 811/1500 [49:34<40:21,  3.51s/it]Epoch  810 | Step  11341/ 21000 | Loss: nan | LR: 1.25e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  810 | Avg Loss: nan | LR: 1.25e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 812/1500 [49:38<40:05,  3.50s/it]Epoch  811 | Step  11355/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  811 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 813/1500 [49:41<40:01,  3.50s/it]Epoch  812 | Step  11369/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  812 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 814/1500 [49:45<40:14,  3.52s/it]Epoch  813 | Step  11383/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  813 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 815/1500 [49:48<39:54,  3.50s/it]Epoch  814 | Step  11397/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  814 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 816/1500 [49:52<39:49,  3.49s/it]Epoch  815 | Step  11411/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  815 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  54%|█████▍    | 817/1500 [49:55<39:48,  3.50s/it]Epoch  816 | Step  11425/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  816 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  55%|█████▍    | 818/1500 [49:59<39:31,  3.48s/it]Epoch  817 | Step  11439/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  817 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  55%|█████▍    | 819/1500 [50:02<39:26,  3.47s/it]Epoch  818 | Step  11453/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  818 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  55%|█████▍    | 820/1500 [50:06<39:09,  3.46s/it]Epoch  819 | Step  11467/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  819 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  55%|█████▍    | 821/1500 [50:09<39:37,  3.50s/it]Epoch  820 | Step  11481/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  820 | Avg Loss: nan | LR: 1.24e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  55%|█████▍    | 822/1500 [50:13<40:34,  3.59s/it]Epoch  821 | Step  11495/ 21000 | Loss: nan | LR: 1.24e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  821 | Avg Loss: nan | LR: 1.23e-03 | Time: 3.8s | Samples: 6,983

Training Flow Model:  55%|█████▍    | 823/1500 [50:17<40:48,  3.62s/it]Epoch  822 | Step  11509/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  822 | Avg Loss: nan | LR: 1.23e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  55%|█████▍    | 824/1500 [50:20<40:39,  3.61s/it]Epoch  823 | Step  11523/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  823 | Avg Loss: nan | LR: 1.23e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  55%|█████▌    | 825/1500 [50:24<40:49,  3.63s/it]Epoch  824 | Step  11537/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  824 | Avg Loss: nan | LR: 1.23e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  55%|█████▌    | 826/1500 [50:27<40:44,  3.63s/it]Epoch  825 | Step  11551/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  825 | Avg Loss: nan | LR: 1.23e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  55%|█████▌    | 827/1500 [50:32<43:15,  3.86s/it]Epoch  826 | Step  11565/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  826 | Avg Loss: nan | LR: 1.23e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  55%|█████▌    | 828/1500 [50:36<44:32,  3.98s/it]Epoch  827 | Step  11579/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  827 | Avg Loss: nan | LR: 1.23e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  55%|█████▌    | 829/1500 [50:41<46:05,  4.12s/it]Epoch  828 | Step  11593/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  828 | Avg Loss: nan | LR: 1.23e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  55%|█████▌    | 830/1500 [50:45<47:57,  4.30s/it]Epoch  829 | Step  11607/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  829 | Avg Loss: nan | LR: 1.23e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  55%|█████▌    | 831/1500 [50:50<48:32,  4.35s/it]Epoch  830 | Step  11621/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  830 | Avg Loss: nan | LR: 1.23e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  55%|█████▌    | 832/1500 [50:54<48:58,  4.40s/it]Epoch  831 | Step  11635/ 21000 | Loss: nan | LR: 1.23e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  831 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 833/1500 [50:59<48:43,  4.38s/it]Epoch  832 | Step  11649/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  832 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 834/1500 [51:03<48:56,  4.41s/it]Epoch  833 | Step  11663/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  833 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 835/1500 [51:08<49:00,  4.42s/it]Epoch  834 | Step  11677/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  834 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 836/1500 [51:12<48:59,  4.43s/it]Epoch  835 | Step  11691/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  835 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 837/1500 [51:16<48:52,  4.42s/it]Epoch  836 | Step  11705/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  836 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 838/1500 [51:21<48:42,  4.41s/it]Epoch  837 | Step  11719/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  837 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 839/1500 [51:25<49:10,  4.46s/it]Epoch  838 | Step  11733/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  838 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 840/1500 [51:30<49:09,  4.47s/it]Epoch  839 | Step  11747/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  839 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 841/1500 [51:34<48:56,  4.46s/it]Epoch  840 | Step  11761/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  840 | Avg Loss: nan | LR: 1.22e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 842/1500 [51:39<48:37,  4.43s/it]Epoch  841 | Step  11775/ 21000 | Loss: nan | LR: 1.22e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  841 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  56%|█████▌    | 843/1500 [51:43<48:34,  4.44s/it]Epoch  842 | Step  11789/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  842 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  56%|█████▋    | 844/1500 [51:48<48:43,  4.46s/it]Epoch  843 | Step  11803/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  843 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  56%|█████▋    | 845/1500 [51:52<48:12,  4.42s/it]Epoch  844 | Step  11817/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  844 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  56%|█████▋    | 846/1500 [51:57<49:04,  4.50s/it]Epoch  845 | Step  11831/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  845 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  56%|█████▋    | 847/1500 [52:01<49:17,  4.53s/it]Epoch  846 | Step  11845/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  846 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 848/1500 [52:06<48:34,  4.47s/it]Epoch  847 | Step  11859/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  847 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 849/1500 [52:10<48:06,  4.43s/it]Epoch  848 | Step  11873/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  848 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 850/1500 [52:14<48:16,  4.46s/it]Epoch  849 | Step  11887/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  849 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 851/1500 [52:19<48:57,  4.53s/it]Epoch  850 | Step  11901/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  850 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 852/1500 [52:24<48:45,  4.51s/it]Epoch  851 | Step  11915/ 21000 | Loss: nan | LR: 1.21e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  851 | Avg Loss: nan | LR: 1.21e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 853/1500 [52:28<48:51,  4.53s/it]Epoch  852 | Step  11929/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  852 | Avg Loss: nan | LR: 1.20e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 854/1500 [52:33<48:36,  4.51s/it]Epoch  853 | Step  11943/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  853 | Avg Loss: nan | LR: 1.20e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 855/1500 [52:36<45:21,  4.22s/it]Epoch  854 | Step  11957/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  854 | Avg Loss: nan | LR: 1.20e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 856/1500 [52:40<43:19,  4.04s/it]Epoch  855 | Step  11971/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  855 | Avg Loss: nan | LR: 1.20e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 857/1500 [52:43<41:32,  3.88s/it]Epoch  856 | Step  11985/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  856 | Avg Loss: nan | LR: 1.20e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 858/1500 [52:47<40:13,  3.76s/it]Epoch  857 | Step  11999/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  857 | Avg Loss: nan | LR: 1.20e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 859/1500 [52:50<39:33,  3.70s/it]Epoch  858 | Step  12013/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  858 | Avg Loss: nan | LR: 1.20e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 860/1500 [52:55<43:40,  4.09s/it]Epoch  859 | Step  12027/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  859 | Avg Loss: nan | LR: 1.20e-03 | Time: 5.0s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 861/1500 [53:00<44:44,  4.20s/it]Epoch  860 | Step  12041/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  860 | Avg Loss: nan | LR: 1.20e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  57%|█████▋    | 862/1500 [53:04<45:38,  4.29s/it]Epoch  861 | Step  12055/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  861 | Avg Loss: nan | LR: 1.20e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 863/1500 [53:09<47:23,  4.46s/it]Epoch  862 | Step  12069/ 21000 | Loss: nan | LR: 1.20e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  862 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.9s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 864/1500 [53:14<48:15,  4.55s/it]Epoch  863 | Step  12083/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  863 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 865/1500 [53:18<47:59,  4.53s/it]Epoch  864 | Step  12097/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  864 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 866/1500 [53:23<48:28,  4.59s/it]Epoch  865 | Step  12111/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  865 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 867/1500 [53:28<49:16,  4.67s/it]Epoch  866 | Step  12125/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  866 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.9s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 868/1500 [53:33<49:21,  4.69s/it]Epoch  867 | Step  12139/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  867 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 869/1500 [53:37<48:20,  4.60s/it]Epoch  868 | Step  12153/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.7h
Epoch  868 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 870/1500 [53:42<48:57,  4.66s/it]Epoch  869 | Step  12167/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  869 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 871/1500 [53:47<49:21,  4.71s/it]Epoch  870 | Step  12181/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  870 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 872/1500 [53:51<48:35,  4.64s/it]Epoch  871 | Step  12195/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  871 | Avg Loss: nan | LR: 1.19e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 873/1500 [53:56<48:09,  4.61s/it]Epoch  872 | Step  12209/ 21000 | Loss: nan | LR: 1.19e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  872 | Avg Loss: nan | LR: 1.18e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 874/1500 [54:01<49:19,  4.73s/it]Epoch  873 | Step  12223/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  873 | Avg Loss: nan | LR: 1.18e-03 | Time: 5.0s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 875/1500 [54:05<49:09,  4.72s/it]Epoch  874 | Step  12237/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  874 | Avg Loss: nan | LR: 1.18e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 876/1500 [54:10<47:47,  4.59s/it]Epoch  875 | Step  12251/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  875 | Avg Loss: nan | LR: 1.18e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  58%|█████▊    | 877/1500 [54:15<48:14,  4.65s/it]Epoch  876 | Step  12265/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  876 | Avg Loss: nan | LR: 1.18e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  59%|█████▊    | 878/1500 [54:19<49:09,  4.74s/it]Epoch  877 | Step  12279/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  877 | Avg Loss: nan | LR: 1.18e-03 | Time: 5.0s | Samples: 6,983

Training Flow Model:  59%|█████▊    | 879/1500 [54:24<48:59,  4.73s/it]Epoch  878 | Step  12293/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  878 | Avg Loss: nan | LR: 1.18e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  59%|█████▊    | 880/1500 [54:29<48:19,  4.68s/it]Epoch  879 | Step  12307/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  879 | Avg Loss: nan | LR: 1.18e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  59%|█████▊    | 881/1500 [54:34<48:39,  4.72s/it]Epoch  880 | Step  12321/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  880 | Avg Loss: nan | LR: 1.18e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 882/1500 [54:37<44:44,  4.34s/it]Epoch  881 | Step  12335/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  881 | Avg Loss: nan | LR: 1.18e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 883/1500 [54:41<43:03,  4.19s/it]Epoch  882 | Step  12349/ 21000 | Loss: nan | LR: 1.18e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  882 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.8s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 884/1500 [54:44<41:00,  3.99s/it]Epoch  883 | Step  12363/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  883 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 885/1500 [54:48<39:36,  3.86s/it]Epoch  884 | Step  12377/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  884 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 886/1500 [54:52<38:45,  3.79s/it]Epoch  885 | Step  12391/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  885 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 887/1500 [54:55<38:04,  3.73s/it]Epoch  886 | Step  12405/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  886 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 888/1500 [54:59<37:47,  3.70s/it]Epoch  887 | Step  12419/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  887 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 889/1500 [55:02<36:46,  3.61s/it]Epoch  888 | Step  12433/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  888 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 890/1500 [55:06<36:04,  3.55s/it]Epoch  889 | Step  12447/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  889 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 891/1500 [55:09<35:36,  3.51s/it]Epoch  890 | Step  12461/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  890 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  59%|█████▉    | 892/1500 [55:13<36:10,  3.57s/it]Epoch  891 | Step  12475/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  891 | Avg Loss: nan | LR: 1.17e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  60%|█████▉    | 893/1500 [55:16<36:24,  3.60s/it]Epoch  892 | Step  12489/ 21000 | Loss: nan | LR: 1.17e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  892 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  60%|█████▉    | 894/1500 [55:20<36:03,  3.57s/it]Epoch  893 | Step  12503/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  893 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  60%|█████▉    | 895/1500 [55:23<35:29,  3.52s/it]Epoch  894 | Step  12517/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  894 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  60%|█████▉    | 896/1500 [55:27<35:09,  3.49s/it]Epoch  895 | Step  12531/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  895 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  60%|█████▉    | 897/1500 [55:30<34:58,  3.48s/it]Epoch  896 | Step  12545/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  896 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  60%|█████▉    | 898/1500 [55:34<34:34,  3.45s/it]Epoch  897 | Step  12559/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  897 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  60%|█████▉    | 899/1500 [55:37<34:28,  3.44s/it]Epoch  898 | Step  12573/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  898 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  60%|██████    | 900/1500 [55:44<44:57,  4.50s/it]Epoch  899 | Step  12587/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  899 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.5s | Samples: 6,983
✓ Checkpoint saved: /data2/edwardsun/flow_checkpoints/amp_flow_model_final_optimized.pth (loss: nan, step: 12600)

Training Flow Model:  60%|██████    | 901/1500 [55:47<41:42,  4.18s/it]Epoch  900 | Step  12601/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  900 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  60%|██████    | 902/1500 [55:51<39:41,  3.98s/it]Epoch  901 | Step  12615/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  901 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  60%|██████    | 903/1500 [55:54<37:49,  3.80s/it]Epoch  902 | Step  12629/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  902 | Avg Loss: nan | LR: 1.16e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  60%|██████    | 904/1500 [55:58<36:48,  3.71s/it]Epoch  903 | Step  12643/ 21000 | Loss: nan | LR: 1.16e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  903 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  60%|██████    | 905/1500 [56:01<35:57,  3.63s/it]Epoch  904 | Step  12657/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  904 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  60%|██████    | 906/1500 [56:05<35:25,  3.58s/it]Epoch  905 | Step  12671/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  905 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  60%|██████    | 907/1500 [56:08<35:14,  3.57s/it]Epoch  906 | Step  12685/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  906 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  61%|██████    | 908/1500 [56:12<34:57,  3.54s/it]Epoch  907 | Step  12699/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  907 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  61%|██████    | 909/1500 [56:15<34:49,  3.54s/it]Epoch  908 | Step  12713/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  908 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  61%|██████    | 910/1500 [56:19<34:54,  3.55s/it]Epoch  909 | Step  12727/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  909 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  61%|██████    | 911/1500 [56:22<35:00,  3.57s/it]Epoch  910 | Step  12741/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  910 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  61%|██████    | 912/1500 [56:26<34:27,  3.52s/it]Epoch  911 | Step  12755/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  911 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  61%|██████    | 913/1500 [56:30<35:06,  3.59s/it]Epoch  912 | Step  12769/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  912 | Avg Loss: nan | LR: 1.15e-03 | Time: 3.8s | Samples: 6,983

Training Flow Model:  61%|██████    | 914/1500 [56:33<34:53,  3.57s/it]Epoch  913 | Step  12783/ 21000 | Loss: nan | LR: 1.15e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  913 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  61%|██████    | 915/1500 [56:37<34:46,  3.57s/it]Epoch  914 | Step  12797/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  914 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  61%|██████    | 916/1500 [56:40<34:45,  3.57s/it]Epoch  915 | Step  12811/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  915 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  61%|██████    | 917/1500 [56:44<34:12,  3.52s/it]Epoch  916 | Step  12825/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  916 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  61%|██████    | 918/1500 [56:47<34:07,  3.52s/it]Epoch  917 | Step  12839/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  917 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  61%|██████▏   | 919/1500 [56:51<33:50,  3.49s/it]Epoch  918 | Step  12853/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  918 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  61%|██████▏   | 920/1500 [56:54<33:43,  3.49s/it]Epoch  919 | Step  12867/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  919 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  61%|██████▏   | 921/1500 [56:58<33:46,  3.50s/it]Epoch  920 | Step  12881/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  920 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  61%|██████▏   | 922/1500 [57:01<33:51,  3.51s/it]Epoch  921 | Step  12895/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  921 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 923/1500 [57:05<34:05,  3.54s/it]Epoch  922 | Step  12909/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  922 | Avg Loss: nan | LR: 1.14e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 924/1500 [57:08<34:01,  3.54s/it]Epoch  923 | Step  12923/ 21000 | Loss: nan | LR: 1.14e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  923 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 925/1500 [57:12<34:03,  3.55s/it]Epoch  924 | Step  12937/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  924 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 926/1500 [57:15<33:49,  3.54s/it]Epoch  925 | Step  12951/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  925 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 927/1500 [57:19<33:40,  3.53s/it]Epoch  926 | Step  12965/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  926 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 928/1500 [57:22<33:28,  3.51s/it]Epoch  927 | Step  12979/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  927 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 929/1500 [57:26<33:12,  3.49s/it]Epoch  928 | Step  12993/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  928 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 930/1500 [57:29<32:54,  3.46s/it]Epoch  929 | Step  13007/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  929 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 931/1500 [57:33<33:19,  3.51s/it]Epoch  930 | Step  13021/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  930 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 932/1500 [57:36<33:30,  3.54s/it]Epoch  931 | Step  13035/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  931 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 933/1500 [57:40<33:19,  3.53s/it]Epoch  932 | Step  13049/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  932 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 934/1500 [57:43<33:27,  3.55s/it]Epoch  933 | Step  13063/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  933 | Avg Loss: nan | LR: 1.13e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 935/1500 [57:47<33:27,  3.55s/it]Epoch  934 | Step  13077/ 21000 | Loss: nan | LR: 1.13e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  934 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 936/1500 [57:51<33:19,  3.54s/it]Epoch  935 | Step  13091/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  935 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  62%|██████▏   | 937/1500 [57:54<33:20,  3.55s/it]Epoch  936 | Step  13105/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  936 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 938/1500 [57:58<33:03,  3.53s/it]Epoch  937 | Step  13119/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  937 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 939/1500 [58:01<33:00,  3.53s/it]Epoch  938 | Step  13133/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  938 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 940/1500 [58:05<32:54,  3.53s/it]Epoch  939 | Step  13147/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  939 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 941/1500 [58:08<33:20,  3.58s/it]Epoch  940 | Step  13161/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  940 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 942/1500 [58:12<33:03,  3.56s/it]Epoch  941 | Step  13175/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  941 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 943/1500 [58:15<32:54,  3.55s/it]Epoch  942 | Step  13189/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  942 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 944/1500 [58:19<32:45,  3.54s/it]Epoch  943 | Step  13203/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  943 | Avg Loss: nan | LR: 1.12e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 945/1500 [58:23<32:59,  3.57s/it]Epoch  944 | Step  13217/ 21000 | Loss: nan | LR: 1.12e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  944 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 946/1500 [58:26<33:02,  3.58s/it]Epoch  945 | Step  13231/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  945 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 947/1500 [58:30<32:48,  3.56s/it]Epoch  946 | Step  13245/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  946 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 948/1500 [58:33<33:01,  3.59s/it]Epoch  947 | Step  13259/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  947 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 949/1500 [58:37<33:01,  3.60s/it]Epoch  948 | Step  13273/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  948 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 950/1500 [58:40<32:38,  3.56s/it]Epoch  949 | Step  13287/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  949 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 951/1500 [58:44<32:43,  3.58s/it]Epoch  950 | Step  13301/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  950 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  63%|██████▎   | 952/1500 [58:48<32:28,  3.56s/it]Epoch  951 | Step  13315/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  951 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  64%|██████▎   | 953/1500 [58:51<32:11,  3.53s/it]Epoch  952 | Step  13329/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  952 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  64%|██████▎   | 954/1500 [58:55<32:04,  3.52s/it]Epoch  953 | Step  13343/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  953 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  64%|██████▎   | 955/1500 [58:58<32:10,  3.54s/it]Epoch  954 | Step  13357/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  954 | Avg Loss: nan | LR: 1.11e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  64%|██████▎   | 956/1500 [59:02<31:41,  3.49s/it]Epoch  955 | Step  13371/ 21000 | Loss: nan | LR: 1.11e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  955 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 957/1500 [59:05<31:48,  3.51s/it]Epoch  956 | Step  13385/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  956 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 958/1500 [59:08<31:30,  3.49s/it]Epoch  957 | Step  13399/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  957 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 959/1500 [59:12<31:47,  3.53s/it]Epoch  958 | Step  13413/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  958 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 960/1500 [59:16<31:56,  3.55s/it]Epoch  959 | Step  13427/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  959 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 961/1500 [59:19<31:36,  3.52s/it]Epoch  960 | Step  13441/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  960 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 962/1500 [59:23<31:28,  3.51s/it]Epoch  961 | Step  13455/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  961 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 963/1500 [59:26<31:16,  3.49s/it]Epoch  962 | Step  13469/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  962 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 964/1500 [59:29<30:56,  3.46s/it]Epoch  963 | Step  13483/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  963 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 965/1500 [59:33<30:56,  3.47s/it]Epoch  964 | Step  13497/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  964 | Avg Loss: nan | LR: 1.10e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 966/1500 [59:36<30:58,  3.48s/it]Epoch  965 | Step  13511/ 21000 | Loss: nan | LR: 1.10e-03 | Speed: 3.8 steps/s | ETA: 0.6h
Epoch  965 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  64%|██████▍   | 967/1500 [59:40<31:06,  3.50s/it]Epoch  966 | Step  13525/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  966 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  65%|██████▍   | 968/1500 [59:44<31:12,  3.52s/it]Epoch  967 | Step  13539/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  967 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  65%|██████▍   | 969/1500 [59:47<31:27,  3.55s/it]Epoch  968 | Step  13553/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  968 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  65%|██████▍   | 970/1500 [59:51<31:15,  3.54s/it]Epoch  969 | Step  13567/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  969 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  65%|██████▍   | 971/1500 [59:54<30:59,  3.51s/it]Epoch  970 | Step  13581/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  970 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  65%|██████▍   | 972/1500 [59:58<30:55,  3.51s/it]Epoch  971 | Step  13595/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  971 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  65%|██████▍   | 973/1500 [1:00:01<30:49,  3.51s/it]Epoch  972 | Step  13609/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  972 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  65%|██████▍   | 974/1500 [1:00:05<30:56,  3.53s/it]Epoch  973 | Step  13623/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  973 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  65%|██████▌   | 975/1500 [1:00:08<30:43,  3.51s/it]Epoch  974 | Step  13637/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  974 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  65%|██████▌   | 976/1500 [1:00:12<30:21,  3.48s/it]Epoch  975 | Step  13651/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  975 | Avg Loss: nan | LR: 1.09e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  65%|██████▌   | 977/1500 [1:00:15<30:23,  3.49s/it]Epoch  976 | Step  13665/ 21000 | Loss: nan | LR: 1.09e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  976 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  65%|██████▌   | 978/1500 [1:00:19<30:20,  3.49s/it]Epoch  977 | Step  13679/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  977 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  65%|██████▌   | 979/1500 [1:00:22<30:23,  3.50s/it]Epoch  978 | Step  13693/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  978 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  65%|██████▌   | 980/1500 [1:00:26<29:57,  3.46s/it]Epoch  979 | Step  13707/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  979 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  65%|██████▌   | 981/1500 [1:00:29<29:57,  3.46s/it]Epoch  980 | Step  13721/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  980 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  65%|██████▌   | 982/1500 [1:00:32<29:48,  3.45s/it]Epoch  981 | Step  13735/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  981 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 983/1500 [1:00:36<29:50,  3.46s/it]Epoch  982 | Step  13749/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  982 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 984/1500 [1:00:39<29:48,  3.47s/it]Epoch  983 | Step  13763/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  983 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 985/1500 [1:00:43<29:34,  3.45s/it]Epoch  984 | Step  13777/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  984 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 986/1500 [1:00:46<29:56,  3.50s/it]Epoch  985 | Step  13791/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  985 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 987/1500 [1:00:50<29:36,  3.46s/it]Epoch  986 | Step  13805/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  986 | Avg Loss: nan | LR: 1.08e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 988/1500 [1:00:53<29:30,  3.46s/it]Epoch  987 | Step  13819/ 21000 | Loss: nan | LR: 1.08e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  987 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 989/1500 [1:00:57<29:29,  3.46s/it]Epoch  988 | Step  13833/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  988 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 990/1500 [1:01:00<29:38,  3.49s/it]Epoch  989 | Step  13847/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  989 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 991/1500 [1:01:04<29:31,  3.48s/it]Epoch  990 | Step  13861/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  990 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 992/1500 [1:01:07<29:19,  3.46s/it]Epoch  991 | Step  13875/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  991 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  66%|██████▌   | 993/1500 [1:01:11<29:23,  3.48s/it]Epoch  992 | Step  13889/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  992 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  66%|██████▋   | 994/1500 [1:01:14<29:46,  3.53s/it]Epoch  993 | Step  13903/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  993 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  66%|██████▋   | 995/1500 [1:01:18<29:28,  3.50s/it]Epoch  994 | Step  13917/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  994 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  66%|██████▋   | 996/1500 [1:01:21<29:23,  3.50s/it]Epoch  995 | Step  13931/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  995 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  66%|██████▋   | 997/1500 [1:01:25<28:51,  3.44s/it]Epoch  996 | Step  13945/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  996 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.3s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 998/1500 [1:01:28<29:03,  3.47s/it]Epoch  997 | Step  13959/ 21000 | Loss: nan | LR: 1.07e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  997 | Avg Loss: nan | LR: 1.07e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 999/1500 [1:01:32<29:43,  3.56s/it]Epoch  998 | Step  13973/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  998 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.8s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1000/1500 [1:01:35<29:46,  3.57s/it]Epoch  999 | Step  13987/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch  999 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1001/1500 [1:01:39<29:39,  3.57s/it]Epoch 1000 | Step  14001/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1000 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1002/1500 [1:01:43<29:32,  3.56s/it]Epoch 1001 | Step  14015/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1001 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1003/1500 [1:01:46<29:09,  3.52s/it]Epoch 1002 | Step  14029/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1002 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1004/1500 [1:01:50<29:23,  3.56s/it]Epoch 1003 | Step  14043/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1003 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1005/1500 [1:01:53<29:00,  3.52s/it]Epoch 1004 | Step  14057/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1004 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1006/1500 [1:01:57<28:57,  3.52s/it]Epoch 1005 | Step  14071/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1005 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1007/1500 [1:02:00<28:56,  3.52s/it]Epoch 1006 | Step  14085/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1006 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1008/1500 [1:02:04<29:06,  3.55s/it]Epoch 1007 | Step  14099/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1007 | Avg Loss: nan | LR: 1.06e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1009/1500 [1:02:07<29:05,  3.55s/it]Epoch 1008 | Step  14113/ 21000 | Loss: nan | LR: 1.06e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1008 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1010/1500 [1:02:11<28:49,  3.53s/it]Epoch 1009 | Step  14127/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1009 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1011/1500 [1:02:14<28:29,  3.50s/it]Epoch 1010 | Step  14141/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1010 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  67%|██████▋   | 1012/1500 [1:02:18<28:27,  3.50s/it]Epoch 1011 | Step  14155/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1011 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1013/1500 [1:02:21<28:32,  3.52s/it]Epoch 1012 | Step  14169/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1012 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1014/1500 [1:02:25<28:24,  3.51s/it]Epoch 1013 | Step  14183/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1013 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1015/1500 [1:02:28<28:30,  3.53s/it]Epoch 1014 | Step  14197/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1014 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1016/1500 [1:02:32<28:04,  3.48s/it]Epoch 1015 | Step  14211/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1015 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1017/1500 [1:02:35<27:57,  3.47s/it]Epoch 1016 | Step  14225/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1016 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1018/1500 [1:02:39<28:09,  3.51s/it]Epoch 1017 | Step  14239/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1017 | Avg Loss: nan | LR: 1.05e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1019/1500 [1:02:43<29:36,  3.69s/it]Epoch 1018 | Step  14253/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1018 | Avg Loss: nan | LR: 1.05e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1020/1500 [1:02:46<29:02,  3.63s/it]Epoch 1019 | Step  14267/ 21000 | Loss: nan | LR: 1.05e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1019 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1021/1500 [1:02:50<28:23,  3.56s/it]Epoch 1020 | Step  14281/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1020 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1022/1500 [1:02:54<30:30,  3.83s/it]Epoch 1021 | Step  14295/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1021 | Avg Loss: nan | LR: 1.04e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1023/1500 [1:02:58<29:19,  3.69s/it]Epoch 1022 | Step  14309/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1022 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1024/1500 [1:03:01<29:03,  3.66s/it]Epoch 1023 | Step  14323/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1023 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1025/1500 [1:03:05<28:23,  3.59s/it]Epoch 1024 | Step  14337/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1024 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1026/1500 [1:03:08<28:05,  3.56s/it]Epoch 1025 | Step  14351/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1025 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  68%|██████▊   | 1027/1500 [1:03:11<27:42,  3.51s/it]Epoch 1026 | Step  14365/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1026 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.4s | Samples: 6,983

Training Flow Model:  69%|██████▊   | 1028/1500 [1:03:15<27:36,  3.51s/it]Epoch 1027 | Step  14379/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1027 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  69%|██████▊   | 1029/1500 [1:03:19<27:51,  3.55s/it]Epoch 1028 | Step  14393/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1028 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  69%|██████▊   | 1030/1500 [1:03:22<27:42,  3.54s/it]Epoch 1029 | Step  14407/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1029 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  69%|██████▊   | 1031/1500 [1:03:26<27:39,  3.54s/it]Epoch 1030 | Step  14421/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1030 | Avg Loss: nan | LR: 1.04e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1032/1500 [1:03:29<27:33,  3.53s/it]Epoch 1031 | Step  14435/ 21000 | Loss: nan | LR: 1.04e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1031 | Avg Loss: nan | LR: 1.03e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1033/1500 [1:03:33<27:45,  3.57s/it]Epoch 1032 | Step  14449/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1032 | Avg Loss: nan | LR: 1.03e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1034/1500 [1:03:36<27:48,  3.58s/it]Epoch 1033 | Step  14463/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1033 | Avg Loss: nan | LR: 1.03e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1035/1500 [1:03:40<27:57,  3.61s/it]Epoch 1034 | Step  14477/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1034 | Avg Loss: nan | LR: 1.03e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1036/1500 [1:03:44<27:58,  3.62s/it]Epoch 1035 | Step  14491/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1035 | Avg Loss: nan | LR: 1.03e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1037/1500 [1:03:48<29:51,  3.87s/it]Epoch 1036 | Step  14505/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1036 | Avg Loss: nan | LR: 1.03e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1038/1500 [1:03:53<31:06,  4.04s/it]Epoch 1037 | Step  14519/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1037 | Avg Loss: nan | LR: 1.03e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1039/1500 [1:03:57<31:53,  4.15s/it]Epoch 1038 | Step  14533/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1038 | Avg Loss: nan | LR: 1.03e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1040/1500 [1:04:01<32:11,  4.20s/it]Epoch 1039 | Step  14547/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1039 | Avg Loss: nan | LR: 1.03e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1041/1500 [1:04:06<32:52,  4.30s/it]Epoch 1040 | Step  14561/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1040 | Avg Loss: nan | LR: 1.03e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  69%|██████▉   | 1042/1500 [1:04:11<33:42,  4.42s/it]Epoch 1041 | Step  14575/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1041 | Avg Loss: nan | LR: 1.03e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  70%|██████▉   | 1043/1500 [1:04:15<33:30,  4.40s/it]Epoch 1042 | Step  14589/ 21000 | Loss: nan | LR: 1.03e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1042 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  70%|██████▉   | 1044/1500 [1:04:20<34:02,  4.48s/it]Epoch 1043 | Step  14603/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1043 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  70%|██████▉   | 1045/1500 [1:04:24<33:56,  4.48s/it]Epoch 1044 | Step  14617/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1044 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  70%|██████▉   | 1046/1500 [1:04:29<34:00,  4.50s/it]Epoch 1045 | Step  14631/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1045 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  70%|██████▉   | 1047/1500 [1:04:33<34:09,  4.52s/it]Epoch 1046 | Step  14645/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1046 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  70%|██████▉   | 1048/1500 [1:04:38<33:50,  4.49s/it]Epoch 1047 | Step  14659/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1047 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  70%|██████▉   | 1049/1500 [1:04:42<33:28,  4.45s/it]Epoch 1048 | Step  14673/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1048 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  70%|███████   | 1050/1500 [1:04:46<33:15,  4.43s/it]Epoch 1049 | Step  14687/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1049 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  70%|███████   | 1051/1500 [1:04:51<33:33,  4.49s/it]Epoch 1050 | Step  14701/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1050 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  70%|███████   | 1052/1500 [1:04:55<33:30,  4.49s/it]Epoch 1051 | Step  14715/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1051 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  70%|███████   | 1053/1500 [1:05:00<33:42,  4.52s/it]Epoch 1052 | Step  14729/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1052 | Avg Loss: nan | LR: 1.02e-03 | Time: 4.6s | Samples: 6,983

Training Flow Model:  70%|███████   | 1054/1500 [1:05:05<33:29,  4.51s/it]Epoch 1053 | Step  14743/ 21000 | Loss: nan | LR: 1.02e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1053 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.5s | Samples: 6,983

Training Flow Model:  70%|███████   | 1055/1500 [1:05:09<33:17,  4.49s/it]Epoch 1054 | Step  14757/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1054 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  70%|███████   | 1056/1500 [1:05:13<32:49,  4.43s/it]Epoch 1055 | Step  14771/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1055 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  70%|███████   | 1057/1500 [1:05:18<32:45,  4.44s/it]Epoch 1056 | Step  14785/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1056 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  71%|███████   | 1058/1500 [1:05:22<33:12,  4.51s/it]Epoch 1057 | Step  14799/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1057 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.7s | Samples: 6,983

Training Flow Model:  71%|███████   | 1059/1500 [1:05:27<32:47,  4.46s/it]Epoch 1058 | Step  14813/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1058 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  71%|███████   | 1060/1500 [1:05:32<33:21,  4.55s/it]Epoch 1059 | Step  14827/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1059 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.8s | Samples: 6,983

Training Flow Model:  71%|███████   | 1061/1500 [1:05:36<33:01,  4.51s/it]Epoch 1060 | Step  14841/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1060 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  71%|███████   | 1062/1500 [1:05:40<32:36,  4.47s/it]Epoch 1061 | Step  14855/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1061 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.4s | Samples: 6,983

Training Flow Model:  71%|███████   | 1063/1500 [1:05:45<32:11,  4.42s/it]Epoch 1062 | Step  14869/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1062 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  71%|███████   | 1064/1500 [1:05:49<31:55,  4.39s/it]Epoch 1063 | Step  14883/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.5h
Epoch 1063 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.3s | Samples: 6,983

Training Flow Model:  71%|███████   | 1065/1500 [1:05:53<31:09,  4.30s/it]Epoch 1064 | Step  14897/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1064 | Avg Loss: nan | LR: 1.01e-03 | Time: 4.1s | Samples: 6,983

Training Flow Model:  71%|███████   | 1066/1500 [1:05:56<29:20,  4.06s/it]Epoch 1065 | Step  14911/ 21000 | Loss: nan | LR: 1.01e-03 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1065 | Avg Loss: nan | LR: 1.00e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  71%|███████   | 1067/1500 [1:06:00<28:09,  3.90s/it]Epoch 1066 | Step  14925/ 21000 | Loss: nan | LR: 1.00e-03 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1066 | Avg Loss: nan | LR: 1.00e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  71%|███████   | 1068/1500 [1:06:04<27:18,  3.79s/it]Epoch 1067 | Step  14939/ 21000 | Loss: nan | LR: 1.00e-03 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1067 | Avg Loss: nan | LR: 1.00e-03 | Time: 3.5s | Samples: 6,983

Training Flow Model:  71%|███████▏  | 1069/1500 [1:06:07<27:03,  3.77s/it]Epoch 1068 | Step  14953/ 21000 | Loss: nan | LR: 1.00e-03 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1068 | Avg Loss: nan | LR: 1.00e-03 | Time: 3.7s | Samples: 6,983

Training Flow Model:  71%|███████▏  | 1070/1500 [1:06:11<26:33,  3.71s/it]Epoch 1069 | Step  14967/ 21000 | Loss: nan | LR: 1.00e-03 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1069 | Avg Loss: nan | LR: 1.00e-03 | Time: 3.6s | Samples: 6,983

Training Flow Model:  71%|███████▏  | 1071/1500 [1:06:16<28:55,  4.05s/it]Epoch 1070 | Step  14981/ 21000 | Loss: nan | LR: 1.00e-03 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1070 | Avg Loss: nan | LR: 1.00e-03 | Time: 4.8s | Samples: 6,983
/data2/edwardsun/flow_home/cfg_dataset.py:360: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  'index': torch.tensor(idx, dtype=torch.long)
/data2/edwardsun/flow_home/cfg_dataset.py:360: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  'index': torch.tensor(idx, dtype=torch.long)
/data2/edwardsun/flow_home/amp_flow_training_single_gpu_full_data.py:392: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with autocast(dtype=torch.bfloat16):

Training Flow Model:  71%|███████▏  | 1072/1500 [1:06:22<32:52,  4.61s/it]Epoch 1071 | Step  14995/ 21000 | Loss: nan | LR: 1.00e-03 | Speed: 3.8 steps/s | ETA: 0.4h
Validation at step 15000: Loss = nan
Epoch 1071 | Avg Loss: nan | LR: 1.00e-03 | Time: 5.9s | Samples: 6,983

Training Flow Model:  72%|███████▏  | 1073/1500 [1:06:26<32:32,  4.57s/it]Epoch 1072 | Step  15009/ 21000 | Loss: nan | LR: 9.99e-04 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1072 | Avg Loss: nan | LR: 9.99e-04 | Time: 4.5s | Samples: 6,983

Training Flow Model:  72%|███████▏  | 1074/1500 [1:06:31<33:15,  4.68s/it]Epoch 1073 | Step  15023/ 21000 | Loss: nan | LR: 9.99e-04 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1073 | Avg Loss: nan | LR: 9.98e-04 | Time: 4.9s | Samples: 6,983

Training Flow Model:  72%|███████▏  | 1075/1500 [1:06:36<33:22,  4.71s/it]Epoch 1074 | Step  15037/ 21000 | Loss: nan | LR: 9.98e-04 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1074 | Avg Loss: nan | LR: 9.97e-04 | Time: 4.8s | Samples: 6,983

Training Flow Model:  72%|███████▏  | 1076/1500 [1:06:40<32:49,  4.65s/it]Epoch 1075 | Step  15051/ 21000 | Loss: nan | LR: 9.97e-04 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1075 | Avg Loss: nan | LR: 9.96e-04 | Time: 4.5s | Samples: 6,983

Training Flow Model:  72%|███████▏  | 1077/1500 [1:06:45<32:34,  4.62s/it]Epoch 1076 | Step  15065/ 21000 | Loss: nan | LR: 9.96e-04 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1076 | Avg Loss: nan | LR: 9.95e-04 | Time: 4.6s | Samples: 6,983

Training Flow Model:  72%|███████▏  | 1078/1500 [1:06:50<32:58,  4.69s/it]Epoch 1077 | Step  15079/ 21000 | Loss: nan | LR: 9.95e-04 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1077 | Avg Loss: nan | LR: 9.94e-04 | Time: 4.8s | Samples: 6,983

Training Flow Model:  72%|███████▏  | 1079/1500 [1:06:54<32:45,  4.67s/it]Epoch 1078 | Step  15093/ 21000 | Loss: nan | LR: 9.94e-04 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1078 | Avg Loss: nan | LR: 9.94e-04 | Time: 4.6s | Samples: 6,983

Training Flow Model:  72%|███████▏  | 1080/1500 [1:06:59<32:25,  4.63s/it]Epoch 1079 | Step  15107/ 21000 | Loss: nan | LR: 9.94e-04 | Speed: 3.8 steps/s | ETA: 0.4h
Epoch 1079 | Avg Loss: nan | LR: 9.93e-04 | Time: 4.6s | Samples: 6,983