Upload logs/limb_sweep_i10_bs1000_ex0.5_nu0.3_gpu6.log

logs/limb_sweep_i10_bs1000_ex0.5_nu0.3_gpu6.log

@@ -0,0 +1,205 @@
+COMMAND: CUDA_VISIBLE_DEVICES=6 WANDB_MODE=offline python main.py fit -c conf/difftree/C_Limb_1gpu.yaml --data.K=10 --data.batch_size=1000 --model.ec_ce_weight=0.5 --model.exaggeration_lat=0.5 --model.nu_lat=0.3 --model.weightrout=0.5 --trainer.logger.init_args.name=limb_sweep_i10_bs1000_ex0.5_nu0.3_gpu6 --trainer.enable_progress_bar=False
+Seed set to 42
+GPU available: True (cuda), used: True
+TPU available: False, using: 0 TPU cores
+HPU available: False, using: 0 HPUs
+wandb: Tracking run with wandb version 0.20.1
+wandb: W&B syncing is set to `offline` in this directory. Run `wandb online` or set WANDB_MODE=online to enable cloud syncing.
+LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [6]
+
+  | Name                | Type               | Params | Mode 
+-------------------------------------------------------------------
+0 | enc                 | TransformerEncoder | 1.6 M  | train
+1 | UNet_model          | AE_layer2          | 64.7 M | train
+2 | UNet_ema            | AE_layer2          | 64.7 M | train
+3 | tree_node_embedding | ModuleList         | 4.1 K  | train
+4 | vis                 | Sequential         | 258 K  | train
+5 | diffusion           | GaussianDiffusion  | 0      | train
+  | other params        | n/a                | 4      | n/a  
+-------------------------------------------------------------------
+131 M     Trainable params
+0         Non-trainable params
+131 M     Total params
+525.041   Total estimated model params size (MB)
+121       Modules in train mode
+0         Modules in eval mode
+Using fully connected network
+data.shape (66633, 500)
+label (66633,)
+load index trainval_index_train_0.8_0.1_Limb.npy trainval_index_val_0.8_0.1_Limb.npy trainval_index_test_0.8_0.1_Limb.npy from data/trainval_index_train_0.8_0.1_Limb.npy
+train_data.shape (53306, 500) train_label.shape (53306,) val_data.shape (6663, 500) val_label.shape (6663,) test_data.shape (6664, 500) test_label.shape (6664,)
+train_val train
+load data from  save_near_index/data_nameLimbK10uselabelFalsepcadim64train_val0.8split_ratio0.1.pkl
+data.shape (66633, 500)
+label (66633,)
+load index trainval_index_train_0.8_0.1_Limb.npy trainval_index_val_0.8_0.1_Limb.npy trainval_index_test_0.8_0.1_Limb.npy from data/trainval_index_train_0.8_0.1_Limb.npy
+train_data.shape (53306, 500) train_label.shape (53306,) val_data.shape (6663, 500) val_label.shape (6663,) test_data.shape (6664, 500) test_label.shape (6664,)
+train_val train
+load data from  save_near_index/data_nameLimbK10uselabelFalsepcadim64train_val0.8split_ratio0.1.pkl
+data.shape (66633, 500)
+label (66633,)
+load index trainval_index_train_0.8_0.1_Limb.npy trainval_index_val_0.8_0.1_Limb.npy trainval_index_test_0.8_0.1_Limb.npy from data/trainval_index_train_0.8_0.1_Limb.npy
+train_data.shape (53306, 500) train_label.shape (53306,) val_data.shape (6663, 500) val_label.shape (6663,) test_data.shape (6664, 500) test_label.shape (6664,)
+train_val val
+load data from  save_near_index/data_nameLimbK10uselabelFalsepcadim64train_val0.8split_ratio0.1.pkl
+data.shape (66633, 500)
+label (66633,)
+load index trainval_index_train_0.8_0.1_Limb.npy trainval_index_val_0.8_0.1_Limb.npy trainval_index_test_0.8_0.1_Limb.npy from data/trainval_index_train_0.8_0.1_Limb.npy
+train_data.shape (53306, 500) train_label.shape (53306,) val_data.shape (6663, 500) val_label.shape (6663,) test_data.shape (6664, 500) test_label.shape (6664,)
+train_val test
+load data from  save_near_index/data_nameLimbK10uselabelFalsepcadim64train_val0.8split_ratio0.1.pkl
+self.training_str step1, epoch 0
+self.training_str step1, epoch 19
+self.training_str step1, epoch 39
+/opt/miniforge3/envs/benchmark/lib/python3.9/site-packages/torch/nn/modules/instancenorm.py:80: UserWarning: input's size at dim=0 does not match num_features. You can silence this warning by not passing in num_features, which is not used because affine=False
+  warnings.warn(f"input's size at dim={feature_dim} does not match num_features. "
+self.training_str step2_s, epoch 59
+leaf L2/7 has 1068 samples
+leaf L3/13 has 508 samples
+leaf L3/12 has 1111 samples
+leaf L1/2 has 1495 samples
+leaf L2/3 has 1446 samples
+leaf L4/11 has 772 samples
+leaf L4/10 has 809 samples
+leaf L3/4 has 1057 samples
+leaf L2/1 has 675 samples
+leaf L2/0 has 1059 samples
+
  0%|          | 0/100 [00:00<?, ?it/s]
 57%|█████▋    | 57/100 [00:00<00:00, 567.84it/s]
+/opt/miniforge3/envs/benchmark/lib/python3.9/site-packages/torch/nn/modules/instancenorm.py:80: UserWarning: input's size at dim=0 does not match num_features. You can silence this warning by not passing in num_features, which is not used because affine=False
+  warnings.warn(f"input's size at dim={feature_dim} does not match num_features. "
+self.training_str step2_s, epoch 79
+leaf L2/7 has 1048 samples
+leaf L2/6 has 1675 samples
+leaf L1/2 has 1522 samples
+leaf L2/3 has 1410 samples
+leaf L4/11 has 781 samples
+leaf L4/10 has 765 samples
+leaf L3/4 has 1052 samples
+leaf L2/1 has 638 samples
+leaf L3/1 has 561 samples
+leaf L3/0 has 548 samples
+
  0%|          | 0/100 [00:00<?, ?it/s]
 51%|█████     | 51/100 [00:00<00:00, 509.10it/s]
+/opt/miniforge3/envs/benchmark/lib/python3.9/site-packages/torch/nn/modules/instancenorm.py:80: UserWarning: input's size at dim=0 does not match num_features. You can silence this warning by not passing in num_features, which is not used because affine=False
+  warnings.warn(f"input's size at dim={feature_dim} does not match num_features. "
+self.training_str step2_s, epoch 99
+leaf L2/7 has 1077 samples
+leaf L3/13 has 493 samples
+leaf L3/12 has 1119 samples
+leaf L1/2 has 1492 samples
+leaf L2/3 has 1516 samples
+leaf L4/11 has 787 samples
+leaf L4/10 has 787 samples
+leaf L3/4 has 986 samples
+leaf L2/1 has 700 samples
+leaf L2/0 has 1043 samples
+
  0%|          | 0/100 [00:00<?, ?it/s]
 91%|█████████ | 91/100 [00:00<00:00, 909.50it/s]
+/opt/miniforge3/envs/benchmark/lib/python3.9/site-packages/torch/nn/modules/instancenorm.py:80: UserWarning: input's size at dim=0 does not match num_features. You can silence this warning by not passing in num_features, which is not used because affine=False
+  warnings.warn(f"input's size at dim={feature_dim} does not match num_features. "
+self.training_str step2_r, epoch 119
+leaf L2/7 has 1040 samples
+leaf L3/13 has 524 samples
+leaf L3/12 has 1155 samples
+leaf L1/2 has 1507 samples
+leaf L2/3 has 1468 samples
+leaf L4/11 has 778 samples
+leaf L4/10 has 765 samples
+leaf L3/4 has 1022 samples
+leaf L2/1 has 718 samples
+leaf L2/0 has 1023 samples
+
  0%|          | 0/100 [00:00<?, ?it/s]
 94%|█████████▍| 94/100 [00:00<00:00, 935.86it/s]
+/opt/miniforge3/envs/benchmark/lib/python3.9/site-packages/torch/nn/modules/instancenorm.py:80: UserWarning: input's size at dim=0 does not match num_features. You can silence this warning by not passing in num_features, which is not used because affine=False
+  warnings.warn(f"input's size at dim={feature_dim} does not match num_features. "
+self.training_str step2_r, epoch 139
+leaf L2/7 has 996 samples
+leaf L3/13 has 510 samples
+leaf L3/12 has 1161 samples
+leaf L1/2 has 1456 samples
+leaf L2/3 has 1451 samples
+leaf L4/11 has 763 samples
+leaf L4/10 has 798 samples
+leaf L3/4 has 1036 samples
+leaf L2/1 has 730 samples
+leaf L2/0 has 1099 samples
+
  0%|          | 0/100 [00:00<?, ?it/s]
+/opt/miniforge3/envs/benchmark/lib/python3.9/site-packages/torch/nn/modules/instancenorm.py:80: UserWarning: input's size at dim=0 does not match num_features. You can silence this warning by not passing in num_features, which is not used because affine=False
+  warnings.warn(f"input's size at dim={feature_dim} does not match num_features. "
+/root/project/hdtree/code_HDTree_review/call_backs/util.py:363: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`). Consider using `matplotlib.pyplot.close()`.
+  fig = plt.figure(figsize=(10, 10))
+self.training_str step2_r, epoch 159
+/root/project/hdtree/code_HDTree_review/call_backs/util.py:747: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`). Consider using `matplotlib.pyplot.close()`.
+  fig, ax = plt.subplots(figsize=(12, 8))  # Adjust according to your needs
+leaf L2/7 has 1045 samples
+leaf L3/13 has 533 samples
+leaf L3/12 has 1053 samples
+leaf L1/2 has 1484 samples
+leaf L2/3 has 1436 samples
+leaf L4/11 has 830 samples
+leaf L4/10 has 755 samples
+leaf L3/4 has 1052 samples
+leaf L2/1 has 736 samples
+leaf L2/0 has 1076 samples
+
  0%|          | 0/100 [00:00<?, ?it/s]
 48%|████▊     | 48/100 [00:00<00:00, 353.82it/s]
 84%|████████▍ | 84/100 [00:00<00:00, 102.94it/s]
+/opt/miniforge3/envs/benchmark/lib/python3.9/site-packages/torch/nn/modules/instancenorm.py:80: UserWarning: input's size at dim=0 does not match num_features. You can silence this warning by not passing in num_features, which is not used because affine=False
+  warnings.warn(f"input's size at dim={feature_dim} does not match num_features. "
+self.training_str step2_r, epoch 179
+leaf L2/7 has 998 samples
+leaf L3/13 has 523 samples
+leaf L3/12 has 1114 samples
+leaf L1/2 has 1554 samples
+leaf L2/3 has 1435 samples
+leaf L4/11 has 802 samples
+leaf L4/10 has 802 samples
+leaf L3/4 has 997 samples
+leaf L2/1 has 665 samples
+leaf L2/0 has 1110 samples
+
  0%|          | 0/100 [00:00<?, ?it/s]
 80%|████████  | 80/100 [00:00<00:00, 790.24it/s]
+`Trainer.fit` stopped: `max_epochs=200` reached.
+wandb:                                                                                
+wandb: 
+wandb: Run history:
+wandb:                      epoch ▁▁▁▁▁▂▂▂▃▃▃▃▃▃▄▄▄▄▄▄▄▅▅▅▅▆▆▆▆▇▇▇▇▇▇▇▇▇▇█
+wandb:                   loss_all █▅▅▅▅▅▅▅▅▅▅▅▅▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
+wandb:                  loss_diff ▅▃▁█████▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
+wandb:                   loss_emb █▂▂▂▂▁▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
+wandb:                   loss_lat █▃▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▂▁▁▁▁▁▁▁▁
+wandb:                  loss_rute ▇▃▂▄▆▄▄▁▁▇▆▁▃▃▄▄▃▂▃▂▃▃▁▅▂▂▄▂▁▅▂▁▁█▁▃█▂▅▂
+wandb:                         lr ▆██████▇▇▇▆▆▆▆▆▅▅▄▄▄▄▄▄▃▃▃▂▂▂▂▂▂▂▁▁▁▁▁▁▁
+wandb:            orthogonal_loss ▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
+wandb:               rout/svc_acc ▁▁▁▁▁▁▁
+wandb:                  train_svc ▁████████▇█
+wandb:              train_svc_rbf ▁██████████
+wandb:        trainer/global_step ▁▂▂▂▂▃▃▃▃▃▃▃▄▄▄▄▄▅▅▅▅▅▆▆▆▆▆▆▆▆▇▇▇▇▇▇▇███
+wandb:                 tree/ari_0 ▇▁▆▆▇▆█
+wandb:         tree/cluster_acc_0 ▇▁▇▇▇▆█
+wandb:                  tree/dp_0 ▆▁▅▆▇▇█
+wandb:      tree/log_likelihood_0 ▁▁▁▁▁▁▁
+wandb:                  tree/lp_0 ▄▁▅█▆▅█
+wandb:                 tree/nmi_0 ▆▁▆▇▆▆█
+wandb: tree/reconstruction_loss_0 ▁▁▁▁▁▁▁
+wandb: 
+wandb: Run summary:
+wandb:                      epoch 199
+wandb:                   loss_all 2e-05
+wandb:                  loss_diff 0.24769
+wandb:                   loss_emb 2.01691
+wandb:                   loss_lat 2.00958
+wandb:                  loss_rute 0.28223
+wandb:                         lr 0.0
+wandb:            orthogonal_loss 0
+wandb:               rout/svc_acc 1
+wandb:                  train_svc 0.53921
+wandb:              train_svc_rbf 0.71421
+wandb:        trainer/global_step 532999
+wandb:                 tree/ari_0 0.3861
+wandb:         tree/cluster_acc_0 0.5286
+wandb:                  tree/dp_0 0.41029
+wandb:      tree/log_likelihood_0 0
+wandb:                  tree/lp_0 0.5837
+wandb:                 tree/nmi_0 0.49042
+wandb: tree/reconstruction_loss_0 0
+wandb: 
+wandb: You can sync this run to the cloud by running:
+wandb: wandb sync wandb/wandb/offline-run-20260518_092347-segtfz2w
+wandb: Find logs at: wandb/wandb/offline-run-20260518_092347-segtfz2w/logs
+self.training_str step2_r, epoch 199
+==== END 2026-05-18_12:11:30 limb_sweep_i10_bs1000_ex0.5_nu0.3_gpu6 status=0 worker=extra_g6 ====