[run] ogbg-molbace view=gin compute=rrog-act mode=stream T=1 ns=3 seed=0 device=cuda:3
python3 rrog/train_ogb_graphprop.py --dataset ogbg-molbace --view gin --compute rrog-act --T 1 --n_sup 3 --hidden 128 --bs 128 --epochs 100 --eval_every 10 --agg_layers 5 --compute_layers 2 --seed 0 --lam_q 0.1 --halt_max_steps 8 --halt_min_steps 2 --halt_target exact --halt_loss_threshold 0.2 --halt_exploration_prob 0.1 --act_train_mode stream --q_warmup_epochs 0 --device cuda:3 --num_workers 0
Downloading http://snap.stanford.edu/ogb/data/graphproppred/csv_mol_download/bace.zip

  0%|          | 0/2 [00:00<?, ?it/s]
Downloaded 0.00 GB:   0%|          | 0/2 [00:00<?, ?it/s]
Downloaded 0.00 GB:  50%|█████     | 1/2 [00:00<00:00,  4.82it/s]
Downloaded 0.00 GB:  50%|█████     | 1/2 [00:00<00:00,  4.82it/s]
Downloaded 0.00 GB: 100%|██████████| 2/2 [00:00<00:00,  9.64it/s]
Processing...
Extracting /home/yurenh2/rrog-gnn-runner/data/ogb/bace.zip
Loading necessary files...
This might take a while.
Processing graphs...

  0%|          | 0/1513 [00:00<?, ?it/s]
100%|██████████| 1513/1513 [00:00<00:00, 163022.63it/s]
Converting graphs into PyG objects...

  0%|          | 0/1513 [00:00<?, ?it/s]
100%|██████████| 1513/1513 [00:00<00:00, 85215.28it/s]
/home/yurenh2/miniconda3/lib/python3.13/site-packages/ogb/graphproppred/dataset_pyg.py:156: UserWarning: The given NumPy array is not writable, and PyTorch does not support non-writable tensors. This means writing to this tensor will result in undefined behavior. You may want to copy the array to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:213.)
  g.y = torch.from_numpy(graph_label[i]).view(1,-1).to(torch.long)
Done!
Saving...
ep10 val_rocauc=0.52894 val_adapt_rocauc=0.53736 adapt_steps=2.04 halt=0.34 train_steps=2.42
ep20 val_rocauc=0.61941 val_adapt_rocauc=0.62161 adapt_steps=2.00 halt=0.46 train_steps=1.68
ep30 val_rocauc=0.53260 val_adapt_rocauc=0.65385 adapt_steps=2.09 halt=0.44 train_steps=1.75
ep40 val_rocauc=0.61319 val_adapt_rocauc=0.66703 adapt_steps=2.00 halt=0.47 train_steps=1.66
ep50 val_rocauc=0.59048 val_adapt_rocauc=0.58315 adapt_steps=2.00 halt=0.47 train_steps=1.58
ep60 val_rocauc=0.66300 val_adapt_rocauc=0.69487 adapt_steps=2.09 halt=0.47 train_steps=1.60
ep70 val_rocauc=0.59011 val_adapt_rocauc=0.62601 adapt_steps=2.00 halt=0.48 train_steps=1.55
ep80 val_rocauc=0.62308 val_adapt_rocauc=0.64579 adapt_steps=2.00 halt=0.47 train_steps=1.59
ep90 val_rocauc=0.60000 val_adapt_rocauc=0.61868 adapt_steps=2.00 halt=0.48 train_steps=1.57
ep100 val_rocauc=0.60549 val_adapt_rocauc=0.61355 adapt_steps=2.01 halt=0.47 train_steps=1.57
[ogbg-molbace_gin_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0] best_ep=60 val={'rocauc': 0.6630036630036631} test={'rocauc': 0.7275256477134411} adaptive={'rocauc': 0.7369153190749435} steps=2.0526315789473686
  wrote /home/yurenh2/rrog-gnn-runner/runs/ogbg-molbace_gin_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0.json
[run] ogbg-molbace view=gcn compute=rrog-act mode=stream T=1 ns=3 seed=0 device=cuda:3
python3 rrog/train_ogb_graphprop.py --dataset ogbg-molbace --view gcn --compute rrog-act --T 1 --n_sup 3 --hidden 128 --bs 128 --epochs 100 --eval_every 10 --agg_layers 5 --compute_layers 2 --seed 0 --lam_q 0.1 --halt_max_steps 8 --halt_min_steps 2 --halt_target exact --halt_loss_threshold 0.2 --halt_exploration_prob 0.1 --act_train_mode stream --q_warmup_epochs 0 --device cuda:3 --num_workers 0
ep10 val_rocauc=0.66154 val_adapt_rocauc=0.68828 adapt_steps=2.16 halt=0.45 train_steps=1.71
ep20 val_rocauc=0.65604 val_adapt_rocauc=0.70476 adapt_steps=2.08 halt=0.47 train_steps=1.66
ep30 val_rocauc=0.57875 val_adapt_rocauc=0.62674 adapt_steps=2.04 halt=0.44 train_steps=1.81
ep40 val_rocauc=0.56337 val_adapt_rocauc=0.58608 adapt_steps=2.02 halt=0.47 train_steps=1.59
ep50 val_rocauc=0.59084 val_adapt_rocauc=0.62527 adapt_steps=2.01 halt=0.46 train_steps=1.60
ep60 val_rocauc=0.57289 val_adapt_rocauc=0.62161 adapt_steps=2.01 halt=0.48 train_steps=1.58
ep70 val_rocauc=0.57399 val_adapt_rocauc=0.60659 adapt_steps=2.01 halt=0.48 train_steps=1.55
ep80 val_rocauc=0.54762 val_adapt_rocauc=0.59560 adapt_steps=2.01 halt=0.48 train_steps=1.59
ep90 val_rocauc=0.55861 val_adapt_rocauc=0.59634 adapt_steps=2.01 halt=0.48 train_steps=1.57
ep100 val_rocauc=0.56264 val_adapt_rocauc=0.60403 adapt_steps=2.00 halt=0.47 train_steps=1.57
[ogbg-molbace_gcn_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0] best_ep=10 val={'rocauc': 0.6615384615384615} test={'rocauc': 0.4418362023995827} adaptive={'rocauc': 0.5362545644235784} steps=2.8421052631578947
  wrote /home/yurenh2/rrog-gnn-runner/runs/ogbg-molbace_gcn_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0.json
[run] ogbg-molbace view=sgc compute=rrog-act mode=stream T=1 ns=3 seed=0 device=cuda:3
python3 rrog/train_ogb_graphprop.py --dataset ogbg-molbace --view sgc --compute rrog-act --T 1 --n_sup 3 --hidden 128 --bs 128 --epochs 100 --eval_every 10 --agg_layers 5 --compute_layers 2 --seed 0 --lam_q 0.1 --halt_max_steps 8 --halt_min_steps 2 --halt_target exact --halt_loss_threshold 0.2 --halt_exploration_prob 0.1 --act_train_mode stream --q_warmup_epochs 0 --device cuda:3 --num_workers 0
ep10 val_rocauc=0.54505 val_adapt_rocauc=0.57875 adapt_steps=2.06 halt=0.44 train_steps=1.70
ep20 val_rocauc=0.58388 val_adapt_rocauc=0.63223 adapt_steps=2.44 halt=0.47 train_steps=1.62
ep30 val_rocauc=0.58168 val_adapt_rocauc=0.58205 adapt_steps=2.07 halt=0.47 train_steps=1.58
ep40 val_rocauc=0.56923 val_adapt_rocauc=0.66410 adapt_steps=2.00 halt=0.48 train_steps=1.58
ep50 val_rocauc=0.59011 val_adapt_rocauc=0.58022 adapt_steps=2.07 halt=0.47 train_steps=1.57
ep60 val_rocauc=0.60037 val_adapt_rocauc=0.59194 adapt_steps=2.03 halt=0.48 train_steps=1.60
ep70 val_rocauc=0.64396 val_adapt_rocauc=0.62271 adapt_steps=2.00 halt=0.48 train_steps=1.55
ep80 val_rocauc=0.63993 val_adapt_rocauc=0.62711 adapt_steps=2.00 halt=0.48 train_steps=1.59
ep90 val_rocauc=0.63883 val_adapt_rocauc=0.62967 adapt_steps=2.00 halt=0.48 train_steps=1.57
ep100 val_rocauc=0.64322 val_adapt_rocauc=0.62491 adapt_steps=2.00 halt=0.47 train_steps=1.57
[ogbg-molbace_sgc_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0] best_ep=70 val={'rocauc': 0.643956043956044} test={'rocauc': 0.7117023126412798} adaptive={'rocauc': 0.7499565292992523} steps=2.013157894736842
  wrote /home/yurenh2/rrog-gnn-runner/runs/ogbg-molbace_sgc_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0.json
[run] ogbg-molbace view=tag compute=rrog-act mode=stream T=1 ns=3 seed=0 device=cuda:3
python3 rrog/train_ogb_graphprop.py --dataset ogbg-molbace --view tag --compute rrog-act --T 1 --n_sup 3 --hidden 128 --bs 128 --epochs 100 --eval_every 10 --agg_layers 5 --compute_layers 2 --seed 0 --lam_q 0.1 --halt_max_steps 8 --halt_min_steps 2 --halt_target exact --halt_loss_threshold 0.2 --halt_exploration_prob 0.1 --act_train_mode stream --q_warmup_epochs 0 --device cuda:3 --num_workers 0
ep10 val_rocauc=0.64799 val_adapt_rocauc=0.66227 adapt_steps=2.66 halt=0.44 train_steps=1.78
ep20 val_rocauc=0.64579 val_adapt_rocauc=0.62674 adapt_steps=2.08 halt=0.43 train_steps=1.80
ep30 val_rocauc=0.65458 val_adapt_rocauc=0.66300 adapt_steps=2.01 halt=0.47 train_steps=1.62
ep40 val_rocauc=0.61941 val_adapt_rocauc=0.60806 adapt_steps=2.01 halt=0.46 train_steps=1.67
ep50 val_rocauc=0.58645 val_adapt_rocauc=0.64908 adapt_steps=2.00 halt=0.47 train_steps=1.58
ep60 val_rocauc=0.57399 val_adapt_rocauc=0.57363 adapt_steps=2.00 halt=0.48 train_steps=1.58
ep70 val_rocauc=0.62161 val_adapt_rocauc=0.63114 adapt_steps=2.00 halt=0.48 train_steps=1.55
ep80 val_rocauc=0.62930 val_adapt_rocauc=0.64286 adapt_steps=2.00 halt=0.48 train_steps=1.59
ep90 val_rocauc=0.63553 val_adapt_rocauc=0.63223 adapt_steps=2.00 halt=0.48 train_steps=1.57
ep100 val_rocauc=0.63736 val_adapt_rocauc=0.63480 adapt_steps=2.00 halt=0.47 train_steps=1.57
[ogbg-molbace_tag_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0] best_ep=30 val={'rocauc': 0.6545787545787546} test={'rocauc': 0.8191618848895844} adaptive={'rocauc': 0.8142931664058426} steps=2.111842105263158
  wrote /home/yurenh2/rrog-gnn-runner/runs/ogbg-molbace_tag_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0.json
[run] ogbg-molbace view=graphconv compute=rrog-act mode=stream T=1 ns=3 seed=0 device=cuda:3
python3 rrog/train_ogb_graphprop.py --dataset ogbg-molbace --view graphconv --compute rrog-act --T 1 --n_sup 3 --hidden 128 --bs 128 --epochs 100 --eval_every 10 --agg_layers 5 --compute_layers 2 --seed 0 --lam_q 0.1 --halt_max_steps 8 --halt_min_steps 2 --halt_target exact --halt_loss_threshold 0.2 --halt_exploration_prob 0.1 --act_train_mode stream --q_warmup_epochs 0 --device cuda:3 --num_workers 0
ep10 val_rocauc=0.67802 val_adapt_rocauc=0.70293 adapt_steps=2.01 halt=0.44 train_steps=1.78
ep20 val_rocauc=0.55751 val_adapt_rocauc=0.55348 adapt_steps=2.03 halt=0.47 train_steps=1.61
ep30 val_rocauc=0.66337 val_adapt_rocauc=0.66300 adapt_steps=2.01 halt=0.47 train_steps=1.61
ep40 val_rocauc=0.68901 val_adapt_rocauc=0.67766 adapt_steps=2.09 halt=0.46 train_steps=1.66
ep50 val_rocauc=0.62857 val_adapt_rocauc=0.71502 adapt_steps=2.00 halt=0.47 train_steps=1.58
ep60 val_rocauc=0.66703 val_adapt_rocauc=0.67033 adapt_steps=2.00 halt=0.48 train_steps=1.58
ep70 val_rocauc=0.69267 val_adapt_rocauc=0.69780 adapt_steps=2.00 halt=0.48 train_steps=1.55
ep80 val_rocauc=0.67766 val_adapt_rocauc=0.66374 adapt_steps=2.00 halt=0.48 train_steps=1.59
ep90 val_rocauc=0.67143 val_adapt_rocauc=0.65238 adapt_steps=2.00 halt=0.48 train_steps=1.57
ep100 val_rocauc=0.68095 val_adapt_rocauc=0.65311 adapt_steps=2.00 halt=0.47 train_steps=1.57
[ogbg-molbace_graphconv_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0] best_ep=70 val={'rocauc': 0.6926739926739928} test={'rocauc': 0.6736219787862981} adaptive={'rocauc': 0.734828725439054} steps=2.0
  wrote /home/yurenh2/rrog-gnn-runner/runs/ogbg-molbace_graphconv_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0.json
[run] ogbg-molbace view=appnp compute=rrog-act mode=stream T=1 ns=3 seed=0 device=cuda:3
python3 rrog/train_ogb_graphprop.py --dataset ogbg-molbace --view appnp --compute rrog-act --T 1 --n_sup 3 --hidden 128 --bs 128 --epochs 100 --eval_every 10 --agg_layers 5 --compute_layers 2 --seed 0 --lam_q 0.1 --halt_max_steps 8 --halt_min_steps 2 --halt_target exact --halt_loss_threshold 0.2 --halt_exploration_prob 0.1 --act_train_mode stream --q_warmup_epochs 0 --device cuda:3 --num_workers 0
ep10 val_rocauc=0.68022 val_adapt_rocauc=0.67143 adapt_steps=2.11 halt=0.48 train_steps=1.58
ep20 val_rocauc=0.64139 val_adapt_rocauc=0.67692 adapt_steps=2.05 halt=0.48 train_steps=1.60
ep30 val_rocauc=0.63516 val_adapt_rocauc=0.63077 adapt_steps=2.02 halt=0.47 train_steps=1.57
ep40 val_rocauc=0.59158 val_adapt_rocauc=0.65128 adapt_steps=2.00 halt=0.48 train_steps=1.59
ep50 val_rocauc=0.51758 val_adapt_rocauc=0.55788 adapt_steps=2.01 halt=0.47 train_steps=1.56
ep60 val_rocauc=0.63626 val_adapt_rocauc=0.63700 adapt_steps=2.00 halt=0.48 train_steps=1.58
ep70 val_rocauc=0.56300 val_adapt_rocauc=0.62271 adapt_steps=2.00 halt=0.48 train_steps=1.55
ep80 val_rocauc=0.55751 val_adapt_rocauc=0.62967 adapt_steps=2.00 halt=0.48 train_steps=1.59
ep90 val_rocauc=0.55495 val_adapt_rocauc=0.61795 adapt_steps=2.00 halt=0.48 train_steps=1.57
ep100 val_rocauc=0.54652 val_adapt_rocauc=0.61722 adapt_steps=2.00 halt=0.47 train_steps=1.57
[ogbg-molbace_appnp_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0] best_ep=10 val={'rocauc': 0.6802197802197802} test={'rocauc': 0.6833594157537818} adaptive={'rocauc': 0.7826464962615197} steps=2.0921052631578947
  wrote /home/yurenh2/rrog-gnn-runner/runs/ogbg-molbace_appnp_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0.json
[run] ogbg-molbace view=pna compute=rrog-act mode=stream T=1 ns=3 seed=0 device=cuda:3
python3 rrog/train_ogb_graphprop.py --dataset ogbg-molbace --view pna --compute rrog-act --T 1 --n_sup 3 --hidden 128 --bs 128 --epochs 100 --eval_every 10 --agg_layers 5 --compute_layers 2 --seed 0 --lam_q 0.1 --halt_max_steps 8 --halt_min_steps 2 --halt_target exact --halt_loss_threshold 0.2 --halt_exploration_prob 0.1 --act_train_mode stream --q_warmup_epochs 0 --device cuda:3 --num_workers 0
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep10 val_rocauc=0.56447 val_adapt_rocauc=0.56557 adapt_steps=2.00 halt=0.35 train_steps=2.27
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep20 val_rocauc=0.65751 val_adapt_rocauc=0.67289 adapt_steps=2.16 halt=0.38 train_steps=2.12
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep30 val_rocauc=0.65897 val_adapt_rocauc=0.65018 adapt_steps=2.56 halt=0.40 train_steps=1.88
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep40 val_rocauc=0.45604 val_adapt_rocauc=0.66996 adapt_steps=2.13 halt=0.46 train_steps=1.68
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep50 val_rocauc=0.49560 val_adapt_rocauc=0.68791 adapt_steps=2.13 halt=0.44 train_steps=1.75
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep60 val_rocauc=0.46227 val_adapt_rocauc=0.68132 adapt_steps=2.05 halt=0.47 train_steps=1.65
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep70 val_rocauc=0.64432 val_adapt_rocauc=0.67912 adapt_steps=2.00 halt=0.47 train_steps=1.57
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep80 val_rocauc=0.60220 val_adapt_rocauc=0.64359 adapt_steps=2.04 halt=0.47 train_steps=1.60
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep90 val_rocauc=0.61136 val_adapt_rocauc=0.68535 adapt_steps=2.01 halt=0.47 train_steps=1.62
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='min')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
/home/yurenh2/miniconda3/lib/python3.13/site-packages/torch_geometric/utils/_scatter.py:91: UserWarning: The usage of `scatter(reduce='max')` can be accelerated via the 'torch-scatter' package, but it was not found
  warnings.warn(
ep100 val_rocauc=0.61978 val_adapt_rocauc=0.68974 adapt_steps=2.01 halt=0.46 train_steps=1.61
[ogbg-molbace_pna_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0] best_ep=30 val={'rocauc': 0.658974358974359} test={'rocauc': 0.7480438184663536} adaptive={'rocauc': 0.7546513649800034} steps=2.6315789473684212
  wrote /home/yurenh2/rrog-gnn-runner/runs/ogbg-molbace_pna_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0.json
[run] ogbg-molbace view=resgated compute=rrog-act mode=stream T=1 ns=3 seed=0 device=cuda:3
python3 rrog/train_ogb_graphprop.py --dataset ogbg-molbace --view resgated --compute rrog-act --T 1 --n_sup 3 --hidden 128 --bs 128 --epochs 100 --eval_every 10 --agg_layers 5 --compute_layers 2 --seed 0 --lam_q 0.1 --halt_max_steps 8 --halt_min_steps 2 --halt_target exact --halt_loss_threshold 0.2 --halt_exploration_prob 0.1 --act_train_mode stream --q_warmup_epochs 0 --device cuda:3 --num_workers 0
ep10 val_rocauc=0.61282 val_adapt_rocauc=0.59304 adapt_steps=2.04 halt=0.43 train_steps=1.89
ep20 val_rocauc=0.68938 val_adapt_rocauc=0.61062 adapt_steps=2.04 halt=0.46 train_steps=1.69
ep30 val_rocauc=0.66740 val_adapt_rocauc=0.70366 adapt_steps=2.12 halt=0.47 train_steps=1.61
ep40 val_rocauc=0.64212 val_adapt_rocauc=0.67582 adapt_steps=2.01 halt=0.47 train_steps=1.67
ep50 val_rocauc=0.55604 val_adapt_rocauc=0.57802 adapt_steps=2.00 halt=0.46 train_steps=1.61
ep60 val_rocauc=0.69414 val_adapt_rocauc=0.69597 adapt_steps=2.00 halt=0.48 train_steps=1.57
ep70 val_rocauc=0.61026 val_adapt_rocauc=0.63590 adapt_steps=2.00 halt=0.48 train_steps=1.55
ep80 val_rocauc=0.66410 val_adapt_rocauc=0.66300 adapt_steps=2.00 halt=0.48 train_steps=1.59
ep90 val_rocauc=0.65934 val_adapt_rocauc=0.64835 adapt_steps=2.00 halt=0.48 train_steps=1.57
ep100 val_rocauc=0.66850 val_adapt_rocauc=0.65128 adapt_steps=2.00 halt=0.47 train_steps=1.57
[ogbg-molbace_resgated_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0] best_ep=60 val={'rocauc': 0.6941391941391942} test={'rocauc': 0.7231785776386715} adaptive={'rocauc': 0.7640410363415059} steps=2.0
  wrote /home/yurenh2/rrog-gnn-runner/runs/ogbg-molbace_resgated_rrog-act_T1_ns3_stream_hm8_hmin2_exact_lq0.1_hex0.1_qw0_h128_e100_s0.json