Skip to content

Instantly share code, notes, and snippets.

@liyaguang

liyaguang/metr-la.log Secret

Created March 8, 2023 05:25
Show Gist options
  • Save liyaguang/f689b004f6aa04f6767bbadaeb808d47 to your computer and use it in GitHub Desktop.
Save liyaguang/f689b004f6aa04f6767bbadaeb808d47 to your computer and use it in GitHub Desktop.
Download ZIP
Training Log of DCRNN
Raw
metr-la.log
python dcrnn_train.py --config_filename=data/model/dcrnn_la.yaml
2020-05-30 22:49:17,853 - model.dcrnn_supervisor - INFO - Log directory: data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/
2020-05-30 22:49:17,855 - model.dcrnn_supervisor - INFO - {'base_dir': 'data/model' 'log_level': 'INFO' 'data': {'batch_size': 64 'dataset_dir': 'data/PEMS-BAY' 'test_batch_size': 64 'val_batch_size': 64 'graph_pkl_filename': 'data/sensor_graph/adj_mx_bay.pkl'} 'model': {'cl_decay_steps': 2000 'filter_type': 'dual_random_walk' 'horizon': 12 'input_dim': 2 'l1_decay': 0 'max_diffusion_step': 2 'num_nodes': 325 'num_rnn_layers': 2 'output_dim': 1 'rnn_units': 64 'seq_len': 12 'use_curriculum_learning': True} 'train': {'base_lr': 0.01 'dropout': 0 'epoch': 0 'epochs': 100 'epsilon': 0.001 'global_step': 0 'lr_decay_ratio': 0.1 'max_grad_norm': 5 'max_to_keep': 100 'min_learning_rate': 2e-06 'optimizer': 'adam' 'patience': 50 'steps': [20 30 40 50] 'test_every_n_epochs': 10}}
2020-05-30 22:49:41,189 - model.dcrnn_supervisor - INFO - ('x_train' (36465 12 325 2))
2020-05-30 22:49:41,191 - model.dcrnn_supervisor - INFO - ('y_train' (36465 12 325 2))
2020-05-30 22:49:41,192 - model.dcrnn_supervisor - INFO - ('x_val' (5209 12 325 2))
2020-05-30 22:49:41,192 - model.dcrnn_supervisor - INFO - ('y_val' (5209 12 325 2))
2020-05-30 22:49:41,194 - model.dcrnn_supervisor - INFO - ('x_test' (10419 12 325 2))
2020-05-30 22:49:41,194 - model.dcrnn_supervisor - INFO - ('y_test' (10419 12 325 2))
2020-05-30 22:49:56,381 - model.dcrnn_supervisor - INFO - Total number of trainable parameters: 372352
2020-05-30 22:49:57,340 - model.dcrnn_supervisor - INFO - Start training ...
2020-05-30 23:01:44,351 - model.dcrnn_supervisor - INFO - Epoch [0/100] (570) train_mae: 1.0133 val_mae: 2.5961 lr:0.010000 706.7s
2020-05-30 23:01:44,703 - model.dcrnn_supervisor - INFO - Val loss decrease from inf to 2.5961 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-2.5961-570
2020-05-30 23:13:23,461 - model.dcrnn_supervisor - INFO - Epoch [1/100] (1140) train_mae: 0.8706 val_mae: 2.2480 lr:0.010000 698.8s
2020-05-30 23:13:23,490 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.5961 to 2.2480 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-2.2480-1140
2020-05-30 23:24:59,156 - model.dcrnn_supervisor - INFO - Epoch [2/100] (1710) train_mae: 0.8587 val_mae: 2.1745 lr:0.010000 695.7s
2020-05-30 23:24:59,184 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.2480 to 2.1745 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-2.1745-1710
2020-05-30 23:36:33,378 - model.dcrnn_supervisor - INFO - Epoch [3/100] (2280) train_mae: 0.8502 val_mae: 2.0298 lr:0.010000 694.2s
2020-05-30 23:36:33,406 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.1745 to 2.0298 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-2.0298-2280
2020-05-30 23:48:10,246 - model.dcrnn_supervisor - INFO - Epoch [4/100] (2850) train_mae: 0.8438 val_mae: 2.0020 lr:0.010000 696.8s
2020-05-30 23:48:10,279 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.0298 to 2.0020 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-2.0020-2850
2020-05-30 23:59:45,598 - model.dcrnn_supervisor - INFO - Epoch [5/100] (3420) train_mae: 0.8364 val_mae: 1.9617 lr:0.010000 695.3s
2020-05-30 23:59:45,626 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.0020 to 1.9617 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.9617-3420
2020-05-31 00:11:19,775 - model.dcrnn_supervisor - INFO - Epoch [6/100] (3990) train_mae: 0.8317 val_mae: 1.9593 lr:0.010000 694.1s
2020-05-31 00:11:19,803 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.9617 to 1.9593 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.9593-3990
2020-05-31 00:22:53,858 - model.dcrnn_supervisor - INFO - Epoch [7/100] (4560) train_mae: 0.8280 val_mae: 1.9376 lr:0.010000 694.1s
2020-05-31 00:22:53,886 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.9593 to 1.9376 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.9376-4560
2020-05-31 00:34:27,898 - model.dcrnn_supervisor - INFO - Epoch [8/100] (5130) train_mae: 0.8248 val_mae: 1.9592 lr:0.010000 694.0s
2020-05-31 00:46:02,021 - model.dcrnn_supervisor - INFO - Epoch [9/100] (5700) train_mae: 0.8212 val_mae: 1.9261 lr:0.010000 694.1s
2020-05-31 00:47:25,688 - model.dcrnn_supervisor - INFO - Horizon 01 0.86 0.0164 1.56
2020-05-31 00:47:26,051 - model.dcrnn_supervisor - INFO - Horizon 02 1.15 0.0229 2.31
2020-05-31 00:47:26,411 - model.dcrnn_supervisor - INFO - Horizon 03 1.36 0.0280 2.92
2020-05-31 00:47:26,777 - model.dcrnn_supervisor - INFO - Horizon 04 1.53 0.0323 3.42
2020-05-31 00:47:27,147 - model.dcrnn_supervisor - INFO - Horizon 05 1.67 0.0361 3.83
2020-05-31 00:47:27,508 - model.dcrnn_supervisor - INFO - Horizon 06 1.80 0.0395 4.19
2020-05-31 00:47:27,872 - model.dcrnn_supervisor - INFO - Horizon 07 1.91 0.0426 4.50
2020-05-31 00:47:28,239 - model.dcrnn_supervisor - INFO - Horizon 08 2.02 0.0455 4.78
2020-05-31 00:47:28,607 - model.dcrnn_supervisor - INFO - Horizon 09 2.12 0.0482 5.03
2020-05-31 00:47:28,977 - model.dcrnn_supervisor - INFO - Horizon 10 2.22 0.0507 5.26
2020-05-31 00:47:29,348 - model.dcrnn_supervisor - INFO - Horizon 11 2.32 0.0531 5.48
2020-05-31 00:47:29,708 - model.dcrnn_supervisor - INFO - Horizon 12 2.41 0.0555 5.70
2020-05-31 00:47:29,787 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.9376 to 1.9261 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.9261-5700
2020-05-31 00:59:03,967 - model.dcrnn_supervisor - INFO - Epoch [10/100] (6270) train_mae: 0.8176 val_mae: 1.8588 lr:0.010000 694.2s
2020-05-31 00:59:03,995 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.9261 to 1.8588 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.8588-6270
2020-05-31 01:10:38,163 - model.dcrnn_supervisor - INFO - Epoch [11/100] (6840) train_mae: 0.8164 val_mae: 1.8513 lr:0.010000 694.2s
2020-05-31 01:10:38,188 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.8588 to 1.8513 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.8513-6840
2020-05-31 01:22:12,281 - model.dcrnn_supervisor - INFO - Epoch [12/100] (7410) train_mae: 0.8158 val_mae: 1.8384 lr:0.010000 694.1s
2020-05-31 01:22:12,311 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.8513 to 1.8384 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.8384-7410
2020-05-31 01:33:46,520 - model.dcrnn_supervisor - INFO - Epoch [13/100] (7980) train_mae: 0.8164 val_mae: 1.9325 lr:0.010000 694.2s
2020-05-31 01:45:20,886 - model.dcrnn_supervisor - INFO - Epoch [14/100] (8550) train_mae: 0.8174 val_mae: 1.8946 lr:0.010000 694.4s
2020-05-31 01:56:55,208 - model.dcrnn_supervisor - INFO - Epoch [15/100] (9120) train_mae: 0.8171 val_mae: 1.7888 lr:0.010000 694.3s
2020-05-31 01:56:55,233 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.8384 to 1.7888 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.7888-9120
2020-05-31 02:08:29,587 - model.dcrnn_supervisor - INFO - Epoch [16/100] (9690) train_mae: 0.8205 val_mae: 1.8581 lr:0.010000 694.4s
2020-05-31 02:20:03,909 - model.dcrnn_supervisor - INFO - Epoch [17/100] (10260) train_mae: 0.8233 val_mae: 1.8907 lr:0.010000 694.3s
2020-05-31 02:31:38,231 - model.dcrnn_supervisor - INFO - Epoch [18/100] (10830) train_mae: 0.8355 val_mae: 1.7662 lr:0.010000 694.3s
2020-05-31 02:31:38,260 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.7888 to 1.7662 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.7662-10830
2020-05-31 02:43:12,728 - model.dcrnn_supervisor - INFO - Epoch [19/100] (11400) train_mae: 0.8397 val_mae: 1.8744 lr:0.010000 694.5s
2020-05-31 02:44:34,607 - model.dcrnn_supervisor - INFO - Horizon 01 0.86 0.0167 1.54
2020-05-31 02:44:34,978 - model.dcrnn_supervisor - INFO - Horizon 02 1.17 0.0243 2.29
2020-05-31 02:44:35,349 - model.dcrnn_supervisor - INFO - Horizon 03 1.39 0.0306 2.91
2020-05-31 02:44:35,721 - model.dcrnn_supervisor - INFO - Horizon 04 1.56 0.0359 3.41
2020-05-31 02:44:36,096 - model.dcrnn_supervisor - INFO - Horizon 05 1.70 0.0405 3.82
2020-05-31 02:44:36,464 - model.dcrnn_supervisor - INFO - Horizon 06 1.82 0.0445 4.15
2020-05-31 02:44:36,832 - model.dcrnn_supervisor - INFO - Horizon 07 1.92 0.0480 4.43
2020-05-31 02:44:37,199 - model.dcrnn_supervisor - INFO - Horizon 08 2.01 0.0509 4.66
2020-05-31 02:44:37,573 - model.dcrnn_supervisor - INFO - Horizon 09 2.09 0.0535 4.85
2020-05-31 02:44:37,945 - model.dcrnn_supervisor - INFO - Horizon 10 2.15 0.0556 5.01
2020-05-31 02:44:38,335 - model.dcrnn_supervisor - INFO - Horizon 11 2.21 0.0575 5.15
2020-05-31 02:44:38,700 - model.dcrnn_supervisor - INFO - Horizon 12 2.27 0.0592 5.27
2020-05-31 02:56:13,233 - model.dcrnn_supervisor - INFO - Epoch [20/100] (11970) train_mae: 0.8391 val_mae: 1.7241 lr:0.001000 694.5s
2020-05-31 02:56:13,261 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.7662 to 1.7241 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.7241-11970
2020-05-31 03:07:47,735 - model.dcrnn_supervisor - INFO - Epoch [21/100] (12540) train_mae: 0.8473 val_mae: 1.7209 lr:0.001000 694.5s
2020-05-31 03:07:47,764 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.7241 to 1.7209 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.7209-12540
2020-05-31 03:19:22,379 - model.dcrnn_supervisor - INFO - Epoch [22/100] (13110) train_mae: 0.8650 val_mae: 1.7055 lr:0.001000 694.6s
2020-05-31 03:19:22,409 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.7209 to 1.7055 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.7055-13110
2020-05-31 03:30:57,044 - model.dcrnn_supervisor - INFO - Epoch [23/100] (13680) train_mae: 0.8876 val_mae: 1.7228 lr:0.001000 694.6s
2020-05-31 03:42:31,947 - model.dcrnn_supervisor - INFO - Epoch [24/100] (14250) train_mae: 0.9127 val_mae: 1.7088 lr:0.001000 694.9s
2020-05-31 03:54:06,637 - model.dcrnn_supervisor - INFO - Epoch [25/100] (14820) train_mae: 0.9396 val_mae: 1.6984 lr:0.001000 694.7s
2020-05-31 03:54:06,671 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.7055 to 1.6984 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6984-14820
2020-05-31 04:05:41,558 - model.dcrnn_supervisor - INFO - Epoch [26/100] (15390) train_mae: 0.9807 val_mae: 1.6902 lr:0.001000 694.9s
2020-05-31 04:05:41,587 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6984 to 1.6902 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6902-15390
2020-05-31 04:17:16,481 - model.dcrnn_supervisor - INFO - Epoch [27/100] (15960) train_mae: 1.0189 val_mae: 1.7730 lr:0.001000 694.9s
2020-05-31 04:28:51,415 - model.dcrnn_supervisor - INFO - Epoch [28/100] (16530) train_mae: 1.0709 val_mae: 1.6733 lr:0.001000 694.9s
2020-05-31 04:28:51,443 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6902 to 1.6733 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6733-16530
2020-05-31 04:40:26,421 - model.dcrnn_supervisor - INFO - Epoch [29/100] (17100) train_mae: 1.1239 val_mae: 1.6849 lr:0.001000 695.0s
2020-05-31 04:41:48,314 - model.dcrnn_supervisor - INFO - Horizon 01 0.85 0.0164 1.54
2020-05-31 04:41:48,684 - model.dcrnn_supervisor - INFO - Horizon 02 1.13 0.0225 2.23
2020-05-31 04:41:49,059 - model.dcrnn_supervisor - INFO - Horizon 03 1.32 0.0272 2.78
2020-05-31 04:41:49,432 - model.dcrnn_supervisor - INFO - Horizon 04 1.47 0.0311 3.20
2020-05-31 04:41:49,796 - model.dcrnn_supervisor - INFO - Horizon 05 1.58 0.0343 3.54
2020-05-31 04:41:50,171 - model.dcrnn_supervisor - INFO - Horizon 06 1.68 0.0370 3.81
2020-05-31 04:41:50,539 - model.dcrnn_supervisor - INFO - Horizon 07 1.76 0.0393 4.03
2020-05-31 04:41:50,901 - model.dcrnn_supervisor - INFO - Horizon 08 1.83 0.0413 4.22
2020-05-31 04:41:51,278 - model.dcrnn_supervisor - INFO - Horizon 09 1.89 0.0430 4.37
2020-05-31 04:41:51,643 - model.dcrnn_supervisor - INFO - Horizon 10 1.95 0.0445 4.51
2020-05-31 04:41:52,009 - model.dcrnn_supervisor - INFO - Horizon 11 2.00 0.0457 4.62
2020-05-31 04:41:52,381 - model.dcrnn_supervisor - INFO - Horizon 12 2.05 0.0469 4.73
2020-05-31 04:53:27,360 - model.dcrnn_supervisor - INFO - Epoch [30/100] (17670) train_mae: 1.1497 val_mae: 1.6515 lr:0.000100 694.9s
2020-05-31 04:53:27,389 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6733 to 1.6515 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6515-17670
2020-05-31 05:05:02,356 - model.dcrnn_supervisor - INFO - Epoch [31/100] (18240) train_mae: 1.1995 val_mae: 1.6483 lr:0.000100 695.0s
2020-05-31 05:05:02,384 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6515 to 1.6483 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6483-18240
2020-05-31 05:16:37,276 - model.dcrnn_supervisor - INFO - Epoch [32/100] (18810) train_mae: 1.2415 val_mae: 1.6540 lr:0.000100 694.9s
2020-05-31 05:28:12,280 - model.dcrnn_supervisor - INFO - Epoch [33/100] (19380) train_mae: 1.2917 val_mae: 1.6449 lr:0.000100 695.0s
2020-05-31 05:28:12,309 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6483 to 1.6449 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6449-19380
2020-05-31 05:39:47,331 - model.dcrnn_supervisor - INFO - Epoch [34/100] (19950) train_mae: 1.3240 val_mae: 1.6436 lr:0.000100 695.0s
2020-05-31 05:39:47,361 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6449 to 1.6436 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6436-19950
2020-05-31 05:51:22,349 - model.dcrnn_supervisor - INFO - Epoch [35/100] (20520) train_mae: 1.3568 val_mae: 1.6405 lr:0.000100 695.0s
2020-05-31 05:51:22,379 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6436 to 1.6405 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6405-20520
2020-05-31 06:02:57,320 - model.dcrnn_supervisor - INFO - Epoch [36/100] (21090) train_mae: 1.3814 val_mae: 1.6387 lr:0.000100 694.9s
2020-05-31 06:02:57,350 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6405 to 1.6387 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6387-21090
2020-05-31 06:14:32,394 - model.dcrnn_supervisor - INFO - Epoch [37/100] (21660) train_mae: 1.3917 val_mae: 1.6380 lr:0.000100 695.0s
2020-05-31 06:14:32,424 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6387 to 1.6380 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6380-21660
2020-05-31 06:26:07,276 - model.dcrnn_supervisor - INFO - Epoch [38/100] (22230) train_mae: 1.4027 val_mae: 1.6359 lr:0.000100 694.8s
2020-05-31 06:26:07,307 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6380 to 1.6359 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6359-22230
2020-05-31 06:37:42,308 - model.dcrnn_supervisor - INFO - Epoch [39/100] (22800) train_mae: 1.4093 val_mae: 1.6348 lr:0.000100 695.0s
2020-05-31 06:39:04,191 - model.dcrnn_supervisor - INFO - Horizon 01 0.85 0.0163 1.54
2020-05-31 06:39:04,560 - model.dcrnn_supervisor - INFO - Horizon 02 1.12 0.0226 2.22
2020-05-31 06:39:04,921 - model.dcrnn_supervisor - INFO - Horizon 03 1.31 0.0275 2.76
2020-05-31 06:39:05,287 - model.dcrnn_supervisor - INFO - Horizon 04 1.46 0.0315 3.18
2020-05-31 06:39:05,656 - model.dcrnn_supervisor - INFO - Horizon 05 1.57 0.0349 3.51
2020-05-31 06:39:06,039 - model.dcrnn_supervisor - INFO - Horizon 06 1.66 0.0377 3.78
2020-05-31 06:39:06,419 - model.dcrnn_supervisor - INFO - Horizon 07 1.73 0.0400 3.99
2020-05-31 06:39:06,791 - model.dcrnn_supervisor - INFO - Horizon 08 1.79 0.0419 4.16
2020-05-31 06:39:07,159 - model.dcrnn_supervisor - INFO - Horizon 09 1.85 0.0435 4.30
2020-05-31 06:39:07,538 - model.dcrnn_supervisor - INFO - Horizon 10 1.90 0.0449 4.42
2020-05-31 06:39:07,910 - model.dcrnn_supervisor - INFO - Horizon 11 1.94 0.0461 4.52
2020-05-31 06:39:08,276 - model.dcrnn_supervisor - INFO - Horizon 12 1.98 0.0472 4.62
2020-05-31 06:39:08,364 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6359 to 1.6348 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6348-22800
2020-05-31 06:50:43,330 - model.dcrnn_supervisor - INFO - Epoch [40/100] (23370) train_mae: 1.4186 val_mae: 1.6320 lr:0.000010 695.0s
2020-05-31 06:50:43,360 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6348 to 1.6320 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6320-23370
2020-05-31 07:02:18,236 - model.dcrnn_supervisor - INFO - Epoch [41/100] (23940) train_mae: 1.4272 val_mae: 1.6323 lr:0.000010 694.9s
2020-05-31 07:13:53,231 - model.dcrnn_supervisor - INFO - Epoch [42/100] (24510) train_mae: 1.4303 val_mae: 1.6321 lr:0.000010 695.0s
2020-05-31 07:25:28,074 - model.dcrnn_supervisor - INFO - Epoch [43/100] (25080) train_mae: 1.4343 val_mae: 1.6321 lr:0.000010 694.8s
2020-05-31 07:37:02,995 - model.dcrnn_supervisor - INFO - Epoch [44/100] (25650) train_mae: 1.4330 val_mae: 1.6320 lr:0.000010 694.9s
2020-05-31 07:48:37,951 - model.dcrnn_supervisor - INFO - Epoch [45/100] (26220) train_mae: 1.4399 val_mae: 1.6321 lr:0.000010 695.0s
2020-05-31 08:00:12,914 - model.dcrnn_supervisor - INFO - Epoch [46/100] (26790) train_mae: 1.4399 val_mae: 1.6317 lr:0.000010 695.0s
2020-05-31 08:00:12,944 - model.dcrnn_supervisor - INFO - Val loss decrease from 1.6320 to 1.6317 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0530224917/models-1.6317-26790
2020-05-31 08:11:47,845 - model.dcrnn_supervisor - INFO - Epoch [47/100] (27360) train_mae: 1.4401 val_mae: 1.6319 lr:0.000010 694.9s
2020-05-31 08:23:22,752 - model.dcrnn_supervisor - INFO - Epoch [48/100] (27930) train_mae: 1.4424 val_mae: 1.6319 lr:0.000010 694.9s
2020-05-31 08:34:57,602 - model.dcrnn_supervisor - INFO - Epoch [49/100] (28500) train_mae: 1.4429 val_mae: 1.6318 lr:0.000010 694.8s
01 0.85 0.0163 1.54
03 1.31 0.0274 2.76
06 1.66 0.0376 3.78
12 1.98 0.0474 4.62
Raw
pems-bay.log
python dcrnn_train.py --config_filename=data/model/dcrnn_bay.yaml
2020-05-31 09:24:44,701 - model.dcrnn_supervisor - INFO - Log directory: data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/
2020-05-31 09:24:44,703 - model.dcrnn_supervisor - INFO - {'base_dir': 'data/model' 'log_level': 'INFO' 'data': {'batch_size': 64 'dataset_dir': 'data/METR-LA' 'test_batch_size': 64 'val_batch_size': 64 'graph_pkl_filename': 'data/sensor_graph/adj_mx.pkl'} 'model': {'cl_decay_steps': 2000 'filter_type': 'dual_random_walk' 'horizon': 12 'input_dim': 2 'l1_decay': 0 'max_diffusion_step': 2 'num_nodes': 207 'num_rnn_layers': 2 'output_dim': 1 'rnn_units': 64 'seq_len': 12 'use_curriculum_learning': True} 'train': {'base_lr': 0.01 'dropout': 0 'epoch': 0 'epochs': 100 'epsilon': 0.001 'global_step': 0 'lr_decay_ratio': 0.1 'max_grad_norm': 5 'max_to_keep': 100 'min_learning_rate': 2e-06 'optimizer': 'adam' 'patience': 50 'steps': [20 30 40 50] 'test_every_n_epochs': 10}}
2020-05-31 09:24:54,964 - model.dcrnn_supervisor - INFO - ('x_train' (23974 12 207 2))
2020-05-31 09:24:54,966 - model.dcrnn_supervisor - INFO - ('y_train' (23974 12 207 2))
2020-05-31 09:24:54,967 - model.dcrnn_supervisor - INFO - ('x_val' (3425 12 207 2))
2020-05-31 09:24:54,967 - model.dcrnn_supervisor - INFO - ('y_val' (3425 12 207 2))
2020-05-31 09:24:54,968 - model.dcrnn_supervisor - INFO - ('x_test' (6850 12 207 2))
2020-05-31 09:24:54,969 - model.dcrnn_supervisor - INFO - ('y_test' (6850 12 207 2))
2020-05-31 09:25:09,799 - model.dcrnn_supervisor - INFO - Total number of trainable parameters: 372352
2020-05-31 09:25:10,771 - model.dcrnn_supervisor - INFO - Start training ...
2020-05-31 09:29:59,045 - model.dcrnn_supervisor - INFO - Epoch [0/100] (375) train_mae: 2.7208 val_mae: 3.6353 lr:0.010000 287.9s
2020-05-31 09:29:59,402 - model.dcrnn_supervisor - INFO - Val loss decrease from inf to 3.6353 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-3.6353-375
2020-05-31 09:34:38,635 - model.dcrnn_supervisor - INFO - Epoch [1/100] (750) train_mae: 2.3438 val_mae: 3.2871 lr:0.010000 279.2s
2020-05-31 09:34:38,663 - model.dcrnn_supervisor - INFO - Val loss decrease from 3.6353 to 3.2871 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-3.2871-750
2020-05-31 09:39:17,048 - model.dcrnn_supervisor - INFO - Epoch [2/100] (1125) train_mae: 2.2602 val_mae: 3.3842 lr:0.010000 278.4s
2020-05-31 09:43:55,232 - model.dcrnn_supervisor - INFO - Epoch [3/100] (1500) train_mae: 2.2054 val_mae: 3.2954 lr:0.010000 278.2s
2020-05-31 09:48:33,467 - model.dcrnn_supervisor - INFO - Epoch [4/100] (1875) train_mae: 2.1714 val_mae: 3.2301 lr:0.010000 278.2s
2020-05-31 09:48:33,494 - model.dcrnn_supervisor - INFO - Val loss decrease from 3.2871 to 3.2301 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-3.2301-1875
2020-05-31 09:53:11,785 - model.dcrnn_supervisor - INFO - Epoch [5/100] (2250) train_mae: 2.1429 val_mae: 3.0727 lr:0.010000 278.3s
2020-05-31 09:53:11,812 - model.dcrnn_supervisor - INFO - Val loss decrease from 3.2301 to 3.0727 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-3.0727-2250
2020-05-31 09:57:50,076 - model.dcrnn_supervisor - INFO - Epoch [6/100] (2625) train_mae: 2.1181 val_mae: 3.0901 lr:0.010000 278.3s
2020-05-31 10:02:32,703 - model.dcrnn_supervisor - INFO - Epoch [7/100] (3000) train_mae: 2.0988 val_mae: 3.1135 lr:0.010000 282.6s
2020-05-31 10:07:23,037 - model.dcrnn_supervisor - INFO - Epoch [8/100] (3375) train_mae: 2.0830 val_mae: 3.0096 lr:0.010000 290.3s
2020-05-31 10:07:23,063 - model.dcrnn_supervisor - INFO - Val loss decrease from 3.0727 to 3.0096 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-3.0096-3375
2020-05-31 10:12:09,941 - model.dcrnn_supervisor - INFO - Epoch [9/100] (3750) train_mae: 2.0683 val_mae: 2.9969 lr:0.010000 286.9s
2020-05-31 10:12:47,532 - model.dcrnn_supervisor - INFO - Horizon 01 2.24 0.0535 3.84
2020-05-31 10:12:47,703 - model.dcrnn_supervisor - INFO - Horizon 02 2.55 0.0646 4.70
2020-05-31 10:12:47,876 - model.dcrnn_supervisor - INFO - Horizon 03 2.78 0.0741 5.36
2020-05-31 10:12:48,042 - model.dcrnn_supervisor - INFO - Horizon 04 2.98 0.0832 5.93
2020-05-31 10:12:48,237 - model.dcrnn_supervisor - INFO - Horizon 05 3.17 0.0921 6.43
2020-05-31 10:12:48,440 - model.dcrnn_supervisor - INFO - Horizon 06 3.34 0.1007 6.89
2020-05-31 10:12:48,646 - model.dcrnn_supervisor - INFO - Horizon 07 3.50 0.1087 7.30
2020-05-31 10:12:48,844 - model.dcrnn_supervisor - INFO - Horizon 08 3.66 0.1165 7.68
2020-05-31 10:12:49,024 - model.dcrnn_supervisor - INFO - Horizon 09 3.80 0.1237 8.02
2020-05-31 10:12:49,200 - model.dcrnn_supervisor - INFO - Horizon 10 3.93 0.1303 8.32
2020-05-31 10:12:49,380 - model.dcrnn_supervisor - INFO - Horizon 11 4.05 0.1362 8.59
2020-05-31 10:12:49,549 - model.dcrnn_supervisor - INFO - Horizon 12 4.16 0.1415 8.82
2020-05-31 10:12:49,607 - model.dcrnn_supervisor - INFO - Val loss decrease from 3.0096 to 2.9969 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.9969-3750
2020-05-31 10:17:34,281 - model.dcrnn_supervisor - INFO - Epoch [10/100] (4125) train_mae: 2.0569 val_mae: 2.9374 lr:0.010000 284.7s
2020-05-31 10:17:34,310 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.9969 to 2.9374 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.9374-4125
2020-05-31 10:22:13,560 - model.dcrnn_supervisor - INFO - Epoch [11/100] (4500) train_mae: 2.0445 val_mae: 2.9948 lr:0.010000 279.2s
2020-05-31 10:26:53,596 - model.dcrnn_supervisor - INFO - Epoch [12/100] (4875) train_mae: 2.0350 val_mae: 2.9911 lr:0.010000 280.0s
2020-05-31 10:31:34,145 - model.dcrnn_supervisor - INFO - Epoch [13/100] (5250) train_mae: 2.0295 val_mae: 2.8719 lr:0.010000 280.5s
2020-05-31 10:31:34,175 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.9374 to 2.8719 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.8719-5250
2020-05-31 10:36:16,977 - model.dcrnn_supervisor - INFO - Epoch [14/100] (5625) train_mae: 2.0212 val_mae: 2.8876 lr:0.010000 282.8s
2020-05-31 10:40:59,011 - model.dcrnn_supervisor - INFO - Epoch [15/100] (6000) train_mae: 2.0130 val_mae: 2.8600 lr:0.010000 282.0s
2020-05-31 10:40:59,040 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.8719 to 2.8600 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.8600-6000
2020-05-31 10:45:38,494 - model.dcrnn_supervisor - INFO - Epoch [16/100] (6375) train_mae: 2.0070 val_mae: 2.8526 lr:0.010000 279.5s
2020-05-31 10:45:38,523 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.8600 to 2.8526 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.8526-6375
2020-05-31 10:50:17,670 - model.dcrnn_supervisor - INFO - Epoch [17/100] (6750) train_mae: 2.0030 val_mae: 2.8707 lr:0.010000 279.1s
2020-05-31 10:54:59,417 - model.dcrnn_supervisor - INFO - Epoch [18/100] (7125) train_mae: 1.9992 val_mae: 2.8395 lr:0.010000 281.7s
2020-05-31 10:54:59,447 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.8526 to 2.8395 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.8395-7125
2020-05-31 10:59:40,285 - model.dcrnn_supervisor - INFO - Epoch [19/100] (7500) train_mae: 1.9970 val_mae: 2.9056 lr:0.010000 280.8s
2020-05-31 11:00:15,056 - model.dcrnn_supervisor - INFO - Horizon 01 2.21 0.0529 3.84
2020-05-31 11:00:15,229 - model.dcrnn_supervisor - INFO - Horizon 02 2.52 0.0632 4.72
2020-05-31 11:00:15,401 - model.dcrnn_supervisor - INFO - Horizon 03 2.74 0.0713 5.35
2020-05-31 11:00:15,565 - model.dcrnn_supervisor - INFO - Horizon 04 2.93 0.0787 5.87
2020-05-31 11:00:15,725 - model.dcrnn_supervisor - INFO - Horizon 05 3.09 0.0855 6.32
2020-05-31 11:00:15,886 - model.dcrnn_supervisor - INFO - Horizon 06 3.23 0.0917 6.70
2020-05-31 11:00:16,058 - model.dcrnn_supervisor - INFO - Horizon 07 3.37 0.0974 7.04
2020-05-31 11:00:16,231 - model.dcrnn_supervisor - INFO - Horizon 08 3.49 0.1028 7.34
2020-05-31 11:00:16,398 - model.dcrnn_supervisor - INFO - Horizon 09 3.60 0.1077 7.61
2020-05-31 11:00:16,569 - model.dcrnn_supervisor - INFO - Horizon 10 3.70 0.1122 7.85
2020-05-31 11:00:16,732 - model.dcrnn_supervisor - INFO - Horizon 11 3.79 0.1164 8.07
2020-05-31 11:00:16,895 - model.dcrnn_supervisor - INFO - Horizon 12 3.88 0.1204 8.27
2020-05-31 11:04:57,772 - model.dcrnn_supervisor - INFO - Epoch [20/100] (7875) train_mae: 1.9683 val_mae: 2.7815 lr:0.001000 280.9s
2020-05-31 11:04:57,802 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.8395 to 2.7815 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7815-7875
2020-05-31 11:09:37,021 - model.dcrnn_supervisor - INFO - Epoch [21/100] (8250) train_mae: 1.9675 val_mae: 2.7796 lr:0.001000 279.2s
2020-05-31 11:09:37,051 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7815 to 2.7796 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7796-8250
2020-05-31 11:14:15,677 - model.dcrnn_supervisor - INFO - Epoch [22/100] (8625) train_mae: 1.9665 val_mae: 2.7778 lr:0.001000 278.6s
2020-05-31 11:14:15,706 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7796 to 2.7778 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7778-8625
2020-05-31 11:18:54,319 - model.dcrnn_supervisor - INFO - Epoch [23/100] (9000) train_mae: 1.9681 val_mae: 2.7763 lr:0.001000 278.6s
2020-05-31 11:18:54,343 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7778 to 2.7763 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7763-9000
2020-05-31 11:23:33,028 - model.dcrnn_supervisor - INFO - Epoch [24/100] (9375) train_mae: 1.9742 val_mae: 2.7748 lr:0.001000 278.7s
2020-05-31 11:23:33,057 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7763 to 2.7748 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7748-9375
2020-05-31 11:28:11,766 - model.dcrnn_supervisor - INFO - Epoch [25/100] (9750) train_mae: 1.9740 val_mae: 2.7748 lr:0.001000 278.7s
2020-05-31 11:28:11,795 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7748 to 2.7748 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7748-9750
2020-05-31 11:32:50,471 - model.dcrnn_supervisor - INFO - Epoch [26/100] (10125) train_mae: 1.9760 val_mae: 2.7770 lr:0.001000 278.7s
2020-05-31 11:37:29,180 - model.dcrnn_supervisor - INFO - Epoch [27/100] (10500) train_mae: 1.9804 val_mae: 2.7757 lr:0.001000 278.7s
2020-05-31 11:42:07,914 - model.dcrnn_supervisor - INFO - Epoch [28/100] (10875) train_mae: 1.9854 val_mae: 2.7711 lr:0.001000 278.7s
2020-05-31 11:42:07,939 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7748 to 2.7711 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7711-10875
2020-05-31 11:46:46,683 - model.dcrnn_supervisor - INFO - Epoch [29/100] (11250) train_mae: 1.9899 val_mae: 2.7718 lr:0.001000 278.7s
2020-05-31 11:47:21,172 - model.dcrnn_supervisor - INFO - Horizon 01 2.17 0.0518 3.77
2020-05-31 11:47:21,333 - model.dcrnn_supervisor - INFO - Horizon 02 2.47 0.0614 4.62
2020-05-31 11:47:21,499 - model.dcrnn_supervisor - INFO - Horizon 03 2.68 0.0685 5.21
2020-05-31 11:47:21,664 - model.dcrnn_supervisor - INFO - Horizon 04 2.84 0.0746 5.68
2020-05-31 11:47:21,833 - model.dcrnn_supervisor - INFO - Horizon 05 2.98 0.0800 6.06
2020-05-31 11:47:22,002 - model.dcrnn_supervisor - INFO - Horizon 06 3.10 0.0849 6.39
2020-05-31 11:47:22,171 - model.dcrnn_supervisor - INFO - Horizon 07 3.21 0.0893 6.68
2020-05-31 11:47:22,334 - model.dcrnn_supervisor - INFO - Horizon 08 3.31 0.0936 6.93
2020-05-31 11:47:22,504 - model.dcrnn_supervisor - INFO - Horizon 09 3.41 0.0974 7.16
2020-05-31 11:47:22,674 - model.dcrnn_supervisor - INFO - Horizon 10 3.49 0.1008 7.37
2020-05-31 11:47:22,843 - model.dcrnn_supervisor - INFO - Horizon 11 3.57 0.1041 7.56
2020-05-31 11:47:23,017 - model.dcrnn_supervisor - INFO - Horizon 12 3.66 0.1073 7.74
2020-05-31 11:52:01,843 - model.dcrnn_supervisor - INFO - Epoch [30/100] (11625) train_mae: 1.9962 val_mae: 2.7691 lr:0.000100 278.8s
2020-05-31 11:52:01,869 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7711 to 2.7691 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7691-11625
2020-05-31 11:56:40,635 - model.dcrnn_supervisor - INFO - Epoch [31/100] (12000) train_mae: 2.0037 val_mae: 2.7688 lr:0.000100 278.8s
2020-05-31 11:56:40,663 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7691 to 2.7688 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7688-12000
2020-05-31 12:01:19,464 - model.dcrnn_supervisor - INFO - Epoch [32/100] (12375) train_mae: 2.0111 val_mae: 2.7648 lr:0.000100 278.8s
2020-05-31 12:01:19,493 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7688 to 2.7648 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7648-12375
2020-05-31 12:05:58,251 - model.dcrnn_supervisor - INFO - Epoch [33/100] (12750) train_mae: 2.0276 val_mae: 2.7678 lr:0.000100 278.8s
2020-05-31 12:10:37,026 - model.dcrnn_supervisor - INFO - Epoch [34/100] (13125) train_mae: 2.0420 val_mae: 2.7656 lr:0.000100 278.8s
2020-05-31 12:15:15,821 - model.dcrnn_supervisor - INFO - Epoch [35/100] (13500) train_mae: 2.0566 val_mae: 2.7602 lr:0.000100 278.8s
2020-05-31 12:15:15,852 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7648 to 2.7602 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7602-13500
2020-05-31 12:19:54,643 - model.dcrnn_supervisor - INFO - Epoch [36/100] (13875) train_mae: 2.0686 val_mae: 2.7622 lr:0.000100 278.8s
2020-05-31 12:24:33,418 - model.dcrnn_supervisor - INFO - Epoch [37/100] (14250) train_mae: 2.0918 val_mae: 2.7555 lr:0.000100 278.8s
2020-05-31 12:24:33,447 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7602 to 2.7555 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7555-14250
2020-05-31 12:29:12,271 - model.dcrnn_supervisor - INFO - Epoch [38/100] (14625) train_mae: 2.1164 val_mae: 2.7570 lr:0.000100 278.8s
2020-05-31 12:33:51,027 - model.dcrnn_supervisor - INFO - Epoch [39/100] (15000) train_mae: 2.1396 val_mae: 2.7540 lr:0.000100 278.8s
2020-05-31 12:34:25,504 - model.dcrnn_supervisor - INFO - Horizon 01 2.17 0.0517 3.77
2020-05-31 12:34:25,669 - model.dcrnn_supervisor - INFO - Horizon 02 2.47 0.0614 4.60
2020-05-31 12:34:25,836 - model.dcrnn_supervisor - INFO - Horizon 03 2.67 0.0687 5.19
2020-05-31 12:34:26,018 - model.dcrnn_supervisor - INFO - Horizon 04 2.83 0.0748 5.65
2020-05-31 12:34:26,190 - model.dcrnn_supervisor - INFO - Horizon 05 2.96 0.0803 6.03
2020-05-31 12:34:26,361 - model.dcrnn_supervisor - INFO - Horizon 06 3.08 0.0851 6.35
2020-05-31 12:34:26,537 - model.dcrnn_supervisor - INFO - Horizon 07 3.19 0.0895 6.64
2020-05-31 12:34:26,703 - model.dcrnn_supervisor - INFO - Horizon 08 3.28 0.0936 6.88
2020-05-31 12:34:26,873 - model.dcrnn_supervisor - INFO - Horizon 09 3.37 0.0974 7.11
2020-05-31 12:34:27,049 - model.dcrnn_supervisor - INFO - Horizon 10 3.45 0.1009 7.31
2020-05-31 12:34:27,217 - model.dcrnn_supervisor - INFO - Horizon 11 3.53 0.1042 7.50
2020-05-31 12:34:27,399 - model.dcrnn_supervisor - INFO - Horizon 12 3.60 0.1074 7.68
2020-05-31 12:34:27,454 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7555 to 2.7540 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7540-15000
2020-05-31 12:39:06,291 - model.dcrnn_supervisor - INFO - Epoch [40/100] (15375) train_mae: 2.1543 val_mae: 2.7496 lr:0.000010 278.8s
2020-05-31 12:39:06,320 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7540 to 2.7496 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7496-15375
2020-05-31 12:43:45,062 - model.dcrnn_supervisor - INFO - Epoch [41/100] (15750) train_mae: 2.1979 val_mae: 2.7493 lr:0.000010 278.7s
2020-05-31 12:43:45,092 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7496 to 2.7493 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7493-15750
2020-05-31 12:48:23,863 - model.dcrnn_supervisor - INFO - Epoch [42/100] (16125) train_mae: 2.2260 val_mae: 2.7470 lr:0.000010 278.8s
2020-05-31 12:48:23,892 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7493 to 2.7470 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7470-16125
2020-05-31 12:53:02,624 - model.dcrnn_supervisor - INFO - Epoch [43/100] (16500) train_mae: 2.2536 val_mae: 2.7459 lr:0.000010 278.7s
2020-05-31 12:53:02,652 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7470 to 2.7459 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7459-16500
2020-05-31 12:57:41,363 - model.dcrnn_supervisor - INFO - Epoch [44/100] (16875) train_mae: 2.3006 val_mae: 2.7471 lr:0.000010 278.7s
2020-05-31 13:02:25,996 - model.dcrnn_supervisor - INFO - Epoch [45/100] (17250) train_mae: 2.3366 val_mae: 2.7452 lr:0.000010 284.6s
2020-05-31 13:02:26,028 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7459 to 2.7452 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7452-17250
2020-05-31 13:07:12,035 - model.dcrnn_supervisor - INFO - Epoch [46/100] (17625) train_mae: 2.3674 val_mae: 2.7400 lr:0.000010 286.0s
2020-05-31 13:07:12,066 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7452 to 2.7400 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7400-17625
2020-05-31 13:12:00,313 - model.dcrnn_supervisor - INFO - Epoch [47/100] (18000) train_mae: 2.4123 val_mae: 2.7431 lr:0.000010 288.2s
2020-05-31 13:16:47,705 - model.dcrnn_supervisor - INFO - Epoch [48/100] (18375) train_mae: 2.4311 val_mae: 2.7412 lr:0.000010 287.4s
2020-05-31 13:21:36,024 - model.dcrnn_supervisor - INFO - Epoch [49/100] (18750) train_mae: 2.4658 val_mae: 2.7392 lr:0.000010 288.3s
2020-05-31 13:22:11,076 - model.dcrnn_supervisor - INFO - Horizon 01 2.18 0.0516 3.77
2020-05-31 13:22:11,261 - model.dcrnn_supervisor - INFO - Horizon 02 2.47 0.0613 4.60
2020-05-31 13:22:11,457 - model.dcrnn_supervisor - INFO - Horizon 03 2.67 0.0684 5.17
2020-05-31 13:22:11,647 - model.dcrnn_supervisor - INFO - Horizon 04 2.83 0.0742 5.63
2020-05-31 13:22:11,838 - model.dcrnn_supervisor - INFO - Horizon 05 2.96 0.0794 6.00
2020-05-31 13:22:12,024 - model.dcrnn_supervisor - INFO - Horizon 06 3.07 0.0839 6.31
2020-05-31 13:22:12,209 - model.dcrnn_supervisor - INFO - Horizon 07 3.18 0.0878 6.57
2020-05-31 13:22:12,401 - model.dcrnn_supervisor - INFO - Horizon 08 3.27 0.0916 6.81
2020-05-31 13:22:12,587 - model.dcrnn_supervisor - INFO - Horizon 09 3.35 0.0949 7.02
2020-05-31 13:22:12,788 - model.dcrnn_supervisor - INFO - Horizon 10 3.42 0.0980 7.21
2020-05-31 13:22:12,984 - model.dcrnn_supervisor - INFO - Horizon 11 3.49 0.1009 7.38
2020-05-31 13:22:13,168 - model.dcrnn_supervisor - INFO - Horizon 12 3.56 0.1037 7.54
2020-05-31 13:22:13,224 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7400 to 2.7392 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7392-18750
2020-05-31 13:26:52,823 - model.dcrnn_supervisor - INFO - Epoch [50/100] (19125) train_mae: 2.4935 val_mae: 2.7394 lr:0.000002 279.6s
2020-05-31 13:31:32,377 - model.dcrnn_supervisor - INFO - Epoch [51/100] (19500) train_mae: 2.5238 val_mae: 2.7390 lr:0.000002 279.5s
2020-05-31 13:31:32,402 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7392 to 2.7390 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7390-19500
2020-05-31 13:36:10,641 - model.dcrnn_supervisor - INFO - Epoch [52/100] (19875) train_mae: 2.5546 val_mae: 2.7395 lr:0.000002 278.2s
2020-05-31 13:40:48,953 - model.dcrnn_supervisor - INFO - Epoch [53/100] (20250) train_mae: 2.5660 val_mae: 2.7394 lr:0.000002 278.3s
2020-05-31 13:45:27,230 - model.dcrnn_supervisor - INFO - Epoch [54/100] (20625) train_mae: 2.5925 val_mae: 2.7385 lr:0.000002 278.3s
2020-05-31 13:45:27,260 - model.dcrnn_supervisor - INFO - Val loss decrease from 2.7390 to 2.7385 saving to data/model\dcrnn_DR_2_h_12_64-64_lr_0.01_bs_64_0531092444/models-2.7385-20625
2020-05-31 13:50:05,509 - model.dcrnn_supervisor - INFO - Epoch [55/100] (21000) train_mae: 2.6133 val_mae: 2.7389 lr:0.000002 278.2s
2020-05-31 13:54:43,765 - model.dcrnn_supervisor - INFO - Epoch [56/100] (21375) train_mae: 2.6235 val_mae: 2.7388 lr:0.000002 278.2s
2020-05-31 13:59:21,988 - model.dcrnn_supervisor - INFO - Epoch [57/100] (21750) train_mae: 2.6347 val_mae: 2.7391 lr:0.000002 278.2s
2020-05-31 14:04:00,140 - model.dcrnn_supervisor - INFO - Epoch [58/100] (22125) train_mae: 2.6509 val_mae: 2.7389 lr:0.000002 278.1s
2020-05-31 14:08:38,292 - model.dcrnn_supervisor - INFO - Epoch [59/100] (22500) train_mae: 2.6649 val_mae: 2.7391 lr:0.000002 278.1s
01 2.18 0.0517 3.77
03 2.67 0.0684 5.17
06 3.08 0.0838 6.30
12 3.56 0.1030 7.52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment