| 2023-10-24 10:30:21,441 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,442 Model: "SequenceTagger( | |
| (embeddings): TransformerWordEmbeddings( | |
| (model): BertModel( | |
| (embeddings): BertEmbeddings( | |
| (word_embeddings): Embedding(64001, 768) | |
| (position_embeddings): Embedding(512, 768) | |
| (token_type_embeddings): Embedding(2, 768) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (encoder): BertEncoder( | |
| (layer): ModuleList( | |
| (0): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (1): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (2): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (3): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (4): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (5): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (6): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (7): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (8): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (9): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (10): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (11): BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (query): Linear(in_features=768, out_features=768, bias=True) | |
| (key): Linear(in_features=768, out_features=768, bias=True) | |
| (value): Linear(in_features=768, out_features=768, bias=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| (output): BertSelfOutput( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| (intermediate): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (dense): Linear(in_features=3072, out_features=768, bias=True) | |
| (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) | |
| (dropout): Dropout(p=0.1, inplace=False) | |
| ) | |
| ) | |
| ) | |
| ) | |
| (pooler): BertPooler( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (activation): Tanh() | |
| ) | |
| ) | |
| ) | |
| (locked_dropout): LockedDropout(p=0.5) | |
| (linear): Linear(in_features=768, out_features=21, bias=True) | |
| (loss_function): CrossEntropyLoss() | |
| )" | |
| 2023-10-24 10:30:21,442 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,442 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences | |
| - NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /home/ubuntu/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator | |
| 2023-10-24 10:30:21,442 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,442 Train: 5901 sentences | |
| 2023-10-24 10:30:21,442 (train_with_dev=False, train_with_test=False) | |
| 2023-10-24 10:30:21,442 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,442 Training Params: | |
| 2023-10-24 10:30:21,442 - learning_rate: "3e-05" | |
| 2023-10-24 10:30:21,442 - mini_batch_size: "4" | |
| 2023-10-24 10:30:21,442 - max_epochs: "10" | |
| 2023-10-24 10:30:21,443 - shuffle: "True" | |
| 2023-10-24 10:30:21,443 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,443 Plugins: | |
| 2023-10-24 10:30:21,443 - TensorboardLogger | |
| 2023-10-24 10:30:21,443 - LinearScheduler | warmup_fraction: '0.1' | |
| 2023-10-24 10:30:21,443 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,443 Final evaluation on model from best epoch (best-model.pt) | |
| 2023-10-24 10:30:21,443 - metric: "('micro avg', 'f1-score')" | |
| 2023-10-24 10:30:21,443 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,443 Computation: | |
| 2023-10-24 10:30:21,443 - compute on device: cuda:0 | |
| 2023-10-24 10:30:21,443 - embedding storage: none | |
| 2023-10-24 10:30:21,443 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,443 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3" | |
| 2023-10-24 10:30:21,443 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,443 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:30:21,443 Logging anything other than scalars to TensorBoard is currently not supported. | |
| 2023-10-24 10:30:30,685 epoch 1 - iter 147/1476 - loss 1.95486685 - time (sec): 9.24 - samples/sec: 1731.50 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 10:30:39,970 epoch 1 - iter 294/1476 - loss 1.26204596 - time (sec): 18.53 - samples/sec: 1712.28 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 10:30:49,068 epoch 1 - iter 441/1476 - loss 1.01783980 - time (sec): 27.62 - samples/sec: 1664.50 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-24 10:30:58,969 epoch 1 - iter 588/1476 - loss 0.82960997 - time (sec): 37.53 - samples/sec: 1720.25 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 10:31:09,389 epoch 1 - iter 735/1476 - loss 0.69238553 - time (sec): 47.95 - samples/sec: 1760.88 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 10:31:18,839 epoch 1 - iter 882/1476 - loss 0.61723371 - time (sec): 57.40 - samples/sec: 1758.66 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 10:31:28,128 epoch 1 - iter 1029/1476 - loss 0.55914623 - time (sec): 66.68 - samples/sec: 1750.72 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 10:31:37,986 epoch 1 - iter 1176/1476 - loss 0.51214581 - time (sec): 76.54 - samples/sec: 1748.28 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-24 10:31:47,268 epoch 1 - iter 1323/1476 - loss 0.47688361 - time (sec): 85.82 - samples/sec: 1743.62 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 10:31:56,801 epoch 1 - iter 1470/1476 - loss 0.44429417 - time (sec): 95.36 - samples/sec: 1740.14 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-24 10:31:57,149 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:31:57,150 EPOCH 1 done: loss 0.4435 - lr: 0.000030 | |
| 2023-10-24 10:32:03,446 DEV : loss 0.1300630122423172 - f1-score (micro avg) 0.7315 | |
| 2023-10-24 10:32:03,467 saving best model | |
| 2023-10-24 10:32:04,027 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:32:13,594 epoch 2 - iter 147/1476 - loss 0.09774384 - time (sec): 9.57 - samples/sec: 1764.56 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-24 10:32:22,802 epoch 2 - iter 294/1476 - loss 0.12373892 - time (sec): 18.77 - samples/sec: 1717.04 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-24 10:32:31,998 epoch 2 - iter 441/1476 - loss 0.13775868 - time (sec): 27.97 - samples/sec: 1678.40 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-24 10:32:41,763 epoch 2 - iter 588/1476 - loss 0.12915200 - time (sec): 37.74 - samples/sec: 1702.64 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-24 10:32:51,061 epoch 2 - iter 735/1476 - loss 0.12918783 - time (sec): 47.03 - samples/sec: 1692.83 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-24 10:33:00,686 epoch 2 - iter 882/1476 - loss 0.12958345 - time (sec): 56.66 - samples/sec: 1704.04 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-24 10:33:09,743 epoch 2 - iter 1029/1476 - loss 0.12887412 - time (sec): 65.71 - samples/sec: 1691.48 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-24 10:33:19,759 epoch 2 - iter 1176/1476 - loss 0.12754934 - time (sec): 75.73 - samples/sec: 1725.62 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 10:33:29,648 epoch 2 - iter 1323/1476 - loss 0.12821778 - time (sec): 85.62 - samples/sec: 1726.78 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 10:33:39,650 epoch 2 - iter 1470/1476 - loss 0.12716691 - time (sec): 95.62 - samples/sec: 1735.61 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 10:33:39,996 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:33:39,997 EPOCH 2 done: loss 0.1270 - lr: 0.000027 | |
| 2023-10-24 10:33:48,525 DEV : loss 0.13529355823993683 - f1-score (micro avg) 0.7895 | |
| 2023-10-24 10:33:48,546 saving best model | |
| 2023-10-24 10:33:49,250 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:33:58,610 epoch 3 - iter 147/1476 - loss 0.06413979 - time (sec): 9.36 - samples/sec: 1629.69 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-24 10:34:08,613 epoch 3 - iter 294/1476 - loss 0.08008883 - time (sec): 19.36 - samples/sec: 1719.49 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-24 10:34:18,091 epoch 3 - iter 441/1476 - loss 0.07801843 - time (sec): 28.84 - samples/sec: 1704.77 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-24 10:34:27,913 epoch 3 - iter 588/1476 - loss 0.07393475 - time (sec): 38.66 - samples/sec: 1742.86 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-24 10:34:37,183 epoch 3 - iter 735/1476 - loss 0.07460277 - time (sec): 47.93 - samples/sec: 1724.93 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-24 10:34:46,875 epoch 3 - iter 882/1476 - loss 0.07823560 - time (sec): 57.62 - samples/sec: 1735.79 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-24 10:34:56,414 epoch 3 - iter 1029/1476 - loss 0.07661923 - time (sec): 67.16 - samples/sec: 1731.96 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-24 10:35:05,800 epoch 3 - iter 1176/1476 - loss 0.07755315 - time (sec): 76.55 - samples/sec: 1727.99 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-24 10:35:15,765 epoch 3 - iter 1323/1476 - loss 0.07765157 - time (sec): 86.51 - samples/sec: 1744.74 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-24 10:35:24,956 epoch 3 - iter 1470/1476 - loss 0.07767577 - time (sec): 95.70 - samples/sec: 1735.25 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-24 10:35:25,291 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:35:25,291 EPOCH 3 done: loss 0.0777 - lr: 0.000023 | |
| 2023-10-24 10:35:33,788 DEV : loss 0.1414514034986496 - f1-score (micro avg) 0.8064 | |
| 2023-10-24 10:35:33,809 saving best model | |
| 2023-10-24 10:35:34,565 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:35:44,207 epoch 4 - iter 147/1476 - loss 0.05130159 - time (sec): 9.64 - samples/sec: 1746.38 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-24 10:35:53,931 epoch 4 - iter 294/1476 - loss 0.04892145 - time (sec): 19.36 - samples/sec: 1811.41 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-24 10:36:03,580 epoch 4 - iter 441/1476 - loss 0.04776918 - time (sec): 29.01 - samples/sec: 1782.70 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-24 10:36:13,245 epoch 4 - iter 588/1476 - loss 0.04668210 - time (sec): 38.68 - samples/sec: 1746.47 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-24 10:36:22,993 epoch 4 - iter 735/1476 - loss 0.04560696 - time (sec): 48.43 - samples/sec: 1754.83 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-24 10:36:32,409 epoch 4 - iter 882/1476 - loss 0.04633596 - time (sec): 57.84 - samples/sec: 1747.73 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 10:36:42,394 epoch 4 - iter 1029/1476 - loss 0.04952081 - time (sec): 67.83 - samples/sec: 1755.06 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 10:36:51,859 epoch 4 - iter 1176/1476 - loss 0.05327110 - time (sec): 77.29 - samples/sec: 1744.99 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 10:37:01,316 epoch 4 - iter 1323/1476 - loss 0.05312525 - time (sec): 86.75 - samples/sec: 1738.37 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-24 10:37:10,558 epoch 4 - iter 1470/1476 - loss 0.05433269 - time (sec): 95.99 - samples/sec: 1726.40 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-24 10:37:10,926 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:37:10,926 EPOCH 4 done: loss 0.0545 - lr: 0.000020 | |
| 2023-10-24 10:37:19,428 DEV : loss 0.17535756528377533 - f1-score (micro avg) 0.8202 | |
| 2023-10-24 10:37:19,449 saving best model | |
| 2023-10-24 10:37:20,150 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:37:29,883 epoch 5 - iter 147/1476 - loss 0.03243990 - time (sec): 9.73 - samples/sec: 1742.69 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-24 10:37:39,519 epoch 5 - iter 294/1476 - loss 0.04635094 - time (sec): 19.37 - samples/sec: 1773.97 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-24 10:37:49,362 epoch 5 - iter 441/1476 - loss 0.04025634 - time (sec): 29.21 - samples/sec: 1779.48 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-24 10:37:58,525 epoch 5 - iter 588/1476 - loss 0.03809557 - time (sec): 38.37 - samples/sec: 1749.42 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-24 10:38:08,520 epoch 5 - iter 735/1476 - loss 0.03722634 - time (sec): 48.37 - samples/sec: 1747.98 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 10:38:17,615 epoch 5 - iter 882/1476 - loss 0.03652541 - time (sec): 57.46 - samples/sec: 1725.74 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 10:38:26,667 epoch 5 - iter 1029/1476 - loss 0.03632487 - time (sec): 66.52 - samples/sec: 1721.17 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 10:38:35,997 epoch 5 - iter 1176/1476 - loss 0.03493445 - time (sec): 75.85 - samples/sec: 1706.50 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-24 10:38:45,476 epoch 5 - iter 1323/1476 - loss 0.03543690 - time (sec): 85.33 - samples/sec: 1712.30 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-24 10:38:55,831 epoch 5 - iter 1470/1476 - loss 0.03603493 - time (sec): 95.68 - samples/sec: 1734.96 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-24 10:38:56,171 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:38:56,172 EPOCH 5 done: loss 0.0362 - lr: 0.000017 | |
| 2023-10-24 10:39:04,666 DEV : loss 0.18856941163539886 - f1-score (micro avg) 0.8052 | |
| 2023-10-24 10:39:04,687 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:39:14,450 epoch 6 - iter 147/1476 - loss 0.03138711 - time (sec): 9.76 - samples/sec: 1826.14 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-24 10:39:24,003 epoch 6 - iter 294/1476 - loss 0.02845150 - time (sec): 19.31 - samples/sec: 1750.17 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-24 10:39:33,538 epoch 6 - iter 441/1476 - loss 0.02731184 - time (sec): 28.85 - samples/sec: 1735.88 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-24 10:39:43,120 epoch 6 - iter 588/1476 - loss 0.02757989 - time (sec): 38.43 - samples/sec: 1737.16 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 10:39:52,417 epoch 6 - iter 735/1476 - loss 0.02426912 - time (sec): 47.73 - samples/sec: 1730.41 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 10:40:02,130 epoch 6 - iter 882/1476 - loss 0.02398610 - time (sec): 57.44 - samples/sec: 1740.41 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 10:40:11,398 epoch 6 - iter 1029/1476 - loss 0.02369056 - time (sec): 66.71 - samples/sec: 1723.81 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-24 10:40:20,772 epoch 6 - iter 1176/1476 - loss 0.02397182 - time (sec): 76.08 - samples/sec: 1726.11 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-24 10:40:30,870 epoch 6 - iter 1323/1476 - loss 0.02409287 - time (sec): 86.18 - samples/sec: 1738.85 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-24 10:40:40,391 epoch 6 - iter 1470/1476 - loss 0.02429223 - time (sec): 95.70 - samples/sec: 1733.92 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-24 10:40:40,734 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:40:40,735 EPOCH 6 done: loss 0.0242 - lr: 0.000013 | |
| 2023-10-24 10:40:49,251 DEV : loss 0.1976252794265747 - f1-score (micro avg) 0.8181 | |
| 2023-10-24 10:40:49,272 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:40:58,821 epoch 7 - iter 147/1476 - loss 0.02416041 - time (sec): 9.55 - samples/sec: 1720.61 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-24 10:41:08,314 epoch 7 - iter 294/1476 - loss 0.02557096 - time (sec): 19.04 - samples/sec: 1705.97 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-24 10:41:18,019 epoch 7 - iter 441/1476 - loss 0.02401158 - time (sec): 28.75 - samples/sec: 1731.71 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 10:41:27,278 epoch 7 - iter 588/1476 - loss 0.02052173 - time (sec): 38.01 - samples/sec: 1714.38 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 10:41:36,455 epoch 7 - iter 735/1476 - loss 0.01942122 - time (sec): 47.18 - samples/sec: 1703.48 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 10:41:46,543 epoch 7 - iter 882/1476 - loss 0.01902328 - time (sec): 57.27 - samples/sec: 1730.36 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-24 10:41:56,034 epoch 7 - iter 1029/1476 - loss 0.01808307 - time (sec): 66.76 - samples/sec: 1730.74 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-24 10:42:05,699 epoch 7 - iter 1176/1476 - loss 0.01833467 - time (sec): 76.43 - samples/sec: 1730.68 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-24 10:42:15,348 epoch 7 - iter 1323/1476 - loss 0.01730497 - time (sec): 86.08 - samples/sec: 1734.66 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-24 10:42:24,846 epoch 7 - iter 1470/1476 - loss 0.01857979 - time (sec): 95.57 - samples/sec: 1734.74 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-24 10:42:25,221 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:42:25,221 EPOCH 7 done: loss 0.0186 - lr: 0.000010 | |
| 2023-10-24 10:42:33,760 DEV : loss 0.2103830873966217 - f1-score (micro avg) 0.8275 | |
| 2023-10-24 10:42:33,781 saving best model | |
| 2023-10-24 10:42:34,480 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:42:43,884 epoch 8 - iter 147/1476 - loss 0.01365644 - time (sec): 9.40 - samples/sec: 1696.78 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-24 10:42:53,017 epoch 8 - iter 294/1476 - loss 0.01467945 - time (sec): 18.54 - samples/sec: 1658.43 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-24 10:43:03,240 epoch 8 - iter 441/1476 - loss 0.01419339 - time (sec): 28.76 - samples/sec: 1760.03 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-24 10:43:12,794 epoch 8 - iter 588/1476 - loss 0.01139119 - time (sec): 38.31 - samples/sec: 1760.19 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-24 10:43:22,562 epoch 8 - iter 735/1476 - loss 0.01070596 - time (sec): 48.08 - samples/sec: 1754.76 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-24 10:43:32,626 epoch 8 - iter 882/1476 - loss 0.01177176 - time (sec): 58.15 - samples/sec: 1762.34 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-24 10:43:41,858 epoch 8 - iter 1029/1476 - loss 0.01286942 - time (sec): 67.38 - samples/sec: 1742.60 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-24 10:43:51,105 epoch 8 - iter 1176/1476 - loss 0.01221774 - time (sec): 76.62 - samples/sec: 1735.10 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-24 10:44:00,445 epoch 8 - iter 1323/1476 - loss 0.01176144 - time (sec): 85.96 - samples/sec: 1731.26 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-24 10:44:10,081 epoch 8 - iter 1470/1476 - loss 0.01227108 - time (sec): 95.60 - samples/sec: 1733.60 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-24 10:44:10,447 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:44:10,447 EPOCH 8 done: loss 0.0123 - lr: 0.000007 | |
| 2023-10-24 10:44:19,008 DEV : loss 0.2190389633178711 - f1-score (micro avg) 0.827 | |
| 2023-10-24 10:44:19,029 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:44:28,424 epoch 9 - iter 147/1476 - loss 0.00656723 - time (sec): 9.39 - samples/sec: 1692.09 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 10:44:38,241 epoch 9 - iter 294/1476 - loss 0.00522978 - time (sec): 19.21 - samples/sec: 1767.71 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 10:44:47,503 epoch 9 - iter 441/1476 - loss 0.00478950 - time (sec): 28.47 - samples/sec: 1720.95 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 10:44:56,695 epoch 9 - iter 588/1476 - loss 0.00468800 - time (sec): 37.66 - samples/sec: 1698.76 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-24 10:45:05,920 epoch 9 - iter 735/1476 - loss 0.00606865 - time (sec): 46.89 - samples/sec: 1700.29 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-24 10:45:15,342 epoch 9 - iter 882/1476 - loss 0.00622321 - time (sec): 56.31 - samples/sec: 1699.13 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-24 10:45:24,857 epoch 9 - iter 1029/1476 - loss 0.00584308 - time (sec): 65.83 - samples/sec: 1709.37 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-24 10:45:34,881 epoch 9 - iter 1176/1476 - loss 0.00648079 - time (sec): 75.85 - samples/sec: 1730.03 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-24 10:45:45,039 epoch 9 - iter 1323/1476 - loss 0.00662559 - time (sec): 86.01 - samples/sec: 1738.06 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-24 10:45:54,542 epoch 9 - iter 1470/1476 - loss 0.00685200 - time (sec): 95.51 - samples/sec: 1737.40 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 10:45:54,884 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:45:54,885 EPOCH 9 done: loss 0.0068 - lr: 0.000003 | |
| 2023-10-24 10:46:03,440 DEV : loss 0.22385086119174957 - f1-score (micro avg) 0.8342 | |
| 2023-10-24 10:46:03,462 saving best model | |
| 2023-10-24 10:46:04,162 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:46:13,588 epoch 10 - iter 147/1476 - loss 0.00352151 - time (sec): 9.42 - samples/sec: 1718.97 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 10:46:22,961 epoch 10 - iter 294/1476 - loss 0.00446608 - time (sec): 18.80 - samples/sec: 1702.67 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 10:46:32,841 epoch 10 - iter 441/1476 - loss 0.00470680 - time (sec): 28.68 - samples/sec: 1742.15 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-24 10:46:42,528 epoch 10 - iter 588/1476 - loss 0.00484924 - time (sec): 38.36 - samples/sec: 1761.90 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-24 10:46:52,776 epoch 10 - iter 735/1476 - loss 0.00598633 - time (sec): 48.61 - samples/sec: 1776.90 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-24 10:47:02,222 epoch 10 - iter 882/1476 - loss 0.00629805 - time (sec): 58.06 - samples/sec: 1762.31 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-24 10:47:12,020 epoch 10 - iter 1029/1476 - loss 0.00622798 - time (sec): 67.86 - samples/sec: 1757.82 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-24 10:47:21,170 epoch 10 - iter 1176/1476 - loss 0.00617265 - time (sec): 77.01 - samples/sec: 1743.69 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-24 10:47:30,350 epoch 10 - iter 1323/1476 - loss 0.00592236 - time (sec): 86.19 - samples/sec: 1734.52 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-24 10:47:39,701 epoch 10 - iter 1470/1476 - loss 0.00547473 - time (sec): 95.54 - samples/sec: 1736.26 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-24 10:47:40,046 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:47:40,046 EPOCH 10 done: loss 0.0055 - lr: 0.000000 | |
| 2023-10-24 10:47:48,614 DEV : loss 0.22415557503700256 - f1-score (micro avg) 0.8404 | |
| 2023-10-24 10:47:48,636 saving best model | |
| 2023-10-24 10:47:49,932 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 10:47:49,932 Loading model from best epoch ... | |
| 2023-10-24 10:47:51,801 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod | |
| 2023-10-24 10:47:58,484 | |
| Results: | |
| - F-score (micro) 0.7974 | |
| - F-score (macro) 0.7083 | |
| - Accuracy 0.6897 | |
| By class: | |
| precision recall f1-score support | |
| loc 0.8366 0.8893 0.8621 858 | |
| pers 0.7656 0.7784 0.7719 537 | |
| org 0.5882 0.6061 0.5970 132 | |
| prod 0.7018 0.6557 0.6780 61 | |
| time 0.5873 0.6852 0.6325 54 | |
| micro avg 0.7806 0.8149 0.7974 1642 | |
| macro avg 0.6959 0.7229 0.7083 1642 | |
| weighted avg 0.7802 0.8149 0.7969 1642 | |
| 2023-10-24 10:47:58,485 ---------------------------------------------------------------------------------------------------- | |