| 2023-10-24 11:33:33,117 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,118 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 11:33:33,118 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,118 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 11:33:33,118 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,118 Train: 5901 sentences | |
| 2023-10-24 11:33:33,118 (train_with_dev=False, train_with_test=False) | |
| 2023-10-24 11:33:33,118 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,118 Training Params: | |
| 2023-10-24 11:33:33,118 - learning_rate: "3e-05" | |
| 2023-10-24 11:33:33,118 - mini_batch_size: "4" | |
| 2023-10-24 11:33:33,118 - max_epochs: "10" | |
| 2023-10-24 11:33:33,118 - shuffle: "True" | |
| 2023-10-24 11:33:33,118 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,118 Plugins: | |
| 2023-10-24 11:33:33,118 - TensorboardLogger | |
| 2023-10-24 11:33:33,118 - LinearScheduler | warmup_fraction: '0.1' | |
| 2023-10-24 11:33:33,118 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,118 Final evaluation on model from best epoch (best-model.pt) | |
| 2023-10-24 11:33:33,119 - metric: "('micro avg', 'f1-score')" | |
| 2023-10-24 11:33:33,119 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,119 Computation: | |
| 2023-10-24 11:33:33,119 - compute on device: cuda:0 | |
| 2023-10-24 11:33:33,119 - embedding storage: none | |
| 2023-10-24 11:33:33,119 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,119 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4" | |
| 2023-10-24 11:33:33,119 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,119 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:33:33,119 Logging anything other than scalars to TensorBoard is currently not supported. | |
| 2023-10-24 11:33:42,246 epoch 1 - iter 147/1476 - loss 2.19536875 - time (sec): 9.13 - samples/sec: 1703.07 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 11:33:52,168 epoch 1 - iter 294/1476 - loss 1.30454022 - time (sec): 19.05 - samples/sec: 1773.95 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 11:34:01,785 epoch 1 - iter 441/1476 - loss 1.00249830 - time (sec): 28.67 - samples/sec: 1775.42 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-24 11:34:11,288 epoch 1 - iter 588/1476 - loss 0.83438704 - time (sec): 38.17 - samples/sec: 1754.02 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 11:34:20,503 epoch 1 - iter 735/1476 - loss 0.72982303 - time (sec): 47.38 - samples/sec: 1741.88 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 11:34:30,629 epoch 1 - iter 882/1476 - loss 0.63143162 - time (sec): 57.51 - samples/sec: 1765.73 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 11:34:40,266 epoch 1 - iter 1029/1476 - loss 0.56910312 - time (sec): 67.15 - samples/sec: 1768.78 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 11:34:49,663 epoch 1 - iter 1176/1476 - loss 0.52180038 - time (sec): 76.54 - samples/sec: 1762.83 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-24 11:34:59,193 epoch 1 - iter 1323/1476 - loss 0.48616883 - time (sec): 86.07 - samples/sec: 1744.55 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 11:35:08,590 epoch 1 - iter 1470/1476 - loss 0.45721393 - time (sec): 95.47 - samples/sec: 1738.79 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-24 11:35:08,928 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:35:08,928 EPOCH 1 done: loss 0.4565 - lr: 0.000030 | |
| 2023-10-24 11:35:14,879 DEV : loss 0.12407723069190979 - f1-score (micro avg) 0.7495 | |
| 2023-10-24 11:35:14,900 saving best model | |
| 2023-10-24 11:35:15,459 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:35:24,516 epoch 2 - iter 147/1476 - loss 0.13332891 - time (sec): 9.06 - samples/sec: 1667.07 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-24 11:35:34,173 epoch 2 - iter 294/1476 - loss 0.14011117 - time (sec): 18.71 - samples/sec: 1675.78 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-24 11:35:43,927 epoch 2 - iter 441/1476 - loss 0.13181640 - time (sec): 28.47 - samples/sec: 1697.25 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-24 11:35:53,440 epoch 2 - iter 588/1476 - loss 0.12886116 - time (sec): 37.98 - samples/sec: 1707.08 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-24 11:36:03,482 epoch 2 - iter 735/1476 - loss 0.12767160 - time (sec): 48.02 - samples/sec: 1727.57 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-24 11:36:13,643 epoch 2 - iter 882/1476 - loss 0.12873025 - time (sec): 58.18 - samples/sec: 1750.18 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-24 11:36:23,066 epoch 2 - iter 1029/1476 - loss 0.12421138 - time (sec): 67.61 - samples/sec: 1754.15 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-24 11:36:32,370 epoch 2 - iter 1176/1476 - loss 0.12289273 - time (sec): 76.91 - samples/sec: 1735.37 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 11:36:41,846 epoch 2 - iter 1323/1476 - loss 0.12330188 - time (sec): 86.39 - samples/sec: 1733.94 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 11:36:51,235 epoch 2 - iter 1470/1476 - loss 0.12313860 - time (sec): 95.78 - samples/sec: 1731.47 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 11:36:51,587 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:36:51,587 EPOCH 2 done: loss 0.1233 - lr: 0.000027 | |
| 2023-10-24 11:37:00,093 DEV : loss 0.1295129507780075 - f1-score (micro avg) 0.7805 | |
| 2023-10-24 11:37:00,114 saving best model | |
| 2023-10-24 11:37:00,822 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:37:10,905 epoch 3 - iter 147/1476 - loss 0.07444805 - time (sec): 10.08 - samples/sec: 1832.61 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-24 11:37:20,538 epoch 3 - iter 294/1476 - loss 0.07028273 - time (sec): 19.72 - samples/sec: 1768.17 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-24 11:37:30,165 epoch 3 - iter 441/1476 - loss 0.07005359 - time (sec): 29.34 - samples/sec: 1756.79 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-24 11:37:40,127 epoch 3 - iter 588/1476 - loss 0.07284502 - time (sec): 39.30 - samples/sec: 1792.76 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-24 11:37:49,293 epoch 3 - iter 735/1476 - loss 0.07496556 - time (sec): 48.47 - samples/sec: 1772.37 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-24 11:37:58,598 epoch 3 - iter 882/1476 - loss 0.07439841 - time (sec): 57.77 - samples/sec: 1760.35 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-24 11:38:08,091 epoch 3 - iter 1029/1476 - loss 0.07434422 - time (sec): 67.27 - samples/sec: 1752.65 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-24 11:38:17,959 epoch 3 - iter 1176/1476 - loss 0.07475674 - time (sec): 77.14 - samples/sec: 1749.95 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-24 11:38:27,481 epoch 3 - iter 1323/1476 - loss 0.07544812 - time (sec): 86.66 - samples/sec: 1731.60 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-24 11:38:37,000 epoch 3 - iter 1470/1476 - loss 0.07534427 - time (sec): 96.18 - samples/sec: 1724.93 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-24 11:38:37,342 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:38:37,342 EPOCH 3 done: loss 0.0751 - lr: 0.000023 | |
| 2023-10-24 11:38:45,917 DEV : loss 0.15013425052165985 - f1-score (micro avg) 0.8109 | |
| 2023-10-24 11:38:45,938 saving best model | |
| 2023-10-24 11:38:46,610 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:38:56,162 epoch 4 - iter 147/1476 - loss 0.06185939 - time (sec): 9.55 - samples/sec: 1718.42 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-24 11:39:05,672 epoch 4 - iter 294/1476 - loss 0.05870855 - time (sec): 19.06 - samples/sec: 1747.27 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-24 11:39:15,400 epoch 4 - iter 441/1476 - loss 0.05814235 - time (sec): 28.79 - samples/sec: 1755.24 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-24 11:39:25,523 epoch 4 - iter 588/1476 - loss 0.05681694 - time (sec): 38.91 - samples/sec: 1781.77 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-24 11:39:34,946 epoch 4 - iter 735/1476 - loss 0.05738224 - time (sec): 48.34 - samples/sec: 1761.48 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-24 11:39:44,387 epoch 4 - iter 882/1476 - loss 0.05743691 - time (sec): 57.78 - samples/sec: 1752.88 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 11:39:54,099 epoch 4 - iter 1029/1476 - loss 0.05689872 - time (sec): 67.49 - samples/sec: 1750.19 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 11:40:03,519 epoch 4 - iter 1176/1476 - loss 0.05774149 - time (sec): 76.91 - samples/sec: 1742.28 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 11:40:12,635 epoch 4 - iter 1323/1476 - loss 0.05709225 - time (sec): 86.02 - samples/sec: 1736.14 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-24 11:40:22,388 epoch 4 - iter 1470/1476 - loss 0.05698675 - time (sec): 95.78 - samples/sec: 1731.88 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-24 11:40:22,740 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:40:22,740 EPOCH 4 done: loss 0.0568 - lr: 0.000020 | |
| 2023-10-24 11:40:31,295 DEV : loss 0.18162360787391663 - f1-score (micro avg) 0.8217 | |
| 2023-10-24 11:40:31,316 saving best model | |
| 2023-10-24 11:40:32,019 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:40:41,069 epoch 5 - iter 147/1476 - loss 0.02214466 - time (sec): 9.05 - samples/sec: 1600.70 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-24 11:40:50,694 epoch 5 - iter 294/1476 - loss 0.03873008 - time (sec): 18.67 - samples/sec: 1669.50 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-24 11:41:00,332 epoch 5 - iter 441/1476 - loss 0.04486905 - time (sec): 28.31 - samples/sec: 1697.77 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-24 11:41:10,485 epoch 5 - iter 588/1476 - loss 0.04328846 - time (sec): 38.47 - samples/sec: 1732.73 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-24 11:41:20,075 epoch 5 - iter 735/1476 - loss 0.03927674 - time (sec): 48.06 - samples/sec: 1729.27 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 11:41:29,733 epoch 5 - iter 882/1476 - loss 0.03772238 - time (sec): 57.71 - samples/sec: 1724.97 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 11:41:39,031 epoch 5 - iter 1029/1476 - loss 0.03912415 - time (sec): 67.01 - samples/sec: 1718.50 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 11:41:48,430 epoch 5 - iter 1176/1476 - loss 0.03900961 - time (sec): 76.41 - samples/sec: 1719.13 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-24 11:41:57,994 epoch 5 - iter 1323/1476 - loss 0.03943131 - time (sec): 85.97 - samples/sec: 1727.94 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-24 11:42:07,675 epoch 5 - iter 1470/1476 - loss 0.03973549 - time (sec): 95.66 - samples/sec: 1732.63 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-24 11:42:08,053 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:42:08,053 EPOCH 5 done: loss 0.0396 - lr: 0.000017 | |
| 2023-10-24 11:42:16,603 DEV : loss 0.18701893091201782 - f1-score (micro avg) 0.8223 | |
| 2023-10-24 11:42:16,625 saving best model | |
| 2023-10-24 11:42:17,335 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:42:26,431 epoch 6 - iter 147/1476 - loss 0.02804029 - time (sec): 9.10 - samples/sec: 1668.22 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-24 11:42:35,842 epoch 6 - iter 294/1476 - loss 0.03043188 - time (sec): 18.51 - samples/sec: 1705.54 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-24 11:42:45,688 epoch 6 - iter 441/1476 - loss 0.02872066 - time (sec): 28.35 - samples/sec: 1717.11 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-24 11:42:55,608 epoch 6 - iter 588/1476 - loss 0.02985256 - time (sec): 38.27 - samples/sec: 1749.52 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 11:43:04,920 epoch 6 - iter 735/1476 - loss 0.03195604 - time (sec): 47.58 - samples/sec: 1727.10 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 11:43:14,749 epoch 6 - iter 882/1476 - loss 0.03203786 - time (sec): 57.41 - samples/sec: 1743.61 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 11:43:23,979 epoch 6 - iter 1029/1476 - loss 0.03051764 - time (sec): 66.64 - samples/sec: 1730.71 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-24 11:43:33,811 epoch 6 - iter 1176/1476 - loss 0.03005521 - time (sec): 76.48 - samples/sec: 1738.38 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-24 11:43:43,375 epoch 6 - iter 1323/1476 - loss 0.02868545 - time (sec): 86.04 - samples/sec: 1734.68 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-24 11:43:52,946 epoch 6 - iter 1470/1476 - loss 0.02869632 - time (sec): 95.61 - samples/sec: 1736.22 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-24 11:43:53,294 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:43:53,294 EPOCH 6 done: loss 0.0286 - lr: 0.000013 | |
| 2023-10-24 11:44:01,846 DEV : loss 0.18682819604873657 - f1-score (micro avg) 0.8323 | |
| 2023-10-24 11:44:01,867 saving best model | |
| 2023-10-24 11:44:02,600 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:44:12,156 epoch 7 - iter 147/1476 - loss 0.02464445 - time (sec): 9.55 - samples/sec: 1681.06 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-24 11:44:21,557 epoch 7 - iter 294/1476 - loss 0.01641864 - time (sec): 18.96 - samples/sec: 1711.58 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-24 11:44:31,099 epoch 7 - iter 441/1476 - loss 0.01616771 - time (sec): 28.50 - samples/sec: 1705.93 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 11:44:41,530 epoch 7 - iter 588/1476 - loss 0.01938895 - time (sec): 38.93 - samples/sec: 1762.53 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 11:44:50,752 epoch 7 - iter 735/1476 - loss 0.02029554 - time (sec): 48.15 - samples/sec: 1739.57 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 11:45:00,625 epoch 7 - iter 882/1476 - loss 0.02059997 - time (sec): 58.02 - samples/sec: 1760.12 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-24 11:45:10,132 epoch 7 - iter 1029/1476 - loss 0.01976046 - time (sec): 67.53 - samples/sec: 1759.00 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-24 11:45:19,388 epoch 7 - iter 1176/1476 - loss 0.01986334 - time (sec): 76.79 - samples/sec: 1745.44 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-24 11:45:29,182 epoch 7 - iter 1323/1476 - loss 0.01944864 - time (sec): 86.58 - samples/sec: 1747.75 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-24 11:45:38,320 epoch 7 - iter 1470/1476 - loss 0.02028984 - time (sec): 95.72 - samples/sec: 1732.06 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-24 11:45:38,675 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:45:38,676 EPOCH 7 done: loss 0.0202 - lr: 0.000010 | |
| 2023-10-24 11:45:47,196 DEV : loss 0.20797935128211975 - f1-score (micro avg) 0.8111 | |
| 2023-10-24 11:45:47,218 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:45:56,470 epoch 8 - iter 147/1476 - loss 0.01332305 - time (sec): 9.25 - samples/sec: 1639.48 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-24 11:46:05,956 epoch 8 - iter 294/1476 - loss 0.01434066 - time (sec): 18.74 - samples/sec: 1702.54 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-24 11:46:15,522 epoch 8 - iter 441/1476 - loss 0.01188966 - time (sec): 28.30 - samples/sec: 1695.47 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-24 11:46:25,247 epoch 8 - iter 588/1476 - loss 0.01123738 - time (sec): 38.03 - samples/sec: 1710.90 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-24 11:46:34,377 epoch 8 - iter 735/1476 - loss 0.01212194 - time (sec): 47.16 - samples/sec: 1705.38 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-24 11:46:43,826 epoch 8 - iter 882/1476 - loss 0.01402160 - time (sec): 56.61 - samples/sec: 1707.35 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-24 11:46:53,196 epoch 8 - iter 1029/1476 - loss 0.01278898 - time (sec): 65.98 - samples/sec: 1705.09 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-24 11:47:02,661 epoch 8 - iter 1176/1476 - loss 0.01370436 - time (sec): 75.44 - samples/sec: 1702.36 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-24 11:47:13,240 epoch 8 - iter 1323/1476 - loss 0.01414650 - time (sec): 86.02 - samples/sec: 1736.26 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-24 11:47:22,917 epoch 8 - iter 1470/1476 - loss 0.01344871 - time (sec): 95.70 - samples/sec: 1733.25 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-24 11:47:23,283 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:47:23,284 EPOCH 8 done: loss 0.0134 - lr: 0.000007 | |
| 2023-10-24 11:47:31,840 DEV : loss 0.20348380506038666 - f1-score (micro avg) 0.8382 | |
| 2023-10-24 11:47:31,861 saving best model | |
| 2023-10-24 11:47:32,563 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:47:41,827 epoch 9 - iter 147/1476 - loss 0.01514449 - time (sec): 9.26 - samples/sec: 1719.99 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 11:47:51,196 epoch 9 - iter 294/1476 - loss 0.01349928 - time (sec): 18.63 - samples/sec: 1685.34 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 11:48:00,922 epoch 9 - iter 441/1476 - loss 0.01272195 - time (sec): 28.36 - samples/sec: 1719.51 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 11:48:10,722 epoch 9 - iter 588/1476 - loss 0.01109428 - time (sec): 38.16 - samples/sec: 1701.76 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-24 11:48:20,429 epoch 9 - iter 735/1476 - loss 0.01137696 - time (sec): 47.87 - samples/sec: 1718.45 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-24 11:48:29,798 epoch 9 - iter 882/1476 - loss 0.01053126 - time (sec): 57.23 - samples/sec: 1710.30 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-24 11:48:39,266 epoch 9 - iter 1029/1476 - loss 0.00958536 - time (sec): 66.70 - samples/sec: 1705.70 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-24 11:48:48,628 epoch 9 - iter 1176/1476 - loss 0.00860303 - time (sec): 76.06 - samples/sec: 1705.96 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-24 11:48:57,868 epoch 9 - iter 1323/1476 - loss 0.00911407 - time (sec): 85.30 - samples/sec: 1704.38 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-24 11:49:08,559 epoch 9 - iter 1470/1476 - loss 0.00933873 - time (sec): 96.00 - samples/sec: 1727.89 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 11:49:08,913 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:49:08,913 EPOCH 9 done: loss 0.0094 - lr: 0.000003 | |
| 2023-10-24 11:49:17,452 DEV : loss 0.22658559679985046 - f1-score (micro avg) 0.833 | |
| 2023-10-24 11:49:17,473 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:49:26,802 epoch 10 - iter 147/1476 - loss 0.00520916 - time (sec): 9.33 - samples/sec: 1666.67 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 11:49:36,575 epoch 10 - iter 294/1476 - loss 0.00900406 - time (sec): 19.10 - samples/sec: 1747.49 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 11:49:46,668 epoch 10 - iter 441/1476 - loss 0.00793928 - time (sec): 29.19 - samples/sec: 1773.84 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-24 11:49:55,920 epoch 10 - iter 588/1476 - loss 0.00756196 - time (sec): 38.45 - samples/sec: 1742.28 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-24 11:50:05,747 epoch 10 - iter 735/1476 - loss 0.00720491 - time (sec): 48.27 - samples/sec: 1746.82 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-24 11:50:15,241 epoch 10 - iter 882/1476 - loss 0.00669091 - time (sec): 57.77 - samples/sec: 1738.90 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-24 11:50:25,224 epoch 10 - iter 1029/1476 - loss 0.00658869 - time (sec): 67.75 - samples/sec: 1752.74 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-24 11:50:34,706 epoch 10 - iter 1176/1476 - loss 0.00770641 - time (sec): 77.23 - samples/sec: 1746.61 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-24 11:50:43,872 epoch 10 - iter 1323/1476 - loss 0.00707407 - time (sec): 86.40 - samples/sec: 1744.59 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-24 11:50:53,041 epoch 10 - iter 1470/1476 - loss 0.00668891 - time (sec): 95.57 - samples/sec: 1734.36 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-24 11:50:53,402 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:50:53,403 EPOCH 10 done: loss 0.0068 - lr: 0.000000 | |
| 2023-10-24 11:51:01,961 DEV : loss 0.23608621954917908 - f1-score (micro avg) 0.8291 | |
| 2023-10-24 11:51:02,537 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 11:51:02,538 Loading model from best epoch ... | |
| 2023-10-24 11:51:04,405 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 11:51:10,712 | |
| Results: | |
| - F-score (micro) 0.792 | |
| - F-score (macro) 0.7045 | |
| - Accuracy 0.6809 | |
| By class: | |
| precision recall f1-score support | |
| loc 0.8640 0.8660 0.8650 858 | |
| pers 0.7417 0.7914 0.7658 537 | |
| org 0.5248 0.5606 0.5421 132 | |
| time 0.5373 0.6667 0.5950 54 | |
| prod 0.8113 0.7049 0.7544 61 | |
| micro avg 0.7798 0.8045 0.7920 1642 | |
| macro avg 0.6958 0.7179 0.7045 1642 | |
| weighted avg 0.7840 0.8045 0.7936 1642 | |
| 2023-10-24 11:51:10,712 ---------------------------------------------------------------------------------------------------- | |