| 2023-10-24 13:26:12,906 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,907 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 13:26:12,907 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,907 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 13:26:12,907 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,907 Train: 5901 sentences | |
| 2023-10-24 13:26:12,907 (train_with_dev=False, train_with_test=False) | |
| 2023-10-24 13:26:12,907 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,907 Training Params: | |
| 2023-10-24 13:26:12,907 - learning_rate: "5e-05" | |
| 2023-10-24 13:26:12,907 - mini_batch_size: "8" | |
| 2023-10-24 13:26:12,907 - max_epochs: "10" | |
| 2023-10-24 13:26:12,907 - shuffle: "True" | |
| 2023-10-24 13:26:12,907 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,907 Plugins: | |
| 2023-10-24 13:26:12,907 - TensorboardLogger | |
| 2023-10-24 13:26:12,907 - LinearScheduler | warmup_fraction: '0.1' | |
| 2023-10-24 13:26:12,907 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,907 Final evaluation on model from best epoch (best-model.pt) | |
| 2023-10-24 13:26:12,907 - metric: "('micro avg', 'f1-score')" | |
| 2023-10-24 13:26:12,907 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,907 Computation: | |
| 2023-10-24 13:26:12,907 - compute on device: cuda:0 | |
| 2023-10-24 13:26:12,907 - embedding storage: none | |
| 2023-10-24 13:26:12,907 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,908 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5" | |
| 2023-10-24 13:26:12,908 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,908 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:26:12,908 Logging anything other than scalars to TensorBoard is currently not supported. | |
| 2023-10-24 13:26:19,585 epoch 1 - iter 73/738 - loss 1.87843397 - time (sec): 6.68 - samples/sec: 2355.81 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-24 13:26:26,648 epoch 1 - iter 146/738 - loss 1.23669691 - time (sec): 13.74 - samples/sec: 2287.20 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-24 13:26:33,333 epoch 1 - iter 219/738 - loss 0.95046007 - time (sec): 20.42 - samples/sec: 2293.57 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 13:26:39,552 epoch 1 - iter 292/738 - loss 0.78741929 - time (sec): 26.64 - samples/sec: 2325.13 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-24 13:26:47,678 epoch 1 - iter 365/738 - loss 0.66312271 - time (sec): 34.77 - samples/sec: 2330.02 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-24 13:26:54,437 epoch 1 - iter 438/738 - loss 0.58329828 - time (sec): 41.53 - samples/sec: 2361.83 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-24 13:27:01,634 epoch 1 - iter 511/738 - loss 0.52013512 - time (sec): 48.73 - samples/sec: 2367.86 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-24 13:27:08,503 epoch 1 - iter 584/738 - loss 0.47832610 - time (sec): 55.59 - samples/sec: 2361.41 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-24 13:27:15,950 epoch 1 - iter 657/738 - loss 0.44013722 - time (sec): 63.04 - samples/sec: 2355.32 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-24 13:27:22,402 epoch 1 - iter 730/738 - loss 0.41193232 - time (sec): 69.49 - samples/sec: 2358.74 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-24 13:27:23,423 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:27:23,423 EPOCH 1 done: loss 0.4080 - lr: 0.000049 | |
| 2023-10-24 13:27:29,681 DEV : loss 0.10953915119171143 - f1-score (micro avg) 0.6996 | |
| 2023-10-24 13:27:29,702 saving best model | |
| 2023-10-24 13:27:30,252 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:27:36,790 epoch 2 - iter 73/738 - loss 0.12856043 - time (sec): 6.54 - samples/sec: 2401.39 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-24 13:27:43,668 epoch 2 - iter 146/738 - loss 0.12842399 - time (sec): 13.42 - samples/sec: 2351.92 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-24 13:27:50,487 epoch 2 - iter 219/738 - loss 0.12815504 - time (sec): 20.23 - samples/sec: 2361.80 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-24 13:27:57,248 epoch 2 - iter 292/738 - loss 0.12409229 - time (sec): 27.00 - samples/sec: 2339.92 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-24 13:28:03,997 epoch 2 - iter 365/738 - loss 0.12199118 - time (sec): 33.74 - samples/sec: 2347.12 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-24 13:28:10,826 epoch 2 - iter 438/738 - loss 0.11973631 - time (sec): 40.57 - samples/sec: 2342.24 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-24 13:28:18,157 epoch 2 - iter 511/738 - loss 0.12078151 - time (sec): 47.90 - samples/sec: 2360.10 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-24 13:28:25,872 epoch 2 - iter 584/738 - loss 0.11713377 - time (sec): 55.62 - samples/sec: 2358.47 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-24 13:28:32,609 epoch 2 - iter 657/738 - loss 0.11656313 - time (sec): 62.36 - samples/sec: 2357.16 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-24 13:28:40,269 epoch 2 - iter 730/738 - loss 0.11463726 - time (sec): 70.02 - samples/sec: 2350.59 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-24 13:28:41,017 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:28:41,017 EPOCH 2 done: loss 0.1145 - lr: 0.000045 | |
| 2023-10-24 13:28:49,488 DEV : loss 0.11087270081043243 - f1-score (micro avg) 0.7895 | |
| 2023-10-24 13:28:49,509 saving best model | |
| 2023-10-24 13:28:50,218 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:28:56,322 epoch 3 - iter 73/738 - loss 0.05848905 - time (sec): 6.10 - samples/sec: 2528.26 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-24 13:29:03,544 epoch 3 - iter 146/738 - loss 0.06303135 - time (sec): 13.33 - samples/sec: 2409.83 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-24 13:29:11,131 epoch 3 - iter 219/738 - loss 0.06634706 - time (sec): 20.91 - samples/sec: 2351.64 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-24 13:29:18,504 epoch 3 - iter 292/738 - loss 0.06169395 - time (sec): 28.28 - samples/sec: 2350.13 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-24 13:29:25,686 epoch 3 - iter 365/738 - loss 0.06197945 - time (sec): 35.47 - samples/sec: 2337.71 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-24 13:29:32,841 epoch 3 - iter 438/738 - loss 0.06398829 - time (sec): 42.62 - samples/sec: 2338.06 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-24 13:29:39,705 epoch 3 - iter 511/738 - loss 0.06397235 - time (sec): 49.49 - samples/sec: 2340.00 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-24 13:29:46,062 epoch 3 - iter 584/738 - loss 0.06498214 - time (sec): 55.84 - samples/sec: 2350.07 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-24 13:29:52,694 epoch 3 - iter 657/738 - loss 0.06519288 - time (sec): 62.48 - samples/sec: 2347.72 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-24 13:29:59,865 epoch 3 - iter 730/738 - loss 0.06817157 - time (sec): 69.65 - samples/sec: 2355.77 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-24 13:30:01,021 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:30:01,021 EPOCH 3 done: loss 0.0681 - lr: 0.000039 | |
| 2023-10-24 13:30:09,509 DEV : loss 0.11597760021686554 - f1-score (micro avg) 0.8001 | |
| 2023-10-24 13:30:09,530 saving best model | |
| 2023-10-24 13:30:10,242 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:30:16,724 epoch 4 - iter 73/738 - loss 0.04098436 - time (sec): 6.48 - samples/sec: 2324.44 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-24 13:30:23,121 epoch 4 - iter 146/738 - loss 0.04478723 - time (sec): 12.88 - samples/sec: 2354.37 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-24 13:30:29,752 epoch 4 - iter 219/738 - loss 0.04880260 - time (sec): 19.51 - samples/sec: 2348.94 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-24 13:30:36,153 epoch 4 - iter 292/738 - loss 0.04742649 - time (sec): 25.91 - samples/sec: 2352.20 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-24 13:30:43,540 epoch 4 - iter 365/738 - loss 0.05043582 - time (sec): 33.30 - samples/sec: 2364.88 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-24 13:30:51,397 epoch 4 - iter 438/738 - loss 0.05054486 - time (sec): 41.15 - samples/sec: 2350.36 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-24 13:30:59,099 epoch 4 - iter 511/738 - loss 0.04763501 - time (sec): 48.86 - samples/sec: 2349.85 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-24 13:31:06,696 epoch 4 - iter 584/738 - loss 0.04763938 - time (sec): 56.45 - samples/sec: 2357.22 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-24 13:31:13,947 epoch 4 - iter 657/738 - loss 0.04791332 - time (sec): 63.70 - samples/sec: 2351.16 - lr: 0.000034 - momentum: 0.000000 | |
| 2023-10-24 13:31:20,299 epoch 4 - iter 730/738 - loss 0.04703381 - time (sec): 70.06 - samples/sec: 2353.03 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-24 13:31:20,938 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:31:20,938 EPOCH 4 done: loss 0.0472 - lr: 0.000033 | |
| 2023-10-24 13:31:29,454 DEV : loss 0.1576639711856842 - f1-score (micro avg) 0.8054 | |
| 2023-10-24 13:31:29,475 saving best model | |
| 2023-10-24 13:31:30,138 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:31:36,865 epoch 5 - iter 73/738 - loss 0.03814569 - time (sec): 6.73 - samples/sec: 2415.22 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-24 13:31:44,135 epoch 5 - iter 146/738 - loss 0.03447645 - time (sec): 14.00 - samples/sec: 2422.92 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-24 13:31:51,116 epoch 5 - iter 219/738 - loss 0.03261789 - time (sec): 20.98 - samples/sec: 2354.95 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-24 13:31:57,985 epoch 5 - iter 292/738 - loss 0.03744440 - time (sec): 27.85 - samples/sec: 2360.46 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-24 13:32:05,585 epoch 5 - iter 365/738 - loss 0.03675531 - time (sec): 35.45 - samples/sec: 2368.90 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-24 13:32:12,307 epoch 5 - iter 438/738 - loss 0.03564464 - time (sec): 42.17 - samples/sec: 2369.51 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-24 13:32:18,805 epoch 5 - iter 511/738 - loss 0.03487785 - time (sec): 48.67 - samples/sec: 2360.38 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-24 13:32:26,703 epoch 5 - iter 584/738 - loss 0.03512836 - time (sec): 56.56 - samples/sec: 2340.57 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-24 13:32:33,278 epoch 5 - iter 657/738 - loss 0.03480734 - time (sec): 63.14 - samples/sec: 2353.68 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-24 13:32:40,558 epoch 5 - iter 730/738 - loss 0.03464501 - time (sec): 70.42 - samples/sec: 2341.86 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-24 13:32:41,297 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:32:41,298 EPOCH 5 done: loss 0.0346 - lr: 0.000028 | |
| 2023-10-24 13:32:49,820 DEV : loss 0.17847341299057007 - f1-score (micro avg) 0.8278 | |
| 2023-10-24 13:32:49,842 saving best model | |
| 2023-10-24 13:32:50,561 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:32:57,858 epoch 6 - iter 73/738 - loss 0.01898276 - time (sec): 7.30 - samples/sec: 2357.37 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 13:33:03,897 epoch 6 - iter 146/738 - loss 0.02062329 - time (sec): 13.34 - samples/sec: 2388.72 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-24 13:33:11,244 epoch 6 - iter 219/738 - loss 0.01861392 - time (sec): 20.68 - samples/sec: 2330.38 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-24 13:33:19,200 epoch 6 - iter 292/738 - loss 0.02328625 - time (sec): 28.64 - samples/sec: 2365.49 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-24 13:33:25,704 epoch 6 - iter 365/738 - loss 0.02353453 - time (sec): 35.14 - samples/sec: 2361.28 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-24 13:33:32,111 epoch 6 - iter 438/738 - loss 0.02268378 - time (sec): 41.55 - samples/sec: 2356.15 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-24 13:33:38,223 epoch 6 - iter 511/738 - loss 0.02475660 - time (sec): 47.66 - samples/sec: 2350.05 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-24 13:33:45,381 epoch 6 - iter 584/738 - loss 0.02481615 - time (sec): 54.82 - samples/sec: 2350.71 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-24 13:33:53,219 epoch 6 - iter 657/738 - loss 0.02437100 - time (sec): 62.66 - samples/sec: 2352.59 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-24 13:34:00,631 epoch 6 - iter 730/738 - loss 0.02404331 - time (sec): 70.07 - samples/sec: 2350.44 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-24 13:34:01,283 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:34:01,284 EPOCH 6 done: loss 0.0239 - lr: 0.000022 | |
| 2023-10-24 13:34:09,822 DEV : loss 0.19103363156318665 - f1-score (micro avg) 0.8177 | |
| 2023-10-24 13:34:09,844 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:34:17,437 epoch 7 - iter 73/738 - loss 0.01986646 - time (sec): 7.59 - samples/sec: 2506.14 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-24 13:34:24,923 epoch 7 - iter 146/738 - loss 0.01706995 - time (sec): 15.08 - samples/sec: 2406.26 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 13:34:31,695 epoch 7 - iter 219/738 - loss 0.01522152 - time (sec): 21.85 - samples/sec: 2366.54 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-24 13:34:38,753 epoch 7 - iter 292/738 - loss 0.01709441 - time (sec): 28.91 - samples/sec: 2354.77 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-24 13:34:45,228 epoch 7 - iter 365/738 - loss 0.01655309 - time (sec): 35.38 - samples/sec: 2363.51 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-24 13:34:51,955 epoch 7 - iter 438/738 - loss 0.01608348 - time (sec): 42.11 - samples/sec: 2356.84 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-24 13:34:58,690 epoch 7 - iter 511/738 - loss 0.01628481 - time (sec): 48.84 - samples/sec: 2347.12 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 13:35:04,990 epoch 7 - iter 584/738 - loss 0.01670341 - time (sec): 55.15 - samples/sec: 2345.70 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-24 13:35:13,111 epoch 7 - iter 657/738 - loss 0.01686839 - time (sec): 63.27 - samples/sec: 2348.50 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-24 13:35:20,233 epoch 7 - iter 730/738 - loss 0.01751030 - time (sec): 70.39 - samples/sec: 2338.05 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-24 13:35:20,903 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:35:20,903 EPOCH 7 done: loss 0.0175 - lr: 0.000017 | |
| 2023-10-24 13:35:29,453 DEV : loss 0.19701939821243286 - f1-score (micro avg) 0.8147 | |
| 2023-10-24 13:35:29,474 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:35:36,177 epoch 8 - iter 73/738 - loss 0.00499695 - time (sec): 6.70 - samples/sec: 2239.76 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-24 13:35:43,356 epoch 8 - iter 146/738 - loss 0.00716743 - time (sec): 13.88 - samples/sec: 2271.84 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-24 13:35:50,568 epoch 8 - iter 219/738 - loss 0.00825664 - time (sec): 21.09 - samples/sec: 2323.82 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 13:35:58,122 epoch 8 - iter 292/738 - loss 0.01241157 - time (sec): 28.65 - samples/sec: 2372.68 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-24 13:36:04,517 epoch 8 - iter 365/738 - loss 0.01179330 - time (sec): 35.04 - samples/sec: 2374.44 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-24 13:36:11,878 epoch 8 - iter 438/738 - loss 0.01120012 - time (sec): 42.40 - samples/sec: 2367.55 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-24 13:36:18,294 epoch 8 - iter 511/738 - loss 0.01084709 - time (sec): 48.82 - samples/sec: 2364.83 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-24 13:36:25,096 epoch 8 - iter 584/738 - loss 0.01068412 - time (sec): 55.62 - samples/sec: 2365.31 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 13:36:32,706 epoch 8 - iter 657/738 - loss 0.01037388 - time (sec): 63.23 - samples/sec: 2359.64 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-24 13:36:39,555 epoch 8 - iter 730/738 - loss 0.01018278 - time (sec): 70.08 - samples/sec: 2347.80 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-24 13:36:40,255 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:36:40,255 EPOCH 8 done: loss 0.0101 - lr: 0.000011 | |
| 2023-10-24 13:36:48,796 DEV : loss 0.2113467901945114 - f1-score (micro avg) 0.8322 | |
| 2023-10-24 13:36:48,817 saving best model | |
| 2023-10-24 13:36:49,516 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:36:56,498 epoch 9 - iter 73/738 - loss 0.00505270 - time (sec): 6.98 - samples/sec: 2316.96 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-24 13:37:04,786 epoch 9 - iter 146/738 - loss 0.00820157 - time (sec): 15.27 - samples/sec: 2400.03 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-24 13:37:11,208 epoch 9 - iter 219/738 - loss 0.00649651 - time (sec): 21.69 - samples/sec: 2406.43 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-24 13:37:17,526 epoch 9 - iter 292/738 - loss 0.00547378 - time (sec): 28.01 - samples/sec: 2418.99 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-24 13:37:24,116 epoch 9 - iter 365/738 - loss 0.00612415 - time (sec): 34.60 - samples/sec: 2391.03 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-24 13:37:31,210 epoch 9 - iter 438/738 - loss 0.00698846 - time (sec): 41.69 - samples/sec: 2378.01 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-24 13:37:37,809 epoch 9 - iter 511/738 - loss 0.00675961 - time (sec): 48.29 - samples/sec: 2378.54 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-24 13:37:44,996 epoch 9 - iter 584/738 - loss 0.00753027 - time (sec): 55.48 - samples/sec: 2370.53 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-24 13:37:52,346 epoch 9 - iter 657/738 - loss 0.00763591 - time (sec): 62.83 - samples/sec: 2367.35 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 13:37:59,594 epoch 9 - iter 730/738 - loss 0.00777319 - time (sec): 70.08 - samples/sec: 2353.84 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-24 13:38:00,322 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:38:00,323 EPOCH 9 done: loss 0.0078 - lr: 0.000006 | |
| 2023-10-24 13:38:08,878 DEV : loss 0.21783244609832764 - f1-score (micro avg) 0.8352 | |
| 2023-10-24 13:38:08,900 saving best model | |
| 2023-10-24 13:38:09,600 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:38:16,920 epoch 10 - iter 73/738 - loss 0.00253360 - time (sec): 7.32 - samples/sec: 2295.08 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-24 13:38:23,358 epoch 10 - iter 146/738 - loss 0.00249723 - time (sec): 13.76 - samples/sec: 2342.58 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-24 13:38:30,009 epoch 10 - iter 219/738 - loss 0.00177836 - time (sec): 20.41 - samples/sec: 2356.11 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-24 13:38:36,780 epoch 10 - iter 292/738 - loss 0.00257288 - time (sec): 27.18 - samples/sec: 2357.09 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 13:38:43,614 epoch 10 - iter 365/738 - loss 0.00349932 - time (sec): 34.01 - samples/sec: 2339.08 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-24 13:38:50,533 epoch 10 - iter 438/738 - loss 0.00352785 - time (sec): 40.93 - samples/sec: 2317.89 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-24 13:38:57,255 epoch 10 - iter 511/738 - loss 0.00323887 - time (sec): 47.65 - samples/sec: 2327.57 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-24 13:39:03,819 epoch 10 - iter 584/738 - loss 0.00431697 - time (sec): 54.22 - samples/sec: 2329.50 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-24 13:39:11,019 epoch 10 - iter 657/738 - loss 0.00445023 - time (sec): 61.42 - samples/sec: 2355.78 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-24 13:39:19,476 epoch 10 - iter 730/738 - loss 0.00508714 - time (sec): 69.87 - samples/sec: 2356.15 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-24 13:39:20,153 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:39:20,153 EPOCH 10 done: loss 0.0050 - lr: 0.000000 | |
| 2023-10-24 13:39:28,707 DEV : loss 0.22154250741004944 - f1-score (micro avg) 0.8321 | |
| 2023-10-24 13:39:29,293 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-24 13:39:29,294 Loading model from best epoch ... | |
| 2023-10-24 13:39:31,161 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 13:39:37,474 | |
| Results: | |
| - F-score (micro) 0.7844 | |
| - F-score (macro) 0.6892 | |
| - Accuracy 0.6719 | |
| By class: | |
| precision recall f1-score support | |
| loc 0.8373 0.8695 0.8531 858 | |
| pers 0.7276 0.7858 0.7556 537 | |
| org 0.5926 0.6061 0.5993 132 | |
| time 0.5231 0.6296 0.5714 54 | |
| prod 0.7400 0.6066 0.6667 61 | |
| micro avg 0.7664 0.8033 0.7844 1642 | |
| macro avg 0.6841 0.6995 0.6892 1642 | |
| weighted avg 0.7678 0.8033 0.7846 1642 | |
| 2023-10-24 13:39:37,474 ---------------------------------------------------------------------------------------------------- | |