| 2023-10-23 21:32:27,709 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,709 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() | |
| )" | |
| 2023-10-23 21:32:27,710 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,710 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences | |
| - NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /home/ubuntu/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator | |
| 2023-10-23 21:32:27,710 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,710 Train: 3575 sentences | |
| 2023-10-23 21:32:27,710 (train_with_dev=False, train_with_test=False) | |
| 2023-10-23 21:32:27,710 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,710 Training Params: | |
| 2023-10-23 21:32:27,710 - learning_rate: "5e-05" | |
| 2023-10-23 21:32:27,710 - mini_batch_size: "4" | |
| 2023-10-23 21:32:27,710 - max_epochs: "10" | |
| 2023-10-23 21:32:27,710 - shuffle: "True" | |
| 2023-10-23 21:32:27,710 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,710 Plugins: | |
| 2023-10-23 21:32:27,710 - TensorboardLogger | |
| 2023-10-23 21:32:27,710 - LinearScheduler | warmup_fraction: '0.1' | |
| 2023-10-23 21:32:27,710 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,710 Final evaluation on model from best epoch (best-model.pt) | |
| 2023-10-23 21:32:27,710 - metric: "('micro avg', 'f1-score')" | |
| 2023-10-23 21:32:27,710 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,710 Computation: | |
| 2023-10-23 21:32:27,710 - compute on device: cuda:0 | |
| 2023-10-23 21:32:27,710 - embedding storage: none | |
| 2023-10-23 21:32:27,710 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,710 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3" | |
| 2023-10-23 21:32:27,710 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,710 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:32:27,710 Logging anything other than scalars to TensorBoard is currently not supported. | |
| 2023-10-23 21:32:33,214 epoch 1 - iter 89/894 - loss 1.85954162 - time (sec): 5.50 - samples/sec: 1522.26 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-23 21:32:39,007 epoch 1 - iter 178/894 - loss 1.13196503 - time (sec): 11.30 - samples/sec: 1545.75 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-23 21:32:44,612 epoch 1 - iter 267/894 - loss 0.87668233 - time (sec): 16.90 - samples/sec: 1554.74 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-23 21:32:50,192 epoch 1 - iter 356/894 - loss 0.72904744 - time (sec): 22.48 - samples/sec: 1556.98 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-23 21:32:56,003 epoch 1 - iter 445/894 - loss 0.63917825 - time (sec): 28.29 - samples/sec: 1561.58 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-23 21:33:01,626 epoch 1 - iter 534/894 - loss 0.57835257 - time (sec): 33.91 - samples/sec: 1553.54 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-23 21:33:07,308 epoch 1 - iter 623/894 - loss 0.52843333 - time (sec): 39.60 - samples/sec: 1544.71 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-23 21:33:12,815 epoch 1 - iter 712/894 - loss 0.48783348 - time (sec): 45.10 - samples/sec: 1544.53 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-23 21:33:18,426 epoch 1 - iter 801/894 - loss 0.46013170 - time (sec): 50.71 - samples/sec: 1538.23 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-23 21:33:24,155 epoch 1 - iter 890/894 - loss 0.43655607 - time (sec): 56.44 - samples/sec: 1525.91 - lr: 0.000050 - momentum: 0.000000 | |
| 2023-10-23 21:33:24,403 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:33:24,404 EPOCH 1 done: loss 0.4350 - lr: 0.000050 | |
| 2023-10-23 21:33:29,212 DEV : loss 0.17315027117729187 - f1-score (micro avg) 0.6335 | |
| 2023-10-23 21:33:29,232 saving best model | |
| 2023-10-23 21:33:29,709 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:33:35,488 epoch 2 - iter 89/894 - loss 0.18260608 - time (sec): 5.78 - samples/sec: 1641.90 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-23 21:33:41,016 epoch 2 - iter 178/894 - loss 0.18828042 - time (sec): 11.31 - samples/sec: 1555.49 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-23 21:33:46,825 epoch 2 - iter 267/894 - loss 0.17496831 - time (sec): 17.12 - samples/sec: 1570.89 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-23 21:33:52,428 epoch 2 - iter 356/894 - loss 0.16664256 - time (sec): 22.72 - samples/sec: 1556.50 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-23 21:33:58,002 epoch 2 - iter 445/894 - loss 0.17404722 - time (sec): 28.29 - samples/sec: 1552.95 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-23 21:34:03,652 epoch 2 - iter 534/894 - loss 0.17916757 - time (sec): 33.94 - samples/sec: 1541.66 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-23 21:34:09,150 epoch 2 - iter 623/894 - loss 0.17434124 - time (sec): 39.44 - samples/sec: 1538.39 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-23 21:34:14,649 epoch 2 - iter 712/894 - loss 0.17000015 - time (sec): 44.94 - samples/sec: 1529.38 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-23 21:34:20,560 epoch 2 - iter 801/894 - loss 0.16586317 - time (sec): 50.85 - samples/sec: 1538.35 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-23 21:34:26,123 epoch 2 - iter 890/894 - loss 0.16468057 - time (sec): 56.41 - samples/sec: 1526.87 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-23 21:34:26,376 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:34:26,376 EPOCH 2 done: loss 0.1643 - lr: 0.000044 | |
| 2023-10-23 21:34:32,851 DEV : loss 0.23010773956775665 - f1-score (micro avg) 0.6613 | |
| 2023-10-23 21:34:32,871 saving best model | |
| 2023-10-23 21:34:33,473 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:34:39,532 epoch 3 - iter 89/894 - loss 0.07179713 - time (sec): 6.06 - samples/sec: 1737.21 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-23 21:34:45,194 epoch 3 - iter 178/894 - loss 0.08496786 - time (sec): 11.72 - samples/sec: 1633.46 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-23 21:34:50,794 epoch 3 - iter 267/894 - loss 0.09730028 - time (sec): 17.32 - samples/sec: 1589.49 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-23 21:34:56,397 epoch 3 - iter 356/894 - loss 0.09225857 - time (sec): 22.92 - samples/sec: 1557.32 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-23 21:35:02,115 epoch 3 - iter 445/894 - loss 0.09329019 - time (sec): 28.64 - samples/sec: 1544.18 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-23 21:35:07,724 epoch 3 - iter 534/894 - loss 0.09372400 - time (sec): 34.25 - samples/sec: 1544.04 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-23 21:35:13,385 epoch 3 - iter 623/894 - loss 0.09448497 - time (sec): 39.91 - samples/sec: 1554.01 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-23 21:35:18,794 epoch 3 - iter 712/894 - loss 0.09826920 - time (sec): 45.32 - samples/sec: 1528.70 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-23 21:35:24,473 epoch 3 - iter 801/894 - loss 0.10299681 - time (sec): 51.00 - samples/sec: 1523.56 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-23 21:35:30,086 epoch 3 - iter 890/894 - loss 0.10044136 - time (sec): 56.61 - samples/sec: 1524.66 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-23 21:35:30,318 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:35:30,318 EPOCH 3 done: loss 0.1002 - lr: 0.000039 | |
| 2023-10-23 21:35:36,796 DEV : loss 0.20172163844108582 - f1-score (micro avg) 0.7141 | |
| 2023-10-23 21:35:36,817 saving best model | |
| 2023-10-23 21:35:37,380 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:35:42,943 epoch 4 - iter 89/894 - loss 0.06516763 - time (sec): 5.56 - samples/sec: 1522.26 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-23 21:35:48,508 epoch 4 - iter 178/894 - loss 0.09442131 - time (sec): 11.13 - samples/sec: 1529.96 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-23 21:35:54,262 epoch 4 - iter 267/894 - loss 0.09252073 - time (sec): 16.88 - samples/sec: 1554.23 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-23 21:35:59,933 epoch 4 - iter 356/894 - loss 0.09139230 - time (sec): 22.55 - samples/sec: 1535.00 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-23 21:36:05,830 epoch 4 - iter 445/894 - loss 0.08513860 - time (sec): 28.45 - samples/sec: 1555.33 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-23 21:36:11,396 epoch 4 - iter 534/894 - loss 0.08819751 - time (sec): 34.02 - samples/sec: 1539.93 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-23 21:36:16,949 epoch 4 - iter 623/894 - loss 0.08464294 - time (sec): 39.57 - samples/sec: 1529.17 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-23 21:36:22,421 epoch 4 - iter 712/894 - loss 0.08580279 - time (sec): 45.04 - samples/sec: 1513.48 - lr: 0.000034 - momentum: 0.000000 | |
| 2023-10-23 21:36:28,192 epoch 4 - iter 801/894 - loss 0.08417899 - time (sec): 50.81 - samples/sec: 1514.32 - lr: 0.000034 - momentum: 0.000000 | |
| 2023-10-23 21:36:33,997 epoch 4 - iter 890/894 - loss 0.08189881 - time (sec): 56.62 - samples/sec: 1523.86 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-23 21:36:34,228 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:36:34,229 EPOCH 4 done: loss 0.0822 - lr: 0.000033 | |
| 2023-10-23 21:36:40,747 DEV : loss 0.21917399764060974 - f1-score (micro avg) 0.731 | |
| 2023-10-23 21:36:40,767 saving best model | |
| 2023-10-23 21:36:41,370 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:36:47,016 epoch 5 - iter 89/894 - loss 0.05884907 - time (sec): 5.65 - samples/sec: 1538.74 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-23 21:36:52,997 epoch 5 - iter 178/894 - loss 0.05271971 - time (sec): 11.63 - samples/sec: 1624.06 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-23 21:36:58,571 epoch 5 - iter 267/894 - loss 0.04804702 - time (sec): 17.20 - samples/sec: 1576.56 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-23 21:37:04,330 epoch 5 - iter 356/894 - loss 0.04906642 - time (sec): 22.96 - samples/sec: 1573.54 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-23 21:37:09,913 epoch 5 - iter 445/894 - loss 0.04994664 - time (sec): 28.54 - samples/sec: 1560.73 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-23 21:37:15,589 epoch 5 - iter 534/894 - loss 0.04889172 - time (sec): 34.22 - samples/sec: 1550.95 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-23 21:37:21,353 epoch 5 - iter 623/894 - loss 0.04744692 - time (sec): 39.98 - samples/sec: 1544.89 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-23 21:37:26,900 epoch 5 - iter 712/894 - loss 0.05021909 - time (sec): 45.53 - samples/sec: 1528.40 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-23 21:37:32,661 epoch 5 - iter 801/894 - loss 0.05064183 - time (sec): 51.29 - samples/sec: 1527.71 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-23 21:37:38,105 epoch 5 - iter 890/894 - loss 0.04987512 - time (sec): 56.73 - samples/sec: 1519.23 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-23 21:37:38,350 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:37:38,351 EPOCH 5 done: loss 0.0498 - lr: 0.000028 | |
| 2023-10-23 21:37:44,863 DEV : loss 0.20383352041244507 - f1-score (micro avg) 0.7381 | |
| 2023-10-23 21:37:44,883 saving best model | |
| 2023-10-23 21:37:45,479 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:37:51,408 epoch 6 - iter 89/894 - loss 0.01904247 - time (sec): 5.93 - samples/sec: 1589.36 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-23 21:37:56,793 epoch 6 - iter 178/894 - loss 0.02897594 - time (sec): 11.31 - samples/sec: 1492.41 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-23 21:38:02,457 epoch 6 - iter 267/894 - loss 0.03701067 - time (sec): 16.98 - samples/sec: 1514.52 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-23 21:38:08,497 epoch 6 - iter 356/894 - loss 0.03468888 - time (sec): 23.02 - samples/sec: 1526.30 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-23 21:38:14,169 epoch 6 - iter 445/894 - loss 0.03392340 - time (sec): 28.69 - samples/sec: 1518.26 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-23 21:38:19,734 epoch 6 - iter 534/894 - loss 0.03288088 - time (sec): 34.25 - samples/sec: 1514.24 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-23 21:38:25,229 epoch 6 - iter 623/894 - loss 0.03392332 - time (sec): 39.75 - samples/sec: 1504.18 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-23 21:38:30,742 epoch 6 - iter 712/894 - loss 0.03501878 - time (sec): 45.26 - samples/sec: 1509.80 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-23 21:38:36,343 epoch 6 - iter 801/894 - loss 0.03644499 - time (sec): 50.86 - samples/sec: 1520.56 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-23 21:38:42,110 epoch 6 - iter 890/894 - loss 0.03540901 - time (sec): 56.63 - samples/sec: 1521.86 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-23 21:38:42,357 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:38:42,357 EPOCH 6 done: loss 0.0355 - lr: 0.000022 | |
| 2023-10-23 21:38:48,853 DEV : loss 0.253505676984787 - f1-score (micro avg) 0.7499 | |
| 2023-10-23 21:38:48,874 saving best model | |
| 2023-10-23 21:38:49,472 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:38:55,299 epoch 7 - iter 89/894 - loss 0.01657475 - time (sec): 5.83 - samples/sec: 1585.69 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-23 21:39:01,391 epoch 7 - iter 178/894 - loss 0.02316210 - time (sec): 11.92 - samples/sec: 1592.58 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-23 21:39:06,969 epoch 7 - iter 267/894 - loss 0.02304160 - time (sec): 17.50 - samples/sec: 1576.26 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-23 21:39:12,479 epoch 7 - iter 356/894 - loss 0.02067861 - time (sec): 23.01 - samples/sec: 1539.39 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-23 21:39:18,127 epoch 7 - iter 445/894 - loss 0.02307564 - time (sec): 28.65 - samples/sec: 1519.76 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-23 21:39:23,764 epoch 7 - iter 534/894 - loss 0.02313820 - time (sec): 34.29 - samples/sec: 1520.77 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-23 21:39:29,376 epoch 7 - iter 623/894 - loss 0.02266132 - time (sec): 39.90 - samples/sec: 1527.61 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-23 21:39:35,012 epoch 7 - iter 712/894 - loss 0.02291381 - time (sec): 45.54 - samples/sec: 1521.63 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-23 21:39:40,592 epoch 7 - iter 801/894 - loss 0.02355819 - time (sec): 51.12 - samples/sec: 1520.94 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-23 21:39:46,210 epoch 7 - iter 890/894 - loss 0.02376008 - time (sec): 56.74 - samples/sec: 1518.40 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-23 21:39:46,480 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:39:46,481 EPOCH 7 done: loss 0.0242 - lr: 0.000017 | |
| 2023-10-23 21:39:52,974 DEV : loss 0.24752555787563324 - f1-score (micro avg) 0.7515 | |
| 2023-10-23 21:39:52,994 saving best model | |
| 2023-10-23 21:39:53,588 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:39:59,363 epoch 8 - iter 89/894 - loss 0.00797625 - time (sec): 5.77 - samples/sec: 1505.38 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-23 21:40:04,885 epoch 8 - iter 178/894 - loss 0.01179722 - time (sec): 11.30 - samples/sec: 1492.29 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-23 21:40:10,569 epoch 8 - iter 267/894 - loss 0.01080412 - time (sec): 16.98 - samples/sec: 1498.19 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-23 21:40:16,067 epoch 8 - iter 356/894 - loss 0.01373510 - time (sec): 22.48 - samples/sec: 1484.42 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-23 21:40:21,773 epoch 8 - iter 445/894 - loss 0.01537574 - time (sec): 28.18 - samples/sec: 1484.17 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-23 21:40:27,350 epoch 8 - iter 534/894 - loss 0.01333527 - time (sec): 33.76 - samples/sec: 1491.14 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-23 21:40:33,254 epoch 8 - iter 623/894 - loss 0.01300049 - time (sec): 39.67 - samples/sec: 1508.45 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-23 21:40:38,826 epoch 8 - iter 712/894 - loss 0.01338439 - time (sec): 45.24 - samples/sec: 1500.02 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-23 21:40:44,528 epoch 8 - iter 801/894 - loss 0.01251899 - time (sec): 50.94 - samples/sec: 1513.88 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-23 21:40:50,299 epoch 8 - iter 890/894 - loss 0.01191349 - time (sec): 56.71 - samples/sec: 1519.86 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-23 21:40:50,543 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:40:50,544 EPOCH 8 done: loss 0.0120 - lr: 0.000011 | |
| 2023-10-23 21:40:57,027 DEV : loss 0.2516254484653473 - f1-score (micro avg) 0.7672 | |
| 2023-10-23 21:40:57,048 saving best model | |
| 2023-10-23 21:40:57,641 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:41:03,127 epoch 9 - iter 89/894 - loss 0.00565499 - time (sec): 5.48 - samples/sec: 1435.26 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-23 21:41:09,113 epoch 9 - iter 178/894 - loss 0.00775794 - time (sec): 11.47 - samples/sec: 1558.87 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-23 21:41:14,835 epoch 9 - iter 267/894 - loss 0.00877010 - time (sec): 17.19 - samples/sec: 1553.70 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-23 21:41:20,509 epoch 9 - iter 356/894 - loss 0.00860547 - time (sec): 22.87 - samples/sec: 1548.72 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-23 21:41:26,256 epoch 9 - iter 445/894 - loss 0.00877409 - time (sec): 28.61 - samples/sec: 1545.27 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-23 21:41:32,176 epoch 9 - iter 534/894 - loss 0.00833655 - time (sec): 34.53 - samples/sec: 1545.37 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-23 21:41:37,747 epoch 9 - iter 623/894 - loss 0.00755503 - time (sec): 40.10 - samples/sec: 1539.71 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-23 21:41:43,212 epoch 9 - iter 712/894 - loss 0.00861964 - time (sec): 45.57 - samples/sec: 1525.39 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-23 21:41:48,681 epoch 9 - iter 801/894 - loss 0.00796957 - time (sec): 51.04 - samples/sec: 1513.43 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-23 21:41:54,306 epoch 9 - iter 890/894 - loss 0.00787224 - time (sec): 56.66 - samples/sec: 1519.07 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-23 21:41:54,556 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:41:54,556 EPOCH 9 done: loss 0.0079 - lr: 0.000006 | |
| 2023-10-23 21:42:00,770 DEV : loss 0.25944018363952637 - f1-score (micro avg) 0.7679 | |
| 2023-10-23 21:42:00,791 saving best model | |
| 2023-10-23 21:42:01,367 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:42:07,589 epoch 10 - iter 89/894 - loss 0.01315473 - time (sec): 6.22 - samples/sec: 1495.99 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-23 21:42:13,189 epoch 10 - iter 178/894 - loss 0.00864657 - time (sec): 11.82 - samples/sec: 1501.56 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-23 21:42:19,148 epoch 10 - iter 267/894 - loss 0.00690061 - time (sec): 17.78 - samples/sec: 1552.87 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-23 21:42:24,658 epoch 10 - iter 356/894 - loss 0.00654629 - time (sec): 23.29 - samples/sec: 1534.16 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-23 21:42:30,236 epoch 10 - iter 445/894 - loss 0.00595186 - time (sec): 28.87 - samples/sec: 1523.07 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-23 21:42:35,875 epoch 10 - iter 534/894 - loss 0.00673341 - time (sec): 34.51 - samples/sec: 1531.08 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-23 21:42:41,401 epoch 10 - iter 623/894 - loss 0.00642803 - time (sec): 40.03 - samples/sec: 1523.41 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-23 21:42:47,195 epoch 10 - iter 712/894 - loss 0.00581504 - time (sec): 45.83 - samples/sec: 1533.00 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-23 21:42:52,647 epoch 10 - iter 801/894 - loss 0.00544909 - time (sec): 51.28 - samples/sec: 1516.60 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-23 21:42:58,299 epoch 10 - iter 890/894 - loss 0.00516166 - time (sec): 56.93 - samples/sec: 1515.13 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-23 21:42:58,532 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:42:58,532 EPOCH 10 done: loss 0.0053 - lr: 0.000000 | |
| 2023-10-23 21:43:04,758 DEV : loss 0.2663462460041046 - f1-score (micro avg) 0.7723 | |
| 2023-10-23 21:43:04,778 saving best model | |
| 2023-10-23 21:43:05,930 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 21:43:05,931 Loading model from best epoch ... | |
| 2023-10-23 21:43:07,654 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-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time | |
| 2023-10-23 21:43:12,469 | |
| Results: | |
| - F-score (micro) 0.7542 | |
| - F-score (macro) 0.6651 | |
| - Accuracy 0.6209 | |
| By class: | |
| precision recall f1-score support | |
| loc 0.8209 0.8691 0.8443 596 | |
| pers 0.7000 0.7568 0.7273 333 | |
| org 0.5364 0.4470 0.4876 132 | |
| prod 0.6327 0.4697 0.5391 66 | |
| time 0.7200 0.7347 0.7273 49 | |
| micro avg 0.7467 0.7619 0.7542 1176 | |
| macro avg 0.6820 0.6554 0.6651 1176 | |
| weighted avg 0.7400 0.7619 0.7491 1176 | |
| 2023-10-23 21:43:12,469 ---------------------------------------------------------------------------------------------------- | |