| 2023-10-23 22:11:00,819 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,820 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 22:11:00,820 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,820 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 22:11:00,820 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,820 Train: 3575 sentences | |
| 2023-10-23 22:11:00,820 (train_with_dev=False, train_with_test=False) | |
| 2023-10-23 22:11:00,820 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,820 Training Params: | |
| 2023-10-23 22:11:00,820 - learning_rate: "5e-05" | |
| 2023-10-23 22:11:00,820 - mini_batch_size: "4" | |
| 2023-10-23 22:11:00,820 - max_epochs: "10" | |
| 2023-10-23 22:11:00,820 - shuffle: "True" | |
| 2023-10-23 22:11:00,820 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,820 Plugins: | |
| 2023-10-23 22:11:00,820 - TensorboardLogger | |
| 2023-10-23 22:11:00,820 - LinearScheduler | warmup_fraction: '0.1' | |
| 2023-10-23 22:11:00,821 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,821 Final evaluation on model from best epoch (best-model.pt) | |
| 2023-10-23 22:11:00,821 - metric: "('micro avg', 'f1-score')" | |
| 2023-10-23 22:11:00,821 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,821 Computation: | |
| 2023-10-23 22:11:00,821 - compute on device: cuda:0 | |
| 2023-10-23 22:11:00,821 - embedding storage: none | |
| 2023-10-23 22:11:00,821 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,821 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4" | |
| 2023-10-23 22:11:00,821 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,821 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:00,821 Logging anything other than scalars to TensorBoard is currently not supported. | |
| 2023-10-23 22:11:06,471 epoch 1 - iter 89/894 - loss 2.24767612 - time (sec): 5.65 - samples/sec: 1465.04 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-23 22:11:12,248 epoch 1 - iter 178/894 - loss 1.32989836 - time (sec): 11.43 - samples/sec: 1509.50 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-23 22:11:17,833 epoch 1 - iter 267/894 - loss 1.01594059 - time (sec): 17.01 - samples/sec: 1500.72 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-23 22:11:23,383 epoch 1 - iter 356/894 - loss 0.85782838 - time (sec): 22.56 - samples/sec: 1499.78 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-23 22:11:28,969 epoch 1 - iter 445/894 - loss 0.73177723 - time (sec): 28.15 - samples/sec: 1508.85 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-23 22:11:34,459 epoch 1 - iter 534/894 - loss 0.65249591 - time (sec): 33.64 - samples/sec: 1506.93 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-23 22:11:40,179 epoch 1 - iter 623/894 - loss 0.58897814 - time (sec): 39.36 - samples/sec: 1514.74 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-23 22:11:45,756 epoch 1 - iter 712/894 - loss 0.54026067 - time (sec): 44.93 - samples/sec: 1517.85 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-23 22:11:51,314 epoch 1 - iter 801/894 - loss 0.50352442 - time (sec): 50.49 - samples/sec: 1515.38 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-23 22:11:57,287 epoch 1 - iter 890/894 - loss 0.47177208 - time (sec): 56.47 - samples/sec: 1522.81 - lr: 0.000050 - momentum: 0.000000 | |
| 2023-10-23 22:11:57,598 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:11:57,598 EPOCH 1 done: loss 0.4717 - lr: 0.000050 | |
| 2023-10-23 22:12:02,458 DEV : loss 0.18034544587135315 - f1-score (micro avg) 0.5217 | |
| 2023-10-23 22:12:02,478 saving best model | |
| 2023-10-23 22:12:02,946 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:12:08,443 epoch 2 - iter 89/894 - loss 0.16827676 - time (sec): 5.50 - samples/sec: 1470.62 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-23 22:12:14,038 epoch 2 - iter 178/894 - loss 0.15766325 - time (sec): 11.09 - samples/sec: 1535.26 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-23 22:12:19,915 epoch 2 - iter 267/894 - loss 0.15542630 - time (sec): 16.97 - samples/sec: 1549.29 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-23 22:12:25,515 epoch 2 - iter 356/894 - loss 0.15957913 - time (sec): 22.57 - samples/sec: 1535.54 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-23 22:12:31,204 epoch 2 - iter 445/894 - loss 0.16341156 - time (sec): 28.26 - samples/sec: 1532.91 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-23 22:12:36,902 epoch 2 - iter 534/894 - loss 0.15864282 - time (sec): 33.96 - samples/sec: 1519.51 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-23 22:12:42,475 epoch 2 - iter 623/894 - loss 0.16171748 - time (sec): 39.53 - samples/sec: 1514.59 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-23 22:12:47,959 epoch 2 - iter 712/894 - loss 0.15483930 - time (sec): 45.01 - samples/sec: 1501.41 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-23 22:12:53,845 epoch 2 - iter 801/894 - loss 0.15163502 - time (sec): 50.90 - samples/sec: 1514.04 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-23 22:12:59,561 epoch 2 - iter 890/894 - loss 0.14974204 - time (sec): 56.61 - samples/sec: 1521.89 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-23 22:12:59,808 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:12:59,809 EPOCH 2 done: loss 0.1492 - lr: 0.000044 | |
| 2023-10-23 22:13:06,298 DEV : loss 0.1743343621492386 - f1-score (micro avg) 0.6821 | |
| 2023-10-23 22:13:06,318 saving best model | |
| 2023-10-23 22:13:06,906 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:13:12,429 epoch 3 - iter 89/894 - loss 0.09273514 - time (sec): 5.52 - samples/sec: 1420.41 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-23 22:13:18,207 epoch 3 - iter 178/894 - loss 0.09432610 - time (sec): 11.30 - samples/sec: 1444.51 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-23 22:13:23,762 epoch 3 - iter 267/894 - loss 0.09201036 - time (sec): 16.86 - samples/sec: 1470.53 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-23 22:13:29,556 epoch 3 - iter 356/894 - loss 0.09877382 - time (sec): 22.65 - samples/sec: 1500.80 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-23 22:13:35,072 epoch 3 - iter 445/894 - loss 0.09586860 - time (sec): 28.17 - samples/sec: 1476.33 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-23 22:13:40,966 epoch 3 - iter 534/894 - loss 0.10865567 - time (sec): 34.06 - samples/sec: 1492.01 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-23 22:13:46,850 epoch 3 - iter 623/894 - loss 0.10472267 - time (sec): 39.94 - samples/sec: 1505.52 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-23 22:13:52,432 epoch 3 - iter 712/894 - loss 0.10079658 - time (sec): 45.53 - samples/sec: 1518.39 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-23 22:13:57,983 epoch 3 - iter 801/894 - loss 0.10204516 - time (sec): 51.08 - samples/sec: 1515.78 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-23 22:14:03,641 epoch 3 - iter 890/894 - loss 0.10044707 - time (sec): 56.73 - samples/sec: 1519.00 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-23 22:14:03,886 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:14:03,886 EPOCH 3 done: loss 0.1009 - lr: 0.000039 | |
| 2023-10-23 22:14:10,389 DEV : loss 0.17964236438274384 - f1-score (micro avg) 0.7016 | |
| 2023-10-23 22:14:10,409 saving best model | |
| 2023-10-23 22:14:11,002 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:14:16,611 epoch 4 - iter 89/894 - loss 0.06929115 - time (sec): 5.61 - samples/sec: 1511.93 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-23 22:14:22,139 epoch 4 - iter 178/894 - loss 0.07558077 - time (sec): 11.14 - samples/sec: 1487.05 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-23 22:14:27,727 epoch 4 - iter 267/894 - loss 0.06465839 - time (sec): 16.72 - samples/sec: 1507.05 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-23 22:14:33,623 epoch 4 - iter 356/894 - loss 0.06402311 - time (sec): 22.62 - samples/sec: 1527.66 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-23 22:14:39,396 epoch 4 - iter 445/894 - loss 0.06469627 - time (sec): 28.39 - samples/sec: 1527.99 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-23 22:14:45,018 epoch 4 - iter 534/894 - loss 0.06445300 - time (sec): 34.02 - samples/sec: 1522.05 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-23 22:14:50,505 epoch 4 - iter 623/894 - loss 0.06592729 - time (sec): 39.50 - samples/sec: 1523.69 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-23 22:14:56,137 epoch 4 - iter 712/894 - loss 0.06504216 - time (sec): 45.13 - samples/sec: 1523.15 - lr: 0.000034 - momentum: 0.000000 | |
| 2023-10-23 22:15:01,901 epoch 4 - iter 801/894 - loss 0.06508479 - time (sec): 50.90 - samples/sec: 1521.92 - lr: 0.000034 - momentum: 0.000000 | |
| 2023-10-23 22:15:07,561 epoch 4 - iter 890/894 - loss 0.06338887 - time (sec): 56.56 - samples/sec: 1524.30 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-23 22:15:07,809 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:15:07,809 EPOCH 4 done: loss 0.0636 - lr: 0.000033 | |
| 2023-10-23 22:15:14,301 DEV : loss 0.2551104426383972 - f1-score (micro avg) 0.7203 | |
| 2023-10-23 22:15:14,321 saving best model | |
| 2023-10-23 22:15:14,914 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:15:20,699 epoch 5 - iter 89/894 - loss 0.03423092 - time (sec): 5.78 - samples/sec: 1555.39 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-23 22:15:26,368 epoch 5 - iter 178/894 - loss 0.03919786 - time (sec): 11.45 - samples/sec: 1529.70 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-23 22:15:31,920 epoch 5 - iter 267/894 - loss 0.03831350 - time (sec): 17.00 - samples/sec: 1519.52 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-23 22:15:37,645 epoch 5 - iter 356/894 - loss 0.03789332 - time (sec): 22.73 - samples/sec: 1535.60 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-23 22:15:43,563 epoch 5 - iter 445/894 - loss 0.03759650 - time (sec): 28.65 - samples/sec: 1560.25 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-23 22:15:49,030 epoch 5 - iter 534/894 - loss 0.03829895 - time (sec): 34.11 - samples/sec: 1540.36 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-23 22:15:54,744 epoch 5 - iter 623/894 - loss 0.03927776 - time (sec): 39.83 - samples/sec: 1531.49 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-23 22:16:00,300 epoch 5 - iter 712/894 - loss 0.03884967 - time (sec): 45.39 - samples/sec: 1533.98 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-23 22:16:05,845 epoch 5 - iter 801/894 - loss 0.03960733 - time (sec): 50.93 - samples/sec: 1521.58 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-23 22:16:11,463 epoch 5 - iter 890/894 - loss 0.03948272 - time (sec): 56.55 - samples/sec: 1519.82 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-23 22:16:11,769 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:16:11,769 EPOCH 5 done: loss 0.0394 - lr: 0.000028 | |
| 2023-10-23 22:16:18,259 DEV : loss 0.2507097125053406 - f1-score (micro avg) 0.7541 | |
| 2023-10-23 22:16:18,279 saving best model | |
| 2023-10-23 22:16:18,871 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:16:24,253 epoch 6 - iter 89/894 - loss 0.03427358 - time (sec): 5.38 - samples/sec: 1390.08 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-23 22:16:29,876 epoch 6 - iter 178/894 - loss 0.03451237 - time (sec): 11.00 - samples/sec: 1458.98 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-23 22:16:35,622 epoch 6 - iter 267/894 - loss 0.03090456 - time (sec): 16.75 - samples/sec: 1510.50 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-23 22:16:41,315 epoch 6 - iter 356/894 - loss 0.03413662 - time (sec): 22.44 - samples/sec: 1515.71 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-23 22:16:47,226 epoch 6 - iter 445/894 - loss 0.03121911 - time (sec): 28.35 - samples/sec: 1538.17 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-23 22:16:52,738 epoch 6 - iter 534/894 - loss 0.03023548 - time (sec): 33.87 - samples/sec: 1526.90 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-23 22:16:58,470 epoch 6 - iter 623/894 - loss 0.03072382 - time (sec): 39.60 - samples/sec: 1531.66 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-23 22:17:04,206 epoch 6 - iter 712/894 - loss 0.02973244 - time (sec): 45.33 - samples/sec: 1523.66 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-23 22:17:09,881 epoch 6 - iter 801/894 - loss 0.02952191 - time (sec): 51.01 - samples/sec: 1523.58 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-23 22:17:15,533 epoch 6 - iter 890/894 - loss 0.02936006 - time (sec): 56.66 - samples/sec: 1521.80 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-23 22:17:15,773 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:17:15,773 EPOCH 6 done: loss 0.0294 - lr: 0.000022 | |
| 2023-10-23 22:17:22,244 DEV : loss 0.2560969591140747 - f1-score (micro avg) 0.7591 | |
| 2023-10-23 22:17:22,265 saving best model | |
| 2023-10-23 22:17:22,856 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:17:28,785 epoch 7 - iter 89/894 - loss 0.01473967 - time (sec): 5.93 - samples/sec: 1607.98 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-23 22:17:34,368 epoch 7 - iter 178/894 - loss 0.01596532 - time (sec): 11.51 - samples/sec: 1544.52 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-23 22:17:39,859 epoch 7 - iter 267/894 - loss 0.01847908 - time (sec): 17.00 - samples/sec: 1507.85 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-23 22:17:45,320 epoch 7 - iter 356/894 - loss 0.01735099 - time (sec): 22.46 - samples/sec: 1486.05 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-23 22:17:51,061 epoch 7 - iter 445/894 - loss 0.01666464 - time (sec): 28.20 - samples/sec: 1491.62 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-23 22:17:56,641 epoch 7 - iter 534/894 - loss 0.01833477 - time (sec): 33.78 - samples/sec: 1494.69 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-23 22:18:02,433 epoch 7 - iter 623/894 - loss 0.01813457 - time (sec): 39.58 - samples/sec: 1508.52 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-23 22:18:08,296 epoch 7 - iter 712/894 - loss 0.01766190 - time (sec): 45.44 - samples/sec: 1531.09 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-23 22:18:13,858 epoch 7 - iter 801/894 - loss 0.01731978 - time (sec): 51.00 - samples/sec: 1525.70 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-23 22:18:19,421 epoch 7 - iter 890/894 - loss 0.01885196 - time (sec): 56.56 - samples/sec: 1523.52 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-23 22:18:19,663 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:18:19,664 EPOCH 7 done: loss 0.0188 - lr: 0.000017 | |
| 2023-10-23 22:18:26,167 DEV : loss 0.25700417160987854 - f1-score (micro avg) 0.7673 | |
| 2023-10-23 22:18:26,188 saving best model | |
| 2023-10-23 22:18:26,776 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:18:32,428 epoch 8 - iter 89/894 - loss 0.00831319 - time (sec): 5.65 - samples/sec: 1519.15 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-23 22:18:38,212 epoch 8 - iter 178/894 - loss 0.00983876 - time (sec): 11.44 - samples/sec: 1510.57 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-23 22:18:43,974 epoch 8 - iter 267/894 - loss 0.01317277 - time (sec): 17.20 - samples/sec: 1532.68 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-23 22:18:49,909 epoch 8 - iter 356/894 - loss 0.01232841 - time (sec): 23.13 - samples/sec: 1547.32 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-23 22:18:55,332 epoch 8 - iter 445/894 - loss 0.01188126 - time (sec): 28.55 - samples/sec: 1522.90 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-23 22:19:00,901 epoch 8 - iter 534/894 - loss 0.01188323 - time (sec): 34.12 - samples/sec: 1525.83 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-23 22:19:06,449 epoch 8 - iter 623/894 - loss 0.01189307 - time (sec): 39.67 - samples/sec: 1523.80 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-23 22:19:11,886 epoch 8 - iter 712/894 - loss 0.01156878 - time (sec): 45.11 - samples/sec: 1508.18 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-23 22:19:17,744 epoch 8 - iter 801/894 - loss 0.01124655 - time (sec): 50.97 - samples/sec: 1518.04 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-23 22:19:23,437 epoch 8 - iter 890/894 - loss 0.01105335 - time (sec): 56.66 - samples/sec: 1522.10 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-23 22:19:23,684 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:19:23,685 EPOCH 8 done: loss 0.0110 - lr: 0.000011 | |
| 2023-10-23 22:19:30,172 DEV : loss 0.2922624945640564 - f1-score (micro avg) 0.7633 | |
| 2023-10-23 22:19:30,192 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:19:35,679 epoch 9 - iter 89/894 - loss 0.00212849 - time (sec): 5.49 - samples/sec: 1488.33 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-23 22:19:41,242 epoch 9 - iter 178/894 - loss 0.00466232 - time (sec): 11.05 - samples/sec: 1484.55 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-23 22:19:46,792 epoch 9 - iter 267/894 - loss 0.00624334 - time (sec): 16.60 - samples/sec: 1508.26 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-23 22:19:52,261 epoch 9 - iter 356/894 - loss 0.00529657 - time (sec): 22.07 - samples/sec: 1503.39 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-23 22:19:57,886 epoch 9 - iter 445/894 - loss 0.00482331 - time (sec): 27.69 - samples/sec: 1508.08 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-23 22:20:03,869 epoch 9 - iter 534/894 - loss 0.00634234 - time (sec): 33.68 - samples/sec: 1536.08 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-23 22:20:09,597 epoch 9 - iter 623/894 - loss 0.00603974 - time (sec): 39.40 - samples/sec: 1530.42 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-23 22:20:15,602 epoch 9 - iter 712/894 - loss 0.00635560 - time (sec): 45.41 - samples/sec: 1537.72 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-23 22:20:21,078 epoch 9 - iter 801/894 - loss 0.00602059 - time (sec): 50.88 - samples/sec: 1524.88 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-23 22:20:26,769 epoch 9 - iter 890/894 - loss 0.00592333 - time (sec): 56.58 - samples/sec: 1525.81 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-23 22:20:27,004 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:20:27,005 EPOCH 9 done: loss 0.0060 - lr: 0.000006 | |
| 2023-10-23 22:20:33,496 DEV : loss 0.29060593247413635 - f1-score (micro avg) 0.7681 | |
| 2023-10-23 22:20:33,516 saving best model | |
| 2023-10-23 22:20:34,105 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:20:39,963 epoch 10 - iter 89/894 - loss 0.00440159 - time (sec): 5.86 - samples/sec: 1526.28 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-23 22:20:45,656 epoch 10 - iter 178/894 - loss 0.00306763 - time (sec): 11.55 - samples/sec: 1499.98 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-23 22:20:51,138 epoch 10 - iter 267/894 - loss 0.00293745 - time (sec): 17.03 - samples/sec: 1528.11 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-23 22:20:56,938 epoch 10 - iter 356/894 - loss 0.00220417 - time (sec): 22.83 - samples/sec: 1548.00 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-23 22:21:02,446 epoch 10 - iter 445/894 - loss 0.00225457 - time (sec): 28.34 - samples/sec: 1522.61 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-23 22:21:07,950 epoch 10 - iter 534/894 - loss 0.00208242 - time (sec): 33.84 - samples/sec: 1517.45 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-23 22:21:13,674 epoch 10 - iter 623/894 - loss 0.00254697 - time (sec): 39.57 - samples/sec: 1519.31 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-23 22:21:19,162 epoch 10 - iter 712/894 - loss 0.00249384 - time (sec): 45.06 - samples/sec: 1512.52 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-23 22:21:24,848 epoch 10 - iter 801/894 - loss 0.00296117 - time (sec): 50.74 - samples/sec: 1514.47 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-23 22:21:30,534 epoch 10 - iter 890/894 - loss 0.00283140 - time (sec): 56.43 - samples/sec: 1514.86 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-23 22:21:31,030 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:21:31,030 EPOCH 10 done: loss 0.0028 - lr: 0.000000 | |
| 2023-10-23 22:21:37,226 DEV : loss 0.291725754737854 - f1-score (micro avg) 0.7739 | |
| 2023-10-23 22:21:37,246 saving best model | |
| 2023-10-23 22:21:38,317 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 22:21:38,317 Loading model from best epoch ... | |
| 2023-10-23 22:21:40,333 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 22:21:44,886 | |
| Results: | |
| - F-score (micro) 0.7501 | |
| - F-score (macro) 0.6739 | |
| - Accuracy 0.6174 | |
| By class: | |
| precision recall f1-score support | |
| loc 0.8147 0.8557 0.8347 596 | |
| pers 0.6868 0.7508 0.7174 333 | |
| org 0.5537 0.5076 0.5296 132 | |
| prod 0.6491 0.5606 0.6016 66 | |
| time 0.6604 0.7143 0.6863 49 | |
| micro avg 0.7363 0.7645 0.7501 1176 | |
| macro avg 0.6729 0.6778 0.6739 1176 | |
| weighted avg 0.7335 0.7645 0.7480 1176 | |
| 2023-10-23 22:21:44,886 ---------------------------------------------------------------------------------------------------- | |