| 2023-10-23 20:34:46,257 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,258 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 20:34:46,258 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,258 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 20:34:46,258 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,258 Train: 3575 sentences | |
| 2023-10-23 20:34:46,258 (train_with_dev=False, train_with_test=False) | |
| 2023-10-23 20:34:46,258 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,258 Training Params: | |
| 2023-10-23 20:34:46,258 - learning_rate: "5e-05" | |
| 2023-10-23 20:34:46,258 - mini_batch_size: "8" | |
| 2023-10-23 20:34:46,258 - max_epochs: "10" | |
| 2023-10-23 20:34:46,258 - shuffle: "True" | |
| 2023-10-23 20:34:46,258 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,258 Plugins: | |
| 2023-10-23 20:34:46,258 - TensorboardLogger | |
| 2023-10-23 20:34:46,258 - LinearScheduler | warmup_fraction: '0.1' | |
| 2023-10-23 20:34:46,258 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,258 Final evaluation on model from best epoch (best-model.pt) | |
| 2023-10-23 20:34:46,258 - metric: "('micro avg', 'f1-score')" | |
| 2023-10-23 20:34:46,258 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,258 Computation: | |
| 2023-10-23 20:34:46,258 - compute on device: cuda:0 | |
| 2023-10-23 20:34:46,258 - embedding storage: none | |
| 2023-10-23 20:34:46,258 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,259 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1" | |
| 2023-10-23 20:34:46,259 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,259 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:34:46,259 Logging anything other than scalars to TensorBoard is currently not supported. | |
| 2023-10-23 20:34:50,214 epoch 1 - iter 44/447 - loss 3.08591702 - time (sec): 3.95 - samples/sec: 2151.57 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-23 20:34:53,927 epoch 1 - iter 88/447 - loss 1.92199617 - time (sec): 7.67 - samples/sec: 2132.97 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-23 20:34:57,836 epoch 1 - iter 132/447 - loss 1.39896841 - time (sec): 11.58 - samples/sec: 2162.64 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-23 20:35:01,868 epoch 1 - iter 176/447 - loss 1.12827946 - time (sec): 15.61 - samples/sec: 2127.79 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-23 20:35:05,662 epoch 1 - iter 220/447 - loss 0.96436501 - time (sec): 19.40 - samples/sec: 2147.43 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-23 20:35:09,538 epoch 1 - iter 264/447 - loss 0.84161150 - time (sec): 23.28 - samples/sec: 2137.60 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-23 20:35:13,581 epoch 1 - iter 308/447 - loss 0.75094933 - time (sec): 27.32 - samples/sec: 2132.17 - lr: 0.000034 - momentum: 0.000000 | |
| 2023-10-23 20:35:18,004 epoch 1 - iter 352/447 - loss 0.67600080 - time (sec): 31.74 - samples/sec: 2139.46 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-23 20:35:22,004 epoch 1 - iter 396/447 - loss 0.62004085 - time (sec): 35.74 - samples/sec: 2148.08 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-23 20:35:26,004 epoch 1 - iter 440/447 - loss 0.58216934 - time (sec): 39.74 - samples/sec: 2148.19 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-23 20:35:26,580 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:35:26,580 EPOCH 1 done: loss 0.5760 - lr: 0.000049 | |
| 2023-10-23 20:35:31,379 DEV : loss 0.1480439156293869 - f1-score (micro avg) 0.663 | |
| 2023-10-23 20:35:31,399 saving best model | |
| 2023-10-23 20:35:31,947 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:35:36,424 epoch 2 - iter 44/447 - loss 0.15891281 - time (sec): 4.48 - samples/sec: 2130.44 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-23 20:35:40,292 epoch 2 - iter 88/447 - loss 0.14907678 - time (sec): 8.34 - samples/sec: 2134.91 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-23 20:35:44,386 epoch 2 - iter 132/447 - loss 0.14353574 - time (sec): 12.44 - samples/sec: 2100.19 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-23 20:35:48,177 epoch 2 - iter 176/447 - loss 0.14493075 - time (sec): 16.23 - samples/sec: 2121.25 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-23 20:35:52,020 epoch 2 - iter 220/447 - loss 0.14099654 - time (sec): 20.07 - samples/sec: 2112.92 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-23 20:35:56,054 epoch 2 - iter 264/447 - loss 0.14001060 - time (sec): 24.11 - samples/sec: 2122.33 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-23 20:36:00,154 epoch 2 - iter 308/447 - loss 0.13478504 - time (sec): 28.21 - samples/sec: 2129.26 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-23 20:36:04,356 epoch 2 - iter 352/447 - loss 0.13430691 - time (sec): 32.41 - samples/sec: 2123.93 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-23 20:36:08,229 epoch 2 - iter 396/447 - loss 0.13148998 - time (sec): 36.28 - samples/sec: 2118.19 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-23 20:36:12,037 epoch 2 - iter 440/447 - loss 0.12964162 - time (sec): 40.09 - samples/sec: 2124.43 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-23 20:36:12,661 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:36:12,661 EPOCH 2 done: loss 0.1306 - lr: 0.000045 | |
| 2023-10-23 20:36:19,168 DEV : loss 0.14241820573806763 - f1-score (micro avg) 0.6282 | |
| 2023-10-23 20:36:19,189 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:36:23,034 epoch 3 - iter 44/447 - loss 0.06032349 - time (sec): 3.84 - samples/sec: 2092.70 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-23 20:36:27,150 epoch 3 - iter 88/447 - loss 0.06341120 - time (sec): 7.96 - samples/sec: 2146.02 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-23 20:36:31,170 epoch 3 - iter 132/447 - loss 0.06583153 - time (sec): 11.98 - samples/sec: 2102.85 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-23 20:36:35,055 epoch 3 - iter 176/447 - loss 0.07127918 - time (sec): 15.87 - samples/sec: 2120.46 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-23 20:36:38,731 epoch 3 - iter 220/447 - loss 0.07341253 - time (sec): 19.54 - samples/sec: 2113.76 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-23 20:36:42,651 epoch 3 - iter 264/447 - loss 0.07021264 - time (sec): 23.46 - samples/sec: 2133.69 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-23 20:36:46,615 epoch 3 - iter 308/447 - loss 0.07048521 - time (sec): 27.43 - samples/sec: 2135.27 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-23 20:36:50,424 epoch 3 - iter 352/447 - loss 0.07189812 - time (sec): 31.23 - samples/sec: 2132.33 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-23 20:36:54,663 epoch 3 - iter 396/447 - loss 0.07322618 - time (sec): 35.47 - samples/sec: 2126.10 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-23 20:36:58,537 epoch 3 - iter 440/447 - loss 0.07165755 - time (sec): 39.35 - samples/sec: 2138.37 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-23 20:36:59,519 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:36:59,520 EPOCH 3 done: loss 0.0713 - lr: 0.000039 | |
| 2023-10-23 20:37:06,006 DEV : loss 0.1675412803888321 - f1-score (micro avg) 0.7561 | |
| 2023-10-23 20:37:06,026 saving best model | |
| 2023-10-23 20:37:06,800 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:37:11,011 epoch 4 - iter 44/447 - loss 0.05463008 - time (sec): 4.21 - samples/sec: 2139.44 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-23 20:37:14,807 epoch 4 - iter 88/447 - loss 0.04619879 - time (sec): 8.01 - samples/sec: 2158.54 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-23 20:37:18,672 epoch 4 - iter 132/447 - loss 0.04878841 - time (sec): 11.87 - samples/sec: 2152.81 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-23 20:37:22,367 epoch 4 - iter 176/447 - loss 0.04429914 - time (sec): 15.57 - samples/sec: 2154.49 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-23 20:37:26,630 epoch 4 - iter 220/447 - loss 0.04506808 - time (sec): 19.83 - samples/sec: 2147.70 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-23 20:37:30,420 epoch 4 - iter 264/447 - loss 0.04554775 - time (sec): 23.62 - samples/sec: 2128.75 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-23 20:37:34,240 epoch 4 - iter 308/447 - loss 0.04617356 - time (sec): 27.44 - samples/sec: 2133.27 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-23 20:37:38,186 epoch 4 - iter 352/447 - loss 0.04610295 - time (sec): 31.39 - samples/sec: 2128.76 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-23 20:37:42,725 epoch 4 - iter 396/447 - loss 0.04655391 - time (sec): 35.92 - samples/sec: 2126.27 - lr: 0.000034 - momentum: 0.000000 | |
| 2023-10-23 20:37:46,676 epoch 4 - iter 440/447 - loss 0.04692162 - time (sec): 39.87 - samples/sec: 2134.89 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-23 20:37:47,359 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:37:47,359 EPOCH 4 done: loss 0.0469 - lr: 0.000033 | |
| 2023-10-23 20:37:53,844 DEV : loss 0.1665799915790558 - f1-score (micro avg) 0.74 | |
| 2023-10-23 20:37:53,865 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:37:57,539 epoch 5 - iter 44/447 - loss 0.04951173 - time (sec): 3.67 - samples/sec: 2080.67 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-23 20:38:01,398 epoch 5 - iter 88/447 - loss 0.04294255 - time (sec): 7.53 - samples/sec: 2096.59 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-23 20:38:05,711 epoch 5 - iter 132/447 - loss 0.04150257 - time (sec): 11.85 - samples/sec: 2092.61 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-23 20:38:09,518 epoch 5 - iter 176/447 - loss 0.03750633 - time (sec): 15.65 - samples/sec: 2115.03 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-23 20:38:13,240 epoch 5 - iter 220/447 - loss 0.03634709 - time (sec): 19.37 - samples/sec: 2126.46 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-23 20:38:17,675 epoch 5 - iter 264/447 - loss 0.03702155 - time (sec): 23.81 - samples/sec: 2132.75 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-23 20:38:21,375 epoch 5 - iter 308/447 - loss 0.03726439 - time (sec): 27.51 - samples/sec: 2144.51 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-23 20:38:25,414 epoch 5 - iter 352/447 - loss 0.03644814 - time (sec): 31.55 - samples/sec: 2155.96 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-23 20:38:29,193 epoch 5 - iter 396/447 - loss 0.03584344 - time (sec): 35.33 - samples/sec: 2146.52 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-23 20:38:33,763 epoch 5 - iter 440/447 - loss 0.03535728 - time (sec): 39.90 - samples/sec: 2136.98 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-23 20:38:34,376 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:38:34,376 EPOCH 5 done: loss 0.0356 - lr: 0.000028 | |
| 2023-10-23 20:38:40,847 DEV : loss 0.2225048989057541 - f1-score (micro avg) 0.7541 | |
| 2023-10-23 20:38:40,867 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:38:44,634 epoch 6 - iter 44/447 - loss 0.02861626 - time (sec): 3.77 - samples/sec: 2196.59 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-23 20:38:48,495 epoch 6 - iter 88/447 - loss 0.03090318 - time (sec): 7.63 - samples/sec: 2182.60 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-23 20:38:53,070 epoch 6 - iter 132/447 - loss 0.03202164 - time (sec): 12.20 - samples/sec: 2163.53 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-23 20:38:57,419 epoch 6 - iter 176/447 - loss 0.03010408 - time (sec): 16.55 - samples/sec: 2132.30 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-23 20:39:01,536 epoch 6 - iter 220/447 - loss 0.02822882 - time (sec): 20.67 - samples/sec: 2131.41 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-23 20:39:05,582 epoch 6 - iter 264/447 - loss 0.02593149 - time (sec): 24.71 - samples/sec: 2134.92 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-23 20:39:09,205 epoch 6 - iter 308/447 - loss 0.02519418 - time (sec): 28.34 - samples/sec: 2124.42 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-23 20:39:12,950 epoch 6 - iter 352/447 - loss 0.02614899 - time (sec): 32.08 - samples/sec: 2131.97 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-23 20:39:16,887 epoch 6 - iter 396/447 - loss 0.02612555 - time (sec): 36.02 - samples/sec: 2129.44 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-23 20:39:20,812 epoch 6 - iter 440/447 - loss 0.02515450 - time (sec): 39.94 - samples/sec: 2136.58 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-23 20:39:21,436 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:39:21,436 EPOCH 6 done: loss 0.0252 - lr: 0.000022 | |
| 2023-10-23 20:39:27,918 DEV : loss 0.24188880622386932 - f1-score (micro avg) 0.7539 | |
| 2023-10-23 20:39:27,939 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:39:32,281 epoch 7 - iter 44/447 - loss 0.03582544 - time (sec): 4.34 - samples/sec: 2165.08 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-23 20:39:36,428 epoch 7 - iter 88/447 - loss 0.02343980 - time (sec): 8.49 - samples/sec: 2104.12 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-23 20:39:40,637 epoch 7 - iter 132/447 - loss 0.02247574 - time (sec): 12.70 - samples/sec: 2128.59 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-23 20:39:44,459 epoch 7 - iter 176/447 - loss 0.02484634 - time (sec): 16.52 - samples/sec: 2120.83 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-23 20:39:48,293 epoch 7 - iter 220/447 - loss 0.02289894 - time (sec): 20.35 - samples/sec: 2111.43 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-23 20:39:52,124 epoch 7 - iter 264/447 - loss 0.02072518 - time (sec): 24.18 - samples/sec: 2123.10 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-23 20:39:56,375 epoch 7 - iter 308/447 - loss 0.02007036 - time (sec): 28.44 - samples/sec: 2125.41 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-23 20:40:00,110 epoch 7 - iter 352/447 - loss 0.01916875 - time (sec): 32.17 - samples/sec: 2143.94 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-23 20:40:04,092 epoch 7 - iter 396/447 - loss 0.01862712 - time (sec): 36.15 - samples/sec: 2127.24 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-23 20:40:07,948 epoch 7 - iter 440/447 - loss 0.01780365 - time (sec): 40.01 - samples/sec: 2138.09 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-23 20:40:08,469 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:40:08,469 EPOCH 7 done: loss 0.0178 - lr: 0.000017 | |
| 2023-10-23 20:40:14,944 DEV : loss 0.26893866062164307 - f1-score (micro avg) 0.7761 | |
| 2023-10-23 20:40:14,964 saving best model | |
| 2023-10-23 20:40:15,678 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:40:19,533 epoch 8 - iter 44/447 - loss 0.00921673 - time (sec): 3.85 - samples/sec: 2158.83 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-23 20:40:23,365 epoch 8 - iter 88/447 - loss 0.01119385 - time (sec): 7.69 - samples/sec: 2183.92 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-23 20:40:27,948 epoch 8 - iter 132/447 - loss 0.01095721 - time (sec): 12.27 - samples/sec: 2125.10 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-23 20:40:31,658 epoch 8 - iter 176/447 - loss 0.01028475 - time (sec): 15.98 - samples/sec: 2150.30 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-23 20:40:35,750 epoch 8 - iter 220/447 - loss 0.00902436 - time (sec): 20.07 - samples/sec: 2145.08 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-23 20:40:39,401 epoch 8 - iter 264/447 - loss 0.00849857 - time (sec): 23.72 - samples/sec: 2135.50 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-23 20:40:43,377 epoch 8 - iter 308/447 - loss 0.00962661 - time (sec): 27.70 - samples/sec: 2130.49 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-23 20:40:47,327 epoch 8 - iter 352/447 - loss 0.01052978 - time (sec): 31.65 - samples/sec: 2137.57 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-23 20:40:51,295 epoch 8 - iter 396/447 - loss 0.01131149 - time (sec): 35.62 - samples/sec: 2139.29 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-23 20:40:55,580 epoch 8 - iter 440/447 - loss 0.01174824 - time (sec): 39.90 - samples/sec: 2136.08 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-23 20:40:56,194 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:40:56,194 EPOCH 8 done: loss 0.0120 - lr: 0.000011 | |
| 2023-10-23 20:41:02,415 DEV : loss 0.2459404468536377 - f1-score (micro avg) 0.7574 | |
| 2023-10-23 20:41:02,436 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:41:06,866 epoch 9 - iter 44/447 - loss 0.03433726 - time (sec): 4.43 - samples/sec: 2016.40 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-23 20:41:11,137 epoch 9 - iter 88/447 - loss 0.03002005 - time (sec): 8.70 - samples/sec: 2071.75 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-23 20:41:15,202 epoch 9 - iter 132/447 - loss 0.02907463 - time (sec): 12.77 - samples/sec: 2064.68 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-23 20:41:18,874 epoch 9 - iter 176/447 - loss 0.02816834 - time (sec): 16.44 - samples/sec: 2056.26 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-23 20:41:22,601 epoch 9 - iter 220/447 - loss 0.02425467 - time (sec): 20.16 - samples/sec: 2068.35 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-23 20:41:26,685 epoch 9 - iter 264/447 - loss 0.02386885 - time (sec): 24.25 - samples/sec: 2078.56 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-23 20:41:30,383 epoch 9 - iter 308/447 - loss 0.02183922 - time (sec): 27.95 - samples/sec: 2096.13 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-23 20:41:34,996 epoch 9 - iter 352/447 - loss 0.02186201 - time (sec): 32.56 - samples/sec: 2127.54 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-23 20:41:38,749 epoch 9 - iter 396/447 - loss 0.02183062 - time (sec): 36.31 - samples/sec: 2134.87 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-23 20:41:42,489 epoch 9 - iter 440/447 - loss 0.02414038 - time (sec): 40.05 - samples/sec: 2131.89 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-23 20:41:43,068 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:41:43,068 EPOCH 9 done: loss 0.0241 - lr: 0.000006 | |
| 2023-10-23 20:41:49,270 DEV : loss 0.277804434299469 - f1-score (micro avg) 0.6578 | |
| 2023-10-23 20:41:49,290 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:41:53,441 epoch 10 - iter 44/447 - loss 0.07121556 - time (sec): 4.15 - samples/sec: 2054.74 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-23 20:41:57,288 epoch 10 - iter 88/447 - loss 0.06814730 - time (sec): 8.00 - samples/sec: 2138.55 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-23 20:42:01,098 epoch 10 - iter 132/447 - loss 0.06359458 - time (sec): 11.81 - samples/sec: 2134.13 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-23 20:42:05,202 epoch 10 - iter 176/447 - loss 0.06000699 - time (sec): 15.91 - samples/sec: 2094.21 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-23 20:42:09,004 epoch 10 - iter 220/447 - loss 0.05728519 - time (sec): 19.71 - samples/sec: 2098.42 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-23 20:42:12,907 epoch 10 - iter 264/447 - loss 0.05401101 - time (sec): 23.62 - samples/sec: 2103.75 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-23 20:42:16,768 epoch 10 - iter 308/447 - loss 0.05440301 - time (sec): 27.48 - samples/sec: 2099.70 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-23 20:42:20,489 epoch 10 - iter 352/447 - loss 0.05284119 - time (sec): 31.20 - samples/sec: 2115.34 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-23 20:42:25,162 epoch 10 - iter 396/447 - loss 0.05118931 - time (sec): 35.87 - samples/sec: 2132.18 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-23 20:42:28,992 epoch 10 - iter 440/447 - loss 0.05044650 - time (sec): 39.70 - samples/sec: 2128.66 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-23 20:42:29,908 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:42:29,908 EPOCH 10 done: loss 0.0503 - lr: 0.000000 | |
| 2023-10-23 20:42:36,109 DEV : loss 0.2561439871788025 - f1-score (micro avg) 0.6432 | |
| 2023-10-23 20:42:36,679 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-23 20:42:36,680 Loading model from best epoch ... | |
| 2023-10-23 20:42:38,467 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 20:42:43,278 | |
| Results: | |
| - F-score (micro) 0.7458 | |
| - F-score (macro) 0.6623 | |
| - Accuracy 0.6117 | |
| By class: | |
| precision recall f1-score support | |
| loc 0.8300 0.8356 0.8328 596 | |
| pers 0.6711 0.7598 0.7127 333 | |
| org 0.5688 0.4697 0.5145 132 | |
| prod 0.7297 0.4091 0.5243 66 | |
| time 0.7200 0.7347 0.7273 49 | |
| micro avg 0.7468 0.7449 0.7458 1176 | |
| macro avg 0.7039 0.6418 0.6623 1176 | |
| weighted avg 0.7455 0.7449 0.7413 1176 | |
| 2023-10-23 20:42:43,279 ---------------------------------------------------------------------------------------------------- | |