2022-02-04 12:18:14,159 ----------------------------------------------------------------------------------------------------
2022-02-04 12:18:14,161 Model: "SequenceTagger(
  (embeddings): TransformerWordEmbeddings(
    (model): CamembertModel(
      (embeddings): RobertaEmbeddings(
        (word_embeddings): Embedding(32005, 768, padding_idx=1)
        (position_embeddings): Embedding(514, 768, padding_idx=1)
        (token_type_embeddings): Embedding(1, 768)
        (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
        (dropout): Dropout(p=0.1, inplace=False)
      )
      (encoder): RobertaEncoder(
        (layer): ModuleList(
          (0): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (1): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (2): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (3): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (4): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (5): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (6): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (7): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (8): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (9): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (10): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (11): RobertaLayer(
            (attention): RobertaAttention(
              (self): RobertaSelfAttention(
                (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): RobertaSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): RobertaIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
            )
            (output): RobertaOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
        )
      )
      (pooler): RobertaPooler(
        (dense): Linear(in_features=768, out_features=768, bias=True)
        (activation): Tanh()
      )
    )
  )
  (word_dropout): WordDropout(p=0.05)
  (locked_dropout): LockedDropout(p=0.5)
  (linear): Linear(in_features=768, out_features=18, bias=True)
  (beta): 1.0
  (weights): None
  (weight_tensor) None
)"
2022-02-04 12:18:14,167 ----------------------------------------------------------------------------------------------------
2022-02-04 12:18:14,167 Corpus: "Corpus: 126973 train + 7037 dev + 7090 test sentences"
2022-02-04 12:18:14,167 ----------------------------------------------------------------------------------------------------
2022-02-04 12:18:14,167 Parameters:
2022-02-04 12:18:14,167  - learning_rate: "5e-05"
2022-02-04 12:18:14,167  - mini_batch_size: "16"
2022-02-04 12:18:14,167  - patience: "3"
2022-02-04 12:18:14,167  - anneal_factor: "0.5"
2022-02-04 12:18:14,167  - max_epochs: "10"
2022-02-04 12:18:14,167  - shuffle: "True"
2022-02-04 12:18:14,167  - train_with_dev: "False"
2022-02-04 12:18:14,167  - batch_growth_annealing: "False"
2022-02-04 12:18:14,167 ----------------------------------------------------------------------------------------------------
2022-02-04 12:18:14,167 Model training base path: "resources/taggers/ner-camembert"
2022-02-04 12:18:14,167 ----------------------------------------------------------------------------------------------------
2022-02-04 12:18:14,167 Device: cuda:0
2022-02-04 12:18:14,167 ----------------------------------------------------------------------------------------------------
2022-02-04 12:18:14,167 Embeddings storage mode: none
2022-02-04 12:18:14,170 ----------------------------------------------------------------------------------------------------
2022-02-04 12:25:23,397 epoch 1 - iter 793/7936 - loss 1.64849782 - samples/sec: 29.56 - lr: 0.000005
2022-02-04 12:33:59,649 epoch 1 - iter 1586/7936 - loss 1.11222779 - samples/sec: 24.58 - lr: 0.000010
2022-02-04 12:41:09,132 epoch 1 - iter 2379/7936 - loss 0.85257016 - samples/sec: 29.55 - lr: 0.000015
2022-02-04 12:47:44,896 epoch 1 - iter 3172/7936 - loss 0.71981753 - samples/sec: 32.07 - lr: 0.000020
2022-02-04 12:55:15,449 epoch 1 - iter 3965/7936 - loss 0.60512907 - samples/sec: 28.16 - lr: 0.000025
2022-02-04 13:02:35,238 epoch 1 - iter 4758/7936 - loss 0.52903622 - samples/sec: 28.85 - lr: 0.000030
2022-02-04 13:09:27,012 epoch 1 - iter 5551/7936 - loss 0.48171220 - samples/sec: 30.82 - lr: 0.000035
2022-02-04 13:15:53,083 epoch 1 - iter 6344/7936 - loss 0.44948661 - samples/sec: 32.87 - lr: 0.000040
2022-02-04 13:22:02,650 epoch 1 - iter 7137/7936 - loss 0.42228564 - samples/sec: 34.34 - lr: 0.000045
2022-02-04 13:28:59,445 epoch 1 - iter 7930/7936 - loss 0.39366725 - samples/sec: 30.45 - lr: 0.000050
2022-02-04 13:29:03,026 ----------------------------------------------------------------------------------------------------
2022-02-04 13:29:03,028 EPOCH 1 done: loss 0.3935 - lr 0.0000500
2022-02-04 13:32:00,102 DEV : loss 0.038586683571338654 - f1-score (micro avg)  0.8195
2022-02-04 13:32:00,155 BAD EPOCHS (no improvement): 4
2022-02-04 13:32:00,156 ----------------------------------------------------------------------------------------------------
2022-02-04 13:39:12,612 epoch 2 - iter 793/7936 - loss 0.14931520 - samples/sec: 29.34 - lr: 0.000049
2022-02-04 13:46:36,550 epoch 2 - iter 1586/7936 - loss 0.14672871 - samples/sec: 28.58 - lr: 0.000049
2022-02-04 13:53:49,885 epoch 2 - iter 2379/7936 - loss 0.14547274 - samples/sec: 29.28 - lr: 0.000048
2022-02-04 14:01:13,739 epoch 2 - iter 3172/7936 - loss 0.14418846 - samples/sec: 28.59 - lr: 0.000048
2022-02-04 14:08:30,985 epoch 2 - iter 3965/7936 - loss 0.14265825 - samples/sec: 29.02 - lr: 0.000047
2022-02-04 14:15:46,742 epoch 2 - iter 4758/7936 - loss 0.14086599 - samples/sec: 29.12 - lr: 0.000047
2022-02-04 14:23:11,181 epoch 2 - iter 5551/7936 - loss 0.13927378 - samples/sec: 28.55 - lr: 0.000046
2022-02-04 14:30:19,706 epoch 2 - iter 6344/7936 - loss 0.13799042 - samples/sec: 29.61 - lr: 0.000046
2022-02-04 14:37:30,554 epoch 2 - iter 7137/7936 - loss 0.13666296 - samples/sec: 29.45 - lr: 0.000045
2022-02-04 14:44:52,886 epoch 2 - iter 7930/7936 - loss 0.13525042 - samples/sec: 28.69 - lr: 0.000044
2022-02-04 14:44:56,060 ----------------------------------------------------------------------------------------------------
2022-02-04 14:44:56,062 EPOCH 2 done: loss 0.1352 - lr 0.0000444
2022-02-04 14:47:40,950 DEV : loss 0.015217592008411884 - f1-score (micro avg)  0.9164
2022-02-04 14:47:41,011 BAD EPOCHS (no improvement): 4
2022-02-04 14:47:41,014 ----------------------------------------------------------------------------------------------------
2022-02-04 14:55:04,697 epoch 3 - iter 793/7936 - loss 0.11742558 - samples/sec: 28.60 - lr: 0.000044
2022-02-04 15:02:16,388 epoch 3 - iter 1586/7936 - loss 0.11679901 - samples/sec: 29.40 - lr: 0.000043
2022-02-04 15:09:29,924 epoch 3 - iter 2379/7936 - loss 0.11557918 - samples/sec: 29.27 - lr: 0.000043
2022-02-04 15:16:54,356 epoch 3 - iter 3172/7936 - loss 0.11469700 - samples/sec: 28.55 - lr: 0.000042
2022-02-04 15:24:11,817 epoch 3 - iter 3965/7936 - loss 0.11351908 - samples/sec: 29.01 - lr: 0.000042
2022-02-04 15:31:20,620 epoch 3 - iter 4758/7936 - loss 0.11266101 - samples/sec: 29.59 - lr: 0.000041
2022-02-04 15:38:42,882 epoch 3 - iter 5551/7936 - loss 0.11158730 - samples/sec: 28.69 - lr: 0.000041
2022-02-04 15:45:50,317 epoch 3 - iter 6344/7936 - loss 0.11067669 - samples/sec: 29.69 - lr: 0.000040
2022-02-04 15:53:16,035 epoch 3 - iter 7137/7936 - loss 0.10955013 - samples/sec: 28.47 - lr: 0.000039
2022-02-04 16:00:25,858 epoch 3 - iter 7930/7936 - loss 0.10859645 - samples/sec: 29.52 - lr: 0.000039
2022-02-04 16:00:29,034 ----------------------------------------------------------------------------------------------------
2022-02-04 16:00:29,035 EPOCH 3 done: loss 0.1086 - lr 0.0000389
2022-02-04 16:03:24,201 DEV : loss 0.015040190890431404 - f1-score (micro avg)  0.9276
2022-02-04 16:03:24,261 BAD EPOCHS (no improvement): 4
2022-02-04 16:03:24,262 ----------------------------------------------------------------------------------------------------
2022-02-04 16:10:35,356 epoch 4 - iter 793/7936 - loss 0.09491620 - samples/sec: 29.44 - lr: 0.000038
2022-02-04 16:17:46,476 epoch 4 - iter 1586/7936 - loss 0.09400900 - samples/sec: 29.43 - lr: 0.000038
2022-02-04 16:25:10,503 epoch 4 - iter 2379/7936 - loss 0.09355228 - samples/sec: 28.58 - lr: 0.000037
2022-02-04 16:32:21,829 epoch 4 - iter 3172/7936 - loss 0.09257257 - samples/sec: 29.42 - lr: 0.000037
2022-02-04 16:39:34,717 epoch 4 - iter 3965/7936 - loss 0.09178491 - samples/sec: 29.31 - lr: 0.000036
2022-02-04 16:46:54,536 epoch 4 - iter 4758/7936 - loss 0.09102086 - samples/sec: 28.85 - lr: 0.000036
2022-02-04 16:54:08,674 epoch 4 - iter 5551/7936 - loss 0.09026061 - samples/sec: 29.23 - lr: 0.000035
2022-02-04 17:01:24,799 epoch 4 - iter 6344/7936 - loss 0.08942621 - samples/sec: 29.10 - lr: 0.000034
2022-02-04 17:08:44,577 epoch 4 - iter 7137/7936 - loss 0.08868927 - samples/sec: 28.85 - lr: 0.000034
2022-02-04 17:15:57,678 epoch 4 - iter 7930/7936 - loss 0.08790466 - samples/sec: 29.30 - lr: 0.000033
2022-02-04 17:16:00,787 ----------------------------------------------------------------------------------------------------
2022-02-04 17:16:00,790 EPOCH 4 done: loss 0.0879 - lr 0.0000333
2022-02-04 17:18:55,805 DEV : loss 0.015710221603512764 - f1-score (micro avg)  0.9308
2022-02-04 17:18:55,865 BAD EPOCHS (no improvement): 4
2022-02-04 17:18:55,873 ----------------------------------------------------------------------------------------------------
2022-02-04 17:26:02,969 epoch 5 - iter 793/7936 - loss 0.07683748 - samples/sec: 29.71 - lr: 0.000033
2022-02-04 17:33:13,355 epoch 5 - iter 1586/7936 - loss 0.07621969 - samples/sec: 29.49 - lr: 0.000032
2022-02-04 17:40:38,247 epoch 5 - iter 2379/7936 - loss 0.07573593 - samples/sec: 28.52 - lr: 0.000032
2022-02-04 17:47:40,269 epoch 5 - iter 3172/7936 - loss 0.07524740 - samples/sec: 30.07 - lr: 0.000031
2022-02-04 17:54:59,036 epoch 5 - iter 3965/7936 - loss 0.07449799 - samples/sec: 28.92 - lr: 0.000031
2022-02-04 18:02:03,686 epoch 5 - iter 4758/7936 - loss 0.07405311 - samples/sec: 29.88 - lr: 0.000030
2022-02-04 18:09:11,646 epoch 5 - iter 5551/7936 - loss 0.07340830 - samples/sec: 29.65 - lr: 0.000029
2022-02-04 18:16:27,240 epoch 5 - iter 6344/7936 - loss 0.07271787 - samples/sec: 29.13 - lr: 0.000029
2022-02-04 18:23:29,669 epoch 5 - iter 7137/7936 - loss 0.07217288 - samples/sec: 30.04 - lr: 0.000028
2022-02-04 18:30:30,597 epoch 5 - iter 7930/7936 - loss 0.07166288 - samples/sec: 30.15 - lr: 0.000028
2022-02-04 18:30:33,919 ----------------------------------------------------------------------------------------------------
2022-02-04 18:30:33,920 EPOCH 5 done: loss 0.0717 - lr 0.0000278
2022-02-04 18:33:23,923 DEV : loss 0.017801353707909584 - f1-score (micro avg)  0.9319
2022-02-04 18:33:23,983 BAD EPOCHS (no improvement): 4
2022-02-04 18:33:23,983 ----------------------------------------------------------------------------------------------------
2022-02-04 18:40:28,017 epoch 6 - iter 793/7936 - loss 0.06265627 - samples/sec: 29.93 - lr: 0.000027
2022-02-04 18:47:46,740 epoch 6 - iter 1586/7936 - loss 0.06168821 - samples/sec: 28.92 - lr: 0.000027
2022-02-04 18:54:59,429 epoch 6 - iter 2379/7936 - loss 0.06137959 - samples/sec: 29.33 - lr: 0.000026
2022-02-04 19:02:08,367 epoch 6 - iter 3172/7936 - loss 0.06101991 - samples/sec: 29.58 - lr: 0.000026
2022-02-04 19:09:34,369 epoch 6 - iter 3965/7936 - loss 0.06073221 - samples/sec: 28.45 - lr: 0.000025
2022-02-04 19:16:53,646 epoch 6 - iter 4758/7936 - loss 0.06031513 - samples/sec: 28.89 - lr: 0.000024
2022-02-04 19:24:05,427 epoch 6 - iter 5551/7936 - loss 0.05997466 - samples/sec: 29.39 - lr: 0.000024
2022-02-04 19:31:27,470 epoch 6 - iter 6344/7936 - loss 0.05952743 - samples/sec: 28.71 - lr: 0.000023
2022-02-04 19:38:37,449 epoch 6 - iter 7137/7936 - loss 0.05906427 - samples/sec: 29.51 - lr: 0.000023
2022-02-04 19:46:02,608 epoch 6 - iter 7930/7936 - loss 0.05868560 - samples/sec: 28.51 - lr: 0.000022
2022-02-04 19:46:05,790 ----------------------------------------------------------------------------------------------------
2022-02-04 19:46:05,791 EPOCH 6 done: loss 0.0587 - lr 0.0000222
2022-02-04 19:48:52,058 DEV : loss 0.018429730087518692 - f1-score (micro avg)  0.9371
2022-02-04 19:48:52,117 BAD EPOCHS (no improvement): 4
2022-02-04 19:48:52,118 ----------------------------------------------------------------------------------------------------
2022-02-04 19:56:15,841 epoch 7 - iter 793/7936 - loss 0.05186660 - samples/sec: 28.60 - lr: 0.000022
2022-02-04 20:03:27,574 epoch 7 - iter 1586/7936 - loss 0.05230029 - samples/sec: 29.39 - lr: 0.000021
2022-02-04 20:10:42,349 epoch 7 - iter 2379/7936 - loss 0.05178480 - samples/sec: 29.19 - lr: 0.000021
2022-02-04 20:18:09,822 epoch 7 - iter 3172/7936 - loss 0.05114746 - samples/sec: 28.36 - lr: 0.000020
2022-02-04 20:25:23,574 epoch 7 - iter 3965/7936 - loss 0.05080701 - samples/sec: 29.26 - lr: 0.000019
2022-02-04 20:32:39,287 epoch 7 - iter 4758/7936 - loss 0.05039880 - samples/sec: 29.12 - lr: 0.000019
2022-02-04 20:40:04,807 epoch 7 - iter 5551/7936 - loss 0.05020234 - samples/sec: 28.48 - lr: 0.000018
2022-02-04 20:47:17,356 epoch 7 - iter 6344/7936 - loss 0.04984342 - samples/sec: 29.34 - lr: 0.000018
2022-02-04 20:54:31,673 epoch 7 - iter 7137/7936 - loss 0.04955538 - samples/sec: 29.22 - lr: 0.000017
2022-02-04 21:01:58,187 epoch 7 - iter 7930/7936 - loss 0.04921375 - samples/sec: 28.42 - lr: 0.000017
2022-02-04 21:02:01,071 ----------------------------------------------------------------------------------------------------
2022-02-04 21:02:01,071 EPOCH 7 done: loss 0.0492 - lr 0.0000167
2022-02-04 21:04:47,460 DEV : loss 0.02109825611114502 - f1-score (micro avg)  0.9362
2022-02-04 21:04:47,519 BAD EPOCHS (no improvement): 4
2022-02-04 21:04:47,519 ----------------------------------------------------------------------------------------------------
2022-02-04 21:12:13,992 epoch 8 - iter 793/7936 - loss 0.04468006 - samples/sec: 28.42 - lr: 0.000016
2022-02-04 21:19:25,811 epoch 8 - iter 1586/7936 - loss 0.04434977 - samples/sec: 29.39 - lr: 0.000016
2022-02-04 21:26:35,161 epoch 8 - iter 2379/7936 - loss 0.04431108 - samples/sec: 29.56 - lr: 0.000015
2022-02-04 21:33:55,512 epoch 8 - iter 3172/7936 - loss 0.04408371 - samples/sec: 28.82 - lr: 0.000014
2022-02-04 21:41:09,449 epoch 8 - iter 3965/7936 - loss 0.04390607 - samples/sec: 29.24 - lr: 0.000014
2022-02-04 21:48:30,449 epoch 8 - iter 4758/7936 - loss 0.04368218 - samples/sec: 28.77 - lr: 0.000013
2022-02-04 21:55:47,346 epoch 8 - iter 5551/7936 - loss 0.04350544 - samples/sec: 29.05 - lr: 0.000013
2022-02-04 22:03:02,107 epoch 8 - iter 6344/7936 - loss 0.04321482 - samples/sec: 29.19 - lr: 0.000012
2022-02-04 22:10:29,225 epoch 8 - iter 7137/7936 - loss 0.04299359 - samples/sec: 28.38 - lr: 0.000012
2022-02-04 22:17:46,915 epoch 8 - iter 7930/7936 - loss 0.04275655 - samples/sec: 28.99 - lr: 0.000011
2022-02-04 22:17:50,251 ----------------------------------------------------------------------------------------------------
2022-02-04 22:17:50,252 EPOCH 8 done: loss 0.0428 - lr 0.0000111
2022-02-04 22:20:46,443 DEV : loss 0.02112417109310627 - f1-score (micro avg)  0.9396
2022-02-04 22:20:46,502 BAD EPOCHS (no improvement): 4
2022-02-04 22:20:46,502 ----------------------------------------------------------------------------------------------------
2022-02-04 22:27:54,677 epoch 9 - iter 793/7936 - loss 0.03874630 - samples/sec: 29.64 - lr: 0.000011
2022-02-04 22:35:07,034 epoch 9 - iter 1586/7936 - loss 0.03916791 - samples/sec: 29.35 - lr: 0.000010
2022-02-04 22:42:33,861 epoch 9 - iter 2379/7936 - loss 0.03903771 - samples/sec: 28.40 - lr: 0.000009
2022-02-04 22:49:45,768 epoch 9 - iter 3172/7936 - loss 0.03915089 - samples/sec: 29.38 - lr: 0.000009
2022-02-04 22:56:49,271 epoch 9 - iter 3965/7936 - loss 0.03903752 - samples/sec: 29.96 - lr: 0.000008
2022-02-04 23:04:02,033 epoch 9 - iter 4758/7936 - loss 0.03886980 - samples/sec: 29.32 - lr: 0.000008
2022-02-04 23:11:05,006 epoch 9 - iter 5551/7936 - loss 0.03870274 - samples/sec: 30.00 - lr: 0.000007
2022-02-04 23:18:05,622 epoch 9 - iter 6344/7936 - loss 0.03860323 - samples/sec: 30.17 - lr: 0.000007
2022-02-04 23:25:20,470 epoch 9 - iter 7137/7936 - loss 0.03844156 - samples/sec: 29.18 - lr: 0.000006
2022-02-04 23:32:20,810 epoch 9 - iter 7930/7936 - loss 0.03839073 - samples/sec: 30.19 - lr: 0.000006
2022-02-04 23:32:23,941 ----------------------------------------------------------------------------------------------------
2022-02-04 23:32:23,942 EPOCH 9 done: loss 0.0384 - lr 0.0000056
2022-02-04 23:35:14,351 DEV : loss 0.02171432413160801 - f1-score (micro avg)  0.9419
2022-02-04 23:35:14,411 BAD EPOCHS (no improvement): 4
2022-02-04 23:35:14,412 ----------------------------------------------------------------------------------------------------
2022-02-04 23:42:16,230 epoch 10 - iter 793/7936 - loss 0.03646154 - samples/sec: 30.08 - lr: 0.000005
2022-02-04 23:49:27,305 epoch 10 - iter 1586/7936 - loss 0.03635515 - samples/sec: 29.44 - lr: 0.000004
2022-02-04 23:56:27,850 epoch 10 - iter 2379/7936 - loss 0.03662968 - samples/sec: 30.17 - lr: 0.000004
2022-02-05 00:03:30,598 epoch 10 - iter 3172/7936 - loss 0.03640152 - samples/sec: 30.02 - lr: 0.000003
2022-02-05 00:10:46,058 epoch 10 - iter 3965/7936 - loss 0.03636994 - samples/sec: 29.14 - lr: 0.000003
2022-02-05 00:17:50,999 epoch 10 - iter 4758/7936 - loss 0.03636800 - samples/sec: 29.86 - lr: 0.000002
2022-02-05 00:24:51,167 epoch 10 - iter 5551/7936 - loss 0.03625499 - samples/sec: 30.20 - lr: 0.000002
2022-02-05 00:32:07,970 epoch 10 - iter 6344/7936 - loss 0.03625737 - samples/sec: 29.05 - lr: 0.000001
2022-02-05 00:39:14,867 epoch 10 - iter 7137/7936 - loss 0.03618156 - samples/sec: 29.73 - lr: 0.000001
2022-02-05 00:46:17,991 epoch 10 - iter 7930/7936 - loss 0.03611184 - samples/sec: 29.99 - lr: 0.000000
2022-02-05 00:46:21,120 ----------------------------------------------------------------------------------------------------
2022-02-05 00:46:21,123 EPOCH 10 done: loss 0.0361 - lr 0.0000000
2022-02-05 00:49:11,421 DEV : loss 0.023424603044986725 - f1-score (micro avg)  0.9417
2022-02-05 00:49:11,486 BAD EPOCHS (no improvement): 4
2022-02-05 00:49:12,641 ----------------------------------------------------------------------------------------------------
2022-02-05 00:49:12,643 Testing using last state of model ...
2022-02-05 00:52:03,154 0.9303	0.9309	0.9306	0.8856
2022-02-05 00:52:03,155 
Results:
- F-score (micro) 0.9306
- F-score (macro) 0.9057
- Accuracy 0.8856

By class:
              precision    recall  f1-score   support

        pers     0.9373    0.9236    0.9304      2734
         loc     0.9140    0.9371    0.9254      1384
      amount     0.9840    0.9840    0.9840       250
        time     0.9447    0.9407    0.9427       236
        func     0.9209    0.9143    0.9176       140
         org     0.8364    0.9388    0.8846        49
        prod     0.7742    0.8889    0.8276        27
       event     0.8333    0.8333    0.8333        12

   micro avg     0.9303    0.9309    0.9306      4832
   macro avg     0.8931    0.9201    0.9057      4832
weighted avg     0.9307    0.9309    0.9307      4832
 samples avg     0.8856    0.8856    0.8856      4832

2022-02-05 00:52:03,155 ----------------------------------------------------------------------------------------------------