2022-10-26 15:28:10,168 ----------------------------------------------------------------------------------------------------
2022-10-26 15:28:10,173 Model: "SequenceTagger(
  (embeddings): TransformerWordEmbeddings(
    (model): XLMRobertaModel(
      (embeddings): RobertaEmbeddings(
        (word_embeddings): Embedding(250002, 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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
              (intermediate_act_fn): GELUActivation()
            )
            (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)
  (embedding2nn): Linear(in_features=768, out_features=768, bias=True)
  (rnn): LSTM(768, 256, batch_first=True, bidirectional=True)
  (linear): Linear(in_features=512, out_features=15, bias=True)
  (loss_function): ViterbiLoss()
  (crf): CRF()
)"
2022-10-26 15:28:10,176 ----------------------------------------------------------------------------------------------------
2022-10-26 15:28:10,180 Corpus: "Corpus: 8551 train + 1425 dev + 1425 test sentences"
2022-10-26 15:28:10,182 ----------------------------------------------------------------------------------------------------
2022-10-26 15:28:10,184 Parameters:
2022-10-26 15:28:10,186  - learning_rate: "0.010000"
2022-10-26 15:28:10,187  - mini_batch_size: "8"
2022-10-26 15:28:10,188  - patience: "3"
2022-10-26 15:28:10,189  - anneal_factor: "0.5"
2022-10-26 15:28:10,191  - max_epochs: "10"
2022-10-26 15:28:10,192  - shuffle: "True"
2022-10-26 15:28:10,193  - train_with_dev: "False"
2022-10-26 15:28:10,194  - batch_growth_annealing: "False"
2022-10-26 15:28:10,196 ----------------------------------------------------------------------------------------------------
2022-10-26 15:28:10,197 Model training base path: "/content/model/xlmr_ner"
2022-10-26 15:28:10,198 ----------------------------------------------------------------------------------------------------
2022-10-26 15:28:10,199 Device: cuda:0
2022-10-26 15:28:10,201 ----------------------------------------------------------------------------------------------------
2022-10-26 15:28:10,202 Embeddings storage mode: none
2022-10-26 15:28:10,203 ----------------------------------------------------------------------------------------------------
2022-10-26 15:30:29,962 epoch 1 - iter 106/1069 - loss 0.55101171 - samples/sec: 6.07 - lr: 0.010000
2022-10-26 15:32:28,714 epoch 1 - iter 212/1069 - loss 0.35636418 - samples/sec: 7.14 - lr: 0.010000
2022-10-26 15:34:23,625 epoch 1 - iter 318/1069 - loss 0.28047260 - samples/sec: 7.38 - lr: 0.010000
2022-10-26 15:36:24,015 epoch 1 - iter 424/1069 - loss 0.23890211 - samples/sec: 7.04 - lr: 0.010000
2022-10-26 15:38:21,987 epoch 1 - iter 530/1069 - loss 0.21322222 - samples/sec: 7.19 - lr: 0.010000
2022-10-26 15:40:22,521 epoch 1 - iter 636/1069 - loss 0.19431796 - samples/sec: 7.04 - lr: 0.010000
2022-10-26 15:42:18,754 epoch 1 - iter 742/1069 - loss 0.18084010 - samples/sec: 7.30 - lr: 0.010000
2022-10-26 15:44:18,344 epoch 1 - iter 848/1069 - loss 0.16975329 - samples/sec: 7.09 - lr: 0.010000
2022-10-26 15:46:14,738 epoch 1 - iter 954/1069 - loss 0.16158584 - samples/sec: 7.29 - lr: 0.010000
2022-10-26 15:48:14,067 epoch 1 - iter 1060/1069 - loss 0.15491697 - samples/sec: 7.11 - lr: 0.010000
2022-10-26 15:48:24,569 ----------------------------------------------------------------------------------------------------
2022-10-26 15:48:24,577 EPOCH 1 done: loss 0.1543 - lr 0.010000
2022-10-26 15:50:17,480 Evaluating as a multi-label problem: False
2022-10-26 15:50:17,512 DEV : loss 0.060714565217494965 - f1-score (micro avg)  0.7908
2022-10-26 15:50:17,553 BAD EPOCHS (no improvement): 0
2022-10-26 15:50:17,554 saving best model
2022-10-26 15:50:23,470 ----------------------------------------------------------------------------------------------------
2022-10-26 15:52:24,219 epoch 2 - iter 106/1069 - loss 0.08869057 - samples/sec: 7.02 - lr: 0.010000
2022-10-26 15:54:21,594 epoch 2 - iter 212/1069 - loss 0.08600343 - samples/sec: 7.23 - lr: 0.010000
2022-10-26 15:56:19,809 epoch 2 - iter 318/1069 - loss 0.08546665 - samples/sec: 7.17 - lr: 0.010000
2022-10-26 15:58:17,214 epoch 2 - iter 424/1069 - loss 0.08476718 - samples/sec: 7.22 - lr: 0.010000
2022-10-26 16:00:16,114 epoch 2 - iter 530/1069 - loss 0.08542624 - samples/sec: 7.13 - lr: 0.010000
2022-10-26 16:02:13,540 epoch 2 - iter 636/1069 - loss 0.08522910 - samples/sec: 7.22 - lr: 0.010000
2022-10-26 16:04:12,854 epoch 2 - iter 742/1069 - loss 0.08502467 - samples/sec: 7.11 - lr: 0.010000
2022-10-26 16:06:13,219 epoch 2 - iter 848/1069 - loss 0.08373459 - samples/sec: 7.05 - lr: 0.010000
2022-10-26 16:08:09,808 epoch 2 - iter 954/1069 - loss 0.08316639 - samples/sec: 7.27 - lr: 0.010000
2022-10-26 16:10:11,036 epoch 2 - iter 1060/1069 - loss 0.08215396 - samples/sec: 7.00 - lr: 0.010000
2022-10-26 16:10:21,246 ----------------------------------------------------------------------------------------------------
2022-10-26 16:10:21,249 EPOCH 2 done: loss 0.0821 - lr 0.010000
2022-10-26 16:12:13,875 Evaluating as a multi-label problem: False
2022-10-26 16:12:13,905 DEV : loss 0.05180404335260391 - f1-score (micro avg)  0.8408
2022-10-26 16:12:13,947 BAD EPOCHS (no improvement): 0
2022-10-26 16:12:13,948 saving best model
2022-10-26 16:12:19,344 ----------------------------------------------------------------------------------------------------
2022-10-26 16:14:19,879 epoch 3 - iter 106/1069 - loss 0.06627178 - samples/sec: 7.04 - lr: 0.010000
2022-10-26 16:16:18,272 epoch 3 - iter 212/1069 - loss 0.07094348 - samples/sec: 7.16 - lr: 0.010000
2022-10-26 16:18:18,453 epoch 3 - iter 318/1069 - loss 0.07194093 - samples/sec: 7.06 - lr: 0.010000
2022-10-26 16:20:15,802 epoch 3 - iter 424/1069 - loss 0.07242840 - samples/sec: 7.23 - lr: 0.010000
2022-10-26 16:22:12,248 epoch 3 - iter 530/1069 - loss 0.07171872 - samples/sec: 7.28 - lr: 0.010000
2022-10-26 16:24:12,231 epoch 3 - iter 636/1069 - loss 0.07162092 - samples/sec: 7.07 - lr: 0.010000
2022-10-26 16:26:10,382 epoch 3 - iter 742/1069 - loss 0.07130310 - samples/sec: 7.18 - lr: 0.010000
2022-10-26 16:28:08,953 epoch 3 - iter 848/1069 - loss 0.07050136 - samples/sec: 7.15 - lr: 0.010000
2022-10-26 16:30:09,728 epoch 3 - iter 954/1069 - loss 0.07070517 - samples/sec: 7.02 - lr: 0.010000
2022-10-26 16:32:08,721 epoch 3 - iter 1060/1069 - loss 0.07033198 - samples/sec: 7.13 - lr: 0.010000
2022-10-26 16:32:18,654 ----------------------------------------------------------------------------------------------------
2022-10-26 16:32:18,656 EPOCH 3 done: loss 0.0702 - lr 0.010000
2022-10-26 16:34:10,956 Evaluating as a multi-label problem: False
2022-10-26 16:34:10,986 DEV : loss 0.04575943946838379 - f1-score (micro avg)  0.8693
2022-10-26 16:34:11,026 BAD EPOCHS (no improvement): 0
2022-10-26 16:34:11,029 saving best model
2022-10-26 16:34:16,564 ----------------------------------------------------------------------------------------------------
2022-10-26 16:36:12,350 epoch 4 - iter 106/1069 - loss 0.06432601 - samples/sec: 7.32 - lr: 0.010000
2022-10-26 16:38:08,474 epoch 4 - iter 212/1069 - loss 0.06376094 - samples/sec: 7.30 - lr: 0.010000
2022-10-26 16:40:03,219 epoch 4 - iter 318/1069 - loss 0.06273795 - samples/sec: 7.39 - lr: 0.010000
2022-10-26 16:41:59,110 epoch 4 - iter 424/1069 - loss 0.06153989 - samples/sec: 7.32 - lr: 0.010000
2022-10-26 16:43:57,347 epoch 4 - iter 530/1069 - loss 0.06137878 - samples/sec: 7.17 - lr: 0.010000
2022-10-26 16:45:55,146 epoch 4 - iter 636/1069 - loss 0.06072772 - samples/sec: 7.20 - lr: 0.010000
2022-10-26 16:47:53,049 epoch 4 - iter 742/1069 - loss 0.06031769 - samples/sec: 7.19 - lr: 0.010000
2022-10-26 16:49:50,705 epoch 4 - iter 848/1069 - loss 0.06084099 - samples/sec: 7.21 - lr: 0.010000
2022-10-26 16:51:49,833 epoch 4 - iter 954/1069 - loss 0.06096388 - samples/sec: 7.12 - lr: 0.010000
2022-10-26 16:53:45,640 epoch 4 - iter 1060/1069 - loss 0.06061743 - samples/sec: 7.32 - lr: 0.010000
2022-10-26 16:53:54,974 ----------------------------------------------------------------------------------------------------
2022-10-26 16:53:54,976 EPOCH 4 done: loss 0.0606 - lr 0.010000
2022-10-26 16:55:45,518 Evaluating as a multi-label problem: False
2022-10-26 16:55:45,548 DEV : loss 0.04747875779867172 - f1-score (micro avg)  0.8627
2022-10-26 16:55:45,589 BAD EPOCHS (no improvement): 1
2022-10-26 16:55:45,590 ----------------------------------------------------------------------------------------------------
2022-10-26 16:57:41,259 epoch 5 - iter 106/1069 - loss 0.05285565 - samples/sec: 7.33 - lr: 0.010000
2022-10-26 16:59:40,296 epoch 5 - iter 212/1069 - loss 0.05049977 - samples/sec: 7.12 - lr: 0.010000
2022-10-26 17:01:35,184 epoch 5 - iter 318/1069 - loss 0.05297933 - samples/sec: 7.38 - lr: 0.010000
2022-10-26 17:03:34,028 epoch 5 - iter 424/1069 - loss 0.05293744 - samples/sec: 7.14 - lr: 0.010000
2022-10-26 17:05:29,295 epoch 5 - iter 530/1069 - loss 0.05359386 - samples/sec: 7.36 - lr: 0.010000
2022-10-26 17:07:25,593 epoch 5 - iter 636/1069 - loss 0.05307424 - samples/sec: 7.29 - lr: 0.010000
2022-10-26 17:09:22,893 epoch 5 - iter 742/1069 - loss 0.05323355 - samples/sec: 7.23 - lr: 0.010000
2022-10-26 17:11:22,602 epoch 5 - iter 848/1069 - loss 0.05272547 - samples/sec: 7.08 - lr: 0.010000
2022-10-26 17:13:22,960 epoch 5 - iter 954/1069 - loss 0.05280553 - samples/sec: 7.05 - lr: 0.010000
2022-10-26 17:15:20,527 epoch 5 - iter 1060/1069 - loss 0.05265360 - samples/sec: 7.21 - lr: 0.010000
2022-10-26 17:15:29,931 ----------------------------------------------------------------------------------------------------
2022-10-26 17:15:29,932 EPOCH 5 done: loss 0.0526 - lr 0.010000
2022-10-26 17:17:21,728 Evaluating as a multi-label problem: False
2022-10-26 17:17:21,760 DEV : loss 0.03879784420132637 - f1-score (micro avg)  0.8864
2022-10-26 17:17:21,803 BAD EPOCHS (no improvement): 0
2022-10-26 17:17:21,804 saving best model
2022-10-26 17:17:27,330 ----------------------------------------------------------------------------------------------------
2022-10-26 17:19:26,401 epoch 6 - iter 106/1069 - loss 0.04801558 - samples/sec: 7.12 - lr: 0.010000
2022-10-26 17:21:22,988 epoch 6 - iter 212/1069 - loss 0.05008290 - samples/sec: 7.27 - lr: 0.010000
2022-10-26 17:23:16,794 epoch 6 - iter 318/1069 - loss 0.04925649 - samples/sec: 7.45 - lr: 0.010000
2022-10-26 17:25:15,532 epoch 6 - iter 424/1069 - loss 0.04786643 - samples/sec: 7.14 - lr: 0.010000
2022-10-26 17:27:13,913 epoch 6 - iter 530/1069 - loss 0.04879792 - samples/sec: 7.16 - lr: 0.010000
2022-10-26 17:29:10,114 epoch 6 - iter 636/1069 - loss 0.04800786 - samples/sec: 7.30 - lr: 0.010000
2022-10-26 17:31:07,810 epoch 6 - iter 742/1069 - loss 0.04755361 - samples/sec: 7.21 - lr: 0.010000
2022-10-26 17:33:04,496 epoch 6 - iter 848/1069 - loss 0.04782375 - samples/sec: 7.27 - lr: 0.010000
2022-10-26 17:35:05,834 epoch 6 - iter 954/1069 - loss 0.04776160 - samples/sec: 6.99 - lr: 0.010000
2022-10-26 17:37:03,878 epoch 6 - iter 1060/1069 - loss 0.04743945 - samples/sec: 7.18 - lr: 0.010000
2022-10-26 17:37:14,466 ----------------------------------------------------------------------------------------------------
2022-10-26 17:37:14,468 EPOCH 6 done: loss 0.0475 - lr 0.010000
2022-10-26 17:39:07,562 Evaluating as a multi-label problem: False
2022-10-26 17:39:07,592 DEV : loss 0.03874654322862625 - f1-score (micro avg)  0.8908
2022-10-26 17:39:07,633 BAD EPOCHS (no improvement): 0
2022-10-26 17:39:07,635 saving best model
2022-10-26 17:39:13,242 ----------------------------------------------------------------------------------------------------
2022-10-26 17:41:11,924 epoch 7 - iter 106/1069 - loss 0.04334369 - samples/sec: 7.15 - lr: 0.010000
2022-10-26 17:43:11,382 epoch 7 - iter 212/1069 - loss 0.04192565 - samples/sec: 7.10 - lr: 0.010000
2022-10-26 17:45:08,087 epoch 7 - iter 318/1069 - loss 0.04115627 - samples/sec: 7.27 - lr: 0.010000
2022-10-26 17:47:06,615 epoch 7 - iter 424/1069 - loss 0.04114928 - samples/sec: 7.16 - lr: 0.010000
2022-10-26 17:49:03,863 epoch 7 - iter 530/1069 - loss 0.04105023 - samples/sec: 7.23 - lr: 0.010000
2022-10-26 17:51:02,216 epoch 7 - iter 636/1069 - loss 0.04125208 - samples/sec: 7.17 - lr: 0.010000
2022-10-26 17:53:04,293 epoch 7 - iter 742/1069 - loss 0.04151765 - samples/sec: 6.95 - lr: 0.010000
2022-10-26 17:55:01,446 epoch 7 - iter 848/1069 - loss 0.04170200 - samples/sec: 7.24 - lr: 0.010000
2022-10-26 17:56:59,848 epoch 7 - iter 954/1069 - loss 0.04180177 - samples/sec: 7.16 - lr: 0.010000
2022-10-26 17:58:56,175 epoch 7 - iter 1060/1069 - loss 0.04203413 - samples/sec: 7.29 - lr: 0.010000
2022-10-26 17:59:05,814 ----------------------------------------------------------------------------------------------------
2022-10-26 17:59:05,816 EPOCH 7 done: loss 0.0420 - lr 0.010000
2022-10-26 18:00:59,457 Evaluating as a multi-label problem: False
2022-10-26 18:00:59,486 DEV : loss 0.04413652420043945 - f1-score (micro avg)  0.8968
2022-10-26 18:00:59,527 BAD EPOCHS (no improvement): 0
2022-10-26 18:00:59,529 saving best model
2022-10-26 18:01:05,372 ----------------------------------------------------------------------------------------------------
2022-10-26 18:03:03,422 epoch 8 - iter 106/1069 - loss 0.03592615 - samples/sec: 7.18 - lr: 0.010000
2022-10-26 18:05:00,466 epoch 8 - iter 212/1069 - loss 0.03676863 - samples/sec: 7.25 - lr: 0.010000
2022-10-26 18:06:58,178 epoch 8 - iter 318/1069 - loss 0.03702258 - samples/sec: 7.20 - lr: 0.010000
2022-10-26 18:08:55,170 epoch 8 - iter 424/1069 - loss 0.03704658 - samples/sec: 7.25 - lr: 0.010000
2022-10-26 18:10:52,222 epoch 8 - iter 530/1069 - loss 0.03711348 - samples/sec: 7.25 - lr: 0.010000
2022-10-26 18:12:51,244 epoch 8 - iter 636/1069 - loss 0.03715815 - samples/sec: 7.13 - lr: 0.010000
2022-10-26 18:14:50,229 epoch 8 - iter 742/1069 - loss 0.03708747 - samples/sec: 7.13 - lr: 0.010000
2022-10-26 18:16:47,946 epoch 8 - iter 848/1069 - loss 0.03734575 - samples/sec: 7.20 - lr: 0.010000
2022-10-26 18:18:45,873 epoch 8 - iter 954/1069 - loss 0.03736843 - samples/sec: 7.19 - lr: 0.010000
2022-10-26 18:20:43,504 epoch 8 - iter 1060/1069 - loss 0.03737578 - samples/sec: 7.21 - lr: 0.010000
2022-10-26 18:20:53,262 ----------------------------------------------------------------------------------------------------
2022-10-26 18:20:53,265 EPOCH 8 done: loss 0.0374 - lr 0.010000
2022-10-26 18:22:46,256 Evaluating as a multi-label problem: False
2022-10-26 18:22:46,293 DEV : loss 0.03726610541343689 - f1-score (micro avg)  0.9117
2022-10-26 18:22:46,336 BAD EPOCHS (no improvement): 0
2022-10-26 18:22:46,337 saving best model
2022-10-26 18:22:51,847 ----------------------------------------------------------------------------------------------------
2022-10-26 18:24:50,402 epoch 9 - iter 106/1069 - loss 0.03606101 - samples/sec: 7.15 - lr: 0.010000
2022-10-26 18:26:47,577 epoch 9 - iter 212/1069 - loss 0.03466163 - samples/sec: 7.24 - lr: 0.010000
2022-10-26 18:28:47,029 epoch 9 - iter 318/1069 - loss 0.03420843 - samples/sec: 7.10 - lr: 0.010000
2022-10-26 18:30:43,235 epoch 9 - iter 424/1069 - loss 0.03406325 - samples/sec: 7.30 - lr: 0.010000
2022-10-26 18:32:41,132 epoch 9 - iter 530/1069 - loss 0.03393077 - samples/sec: 7.19 - lr: 0.010000
2022-10-26 18:34:35,953 epoch 9 - iter 636/1069 - loss 0.03438052 - samples/sec: 7.39 - lr: 0.010000
2022-10-26 18:36:33,872 epoch 9 - iter 742/1069 - loss 0.03435922 - samples/sec: 7.19 - lr: 0.010000
2022-10-26 18:38:30,457 epoch 9 - iter 848/1069 - loss 0.03351594 - samples/sec: 7.27 - lr: 0.010000
2022-10-26 18:40:26,775 epoch 9 - iter 954/1069 - loss 0.03363514 - samples/sec: 7.29 - lr: 0.010000
2022-10-26 18:42:26,040 epoch 9 - iter 1060/1069 - loss 0.03301736 - samples/sec: 7.11 - lr: 0.010000
2022-10-26 18:42:34,477 ----------------------------------------------------------------------------------------------------
2022-10-26 18:42:34,480 EPOCH 9 done: loss 0.0330 - lr 0.010000
2022-10-26 18:44:24,572 Evaluating as a multi-label problem: False
2022-10-26 18:44:24,602 DEV : loss 0.04557322338223457 - f1-score (micro avg)  0.9084
2022-10-26 18:44:24,644 BAD EPOCHS (no improvement): 1
2022-10-26 18:44:24,646 ----------------------------------------------------------------------------------------------------
2022-10-26 18:46:21,774 epoch 10 - iter 106/1069 - loss 0.02992093 - samples/sec: 7.24 - lr: 0.010000
2022-10-26 18:48:20,730 epoch 10 - iter 212/1069 - loss 0.02886380 - samples/sec: 7.13 - lr: 0.010000
2022-10-26 18:50:20,679 epoch 10 - iter 318/1069 - loss 0.03109654 - samples/sec: 7.07 - lr: 0.010000
2022-10-26 18:52:14,564 epoch 10 - iter 424/1069 - loss 0.03091892 - samples/sec: 7.45 - lr: 0.010000
2022-10-26 18:54:14,888 epoch 10 - iter 530/1069 - loss 0.02977117 - samples/sec: 7.05 - lr: 0.010000
2022-10-26 18:56:13,992 epoch 10 - iter 636/1069 - loss 0.02969566 - samples/sec: 7.12 - lr: 0.010000
2022-10-26 18:58:12,618 epoch 10 - iter 742/1069 - loss 0.02979601 - samples/sec: 7.15 - lr: 0.010000
2022-10-26 19:00:10,398 epoch 10 - iter 848/1069 - loss 0.03040781 - samples/sec: 7.20 - lr: 0.010000
2022-10-26 19:02:06,063 epoch 10 - iter 954/1069 - loss 0.03029135 - samples/sec: 7.33 - lr: 0.010000
2022-10-26 19:04:05,626 epoch 10 - iter 1060/1069 - loss 0.03035206 - samples/sec: 7.09 - lr: 0.010000
2022-10-26 19:04:15,538 ----------------------------------------------------------------------------------------------------
2022-10-26 19:04:15,540 EPOCH 10 done: loss 0.0303 - lr 0.010000
2022-10-26 19:06:06,586 Evaluating as a multi-label problem: False
2022-10-26 19:06:06,621 DEV : loss 0.03892701491713524 - f1-score (micro avg)  0.9132
2022-10-26 19:06:06,663 BAD EPOCHS (no improvement): 0
2022-10-26 19:06:06,665 saving best model
2022-10-26 19:06:17,597 ----------------------------------------------------------------------------------------------------
2022-10-26 19:06:17,723 loading file /content/model/xlmr_ner/best-model.pt
2022-10-26 19:06:24,597 SequenceTagger predicts: Dictionary with 15 tags: O, S-PER, B-PER, E-PER, I-PER, S-MISC, B-MISC, E-MISC, I-MISC, S-LOC, B-LOC, E-LOC, I-LOC, <START>, <STOP>
2022-10-26 19:08:17,003 Evaluating as a multi-label problem: False
2022-10-26 19:08:17,040 0.9053	0.9316	0.9182	0.8955
2022-10-26 19:08:17,041 
Results:
- F-score (micro) 0.9182
- F-score (macro) 0.8875
- Accuracy 0.8955

By class:
              precision    recall  f1-score   support

         PER     0.9339    0.9633    0.9484      2127
        MISC     0.8469    0.9250    0.8842       933
         LOC     0.8955    0.7732    0.8299       388

   micro avg     0.9053    0.9316    0.9182      3448
   macro avg     0.8921    0.8872    0.8875      3448
weighted avg     0.9060    0.9316    0.9177      3448

2022-10-26 19:08:17,045 ----------------------------------------------------------------------------------------------------