Training in progress, epoch 1

README.md

@@ -0,0 +1,103 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: PlanTL-GOB-ES/bsc-bio-ehr-es
+tags:
+- token-classification
+- generated_from_trainer
+datasets:
+- Rodrigo1771/distemist-fasttext-75-ner
+metrics:
+- precision
+- recall
+- f1
+- accuracy
+model-index:
+- name: output
+  results:
+  - task:
+      name: Token Classification
+      type: token-classification
+    dataset:
+      name: Rodrigo1771/distemist-fasttext-75-ner
+      type: Rodrigo1771/distemist-fasttext-75-ner
+      config: DisTEMIST NER
+      split: validation
+      args: DisTEMIST NER
+    metrics:
+    - name: Precision
+      type: precision
+      value: 0.7991246256622898
+    - name: Recall
+      type: recall
+      value: 0.8116518483855872
+    - name: F1
+      type: f1
+      value: 0.8053395240858967
+    - name: Accuracy
+      type: accuracy
+      value: 0.9758743323218038
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# output
+
+This model is a fine-tuned version of [PlanTL-GOB-ES/bsc-bio-ehr-es](https://huggingface.co/PlanTL-GOB-ES/bsc-bio-ehr-es) on the Rodrigo1771/distemist-fasttext-75-ner dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1711
+- Precision: 0.7991
+- Recall: 0.8117
+- F1: 0.8053
+- Accuracy: 0.9759
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 32
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 10.0
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.1165        | 0.9993 | 702  | 0.0809          | 0.7455    | 0.8039 | 0.7736 | 0.9735   |
+| 0.0461        | 2.0    | 1405 | 0.0956          | 0.7611    | 0.8067 | 0.7833 | 0.9747   |
+| 0.0165        | 2.9993 | 2107 | 0.1057          | 0.7721    | 0.7990 | 0.7853 | 0.9744   |
+| 0.011         | 4.0    | 2810 | 0.1274          | 0.7759    | 0.8196 | 0.7971 | 0.9751   |
+| 0.006         | 4.9993 | 3512 | 0.1358          | 0.7904    | 0.8049 | 0.7976 | 0.9745   |
+| 0.0045        | 6.0    | 4215 | 0.1420          | 0.7911    | 0.7985 | 0.7948 | 0.9746   |
+| 0.0037        | 6.9993 | 4917 | 0.1601          | 0.7925    | 0.8000 | 0.7962 | 0.9749   |
+| 0.0022        | 8.0    | 5620 | 0.1621          | 0.8000    | 0.8102 | 0.8051 | 0.9758   |
+| 0.0016        | 8.9993 | 6322 | 0.1681          | 0.7972    | 0.8086 | 0.8029 | 0.9758   |
+| 0.0013        | 9.9929 | 7020 | 0.1711          | 0.7991    | 0.8117 | 0.8053 | 0.9759   |
+
+
+### Framework versions
+
+- Transformers 4.44.2
+- Pytorch 2.4.0+cu121
+- Datasets 2.21.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,26 @@
+{
+    "epoch": 9.99288256227758,
+    "eval_accuracy": 0.9758743323218038,
+    "eval_f1": 0.8053395240858967,
+    "eval_loss": 0.17113561928272247,
+    "eval_precision": 0.7991246256622898,
+    "eval_recall": 0.8116518483855872,
+    "eval_runtime": 14.1813,
+    "eval_samples": 6810,
+    "eval_samples_per_second": 480.21,
+    "eval_steps_per_second": 60.079,
+    "predict_accuracy": 0.9755136059661639,
+    "predict_f1": 0.8018404526518684,
+    "predict_loss": 0.161112979054451,
+    "predict_precision": 0.7862455798073406,
+    "predict_recall": 0.818066480588683,
+    "predict_runtime": 29.1457,
+    "predict_samples_per_second": 501.411,
+    "predict_steps_per_second": 62.685,
+    "total_flos": 2.130317416831723e+16,
+    "train_loss": 0.017706579814779112,
+    "train_runtime": 1901.5665,
+    "train_samples": 44938,
+    "train_samples_per_second": 236.321,
+    "train_steps_per_second": 3.692
+}

config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "PlanTL-GOB-ES/bsc-bio-ehr-es",
+  "architectures": [
+    "RobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "ner",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "O",
+    "1": "B-FARMACO",
+    "2": "I-FARMACO"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "B-FARMACO": 1,
+    "I-FARMACO": 2,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50262
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 9.99288256227758,
+    "eval_accuracy": 0.9758743323218038,
+    "eval_f1": 0.8053395240858967,
+    "eval_loss": 0.17113561928272247,
+    "eval_precision": 0.7991246256622898,
+    "eval_recall": 0.8116518483855872,
+    "eval_runtime": 14.1813,
+    "eval_samples": 6810,
+    "eval_samples_per_second": 480.21,
+    "eval_steps_per_second": 60.079
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3aebeaa296e85a82ea161a93feaf408a33f23f38fe04044a7f0fda8f18f0a2fe
+size 496244100

predict_results.json

@@ -0,0 +1,10 @@
+{
+    "predict_accuracy": 0.9755136059661639,
+    "predict_f1": 0.8018404526518684,
+    "predict_loss": 0.161112979054451,
+    "predict_precision": 0.7862455798073406,
+    "predict_recall": 0.818066480588683,
+    "predict_runtime": 29.1457,
+    "predict_samples_per_second": 501.411,
+    "predict_steps_per_second": 62.685
+}

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tb/events.out.tfevents.1725911599.0ada7e7d1d89.3073.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5816f5dda3b32e02f4eaf786675db7f9a58883db7fed71ab3a319cb3cdef0814
+size 13213

tb/events.out.tfevents.1725913533.0ada7e7d1d89.3073.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bac376227477ed56be640720bee3901400d0b77a1b47bd01abe47f5a9812732
+size 560

tb/events.out.tfevents.1725913932.0ada7e7d1d89.13010.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:674da0d93ba53762b090167b15a68a72f2dfa2143b514065154e098e35eb57c1
+size 5645

tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50261": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
+  "max_len": 512,
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

train.log

@@ -0,0 +1,412 @@
+2024-09-09 20:31:42.982639: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
+2024-09-09 20:31:43.001035: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:485] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
+2024-09-09 20:31:43.022596: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:8454] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
+2024-09-09 20:31:43.029063: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1452] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
+2024-09-09 20:31:43.044988: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
+To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
+2024-09-09 20:31:44.304336: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
+/usr/local/lib/python3.10/dist-packages/transformers/training_args.py:1525: FutureWarning: `evaluation_strategy` is deprecated and will be removed in version 4.46 of 🤗 Transformers. Use `eval_strategy` instead
+  warnings.warn(
+09/09/2024 20:31:45 - WARNING - __main__ -   Process rank: 0, device: cuda:0, n_gpu: 1distributed training: True, 16-bits training: False
+09/09/2024 20:31:45 - INFO - __main__ -   Training/evaluation parameters TrainingArguments(
+_n_gpu=1,
+accelerator_config={'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False},
+adafactor=False,
+adam_beta1=0.9,
+adam_beta2=0.999,
+adam_epsilon=1e-08,
+auto_find_batch_size=False,
+batch_eval_metrics=False,
+bf16=False,
+bf16_full_eval=False,
+data_seed=None,
+dataloader_drop_last=False,
+dataloader_num_workers=0,
+dataloader_persistent_workers=False,
+dataloader_pin_memory=True,
+dataloader_prefetch_factor=None,
+ddp_backend=None,
+ddp_broadcast_buffers=None,
+ddp_bucket_cap_mb=None,
+ddp_find_unused_parameters=None,
+ddp_timeout=1800,
+debug=[],
+deepspeed=None,
+disable_tqdm=False,
+dispatch_batches=None,
+do_eval=True,
+do_predict=True,
+do_train=True,
+eval_accumulation_steps=None,
+eval_delay=0,
+eval_do_concat_batches=True,
+eval_on_start=False,
+eval_steps=None,
+eval_strategy=epoch,
+eval_use_gather_object=False,
+evaluation_strategy=epoch,
+fp16=False,
+fp16_backend=auto,
+fp16_full_eval=False,
+fp16_opt_level=O1,
+fsdp=[],
+fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False},
+fsdp_min_num_params=0,
+fsdp_transformer_layer_cls_to_wrap=None,
+full_determinism=False,
+gradient_accumulation_steps=2,
+gradient_checkpointing=False,
+gradient_checkpointing_kwargs=None,
+greater_is_better=True,
+group_by_length=False,
+half_precision_backend=auto,
+hub_always_push=False,
+hub_model_id=None,
+hub_private_repo=False,
+hub_strategy=every_save,
+hub_token=<HUB_TOKEN>,
+ignore_data_skip=False,
+include_inputs_for_metrics=False,
+include_num_input_tokens_seen=False,
+include_tokens_per_second=False,
+jit_mode_eval=False,
+label_names=None,
+label_smoothing_factor=0.0,
+learning_rate=5e-05,
+length_column_name=length,
+load_best_model_at_end=True,
+local_rank=0,
+log_level=passive,
+log_level_replica=warning,
+log_on_each_node=True,
+logging_dir=/content/dissertation/scripts/ner/output/tb,
+logging_first_step=False,
+logging_nan_inf_filter=True,
+logging_steps=500,
+logging_strategy=steps,
+lr_scheduler_kwargs={},
+lr_scheduler_type=linear,
+max_grad_norm=1.0,
+max_steps=-1,
+metric_for_best_model=f1,
+mp_parameters=,
+neftune_noise_alpha=None,
+no_cuda=False,
+num_train_epochs=10.0,
+optim=adamw_torch,
+optim_args=None,
+optim_target_modules=None,
+output_dir=/content/dissertation/scripts/ner/output,
+overwrite_output_dir=True,
+past_index=-1,
+per_device_eval_batch_size=8,
+per_device_train_batch_size=32,
+prediction_loss_only=False,
+push_to_hub=True,
+push_to_hub_model_id=None,
+push_to_hub_organization=None,
+push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
+ray_scope=last,
+remove_unused_columns=True,
+report_to=['tensorboard'],
+restore_callback_states_from_checkpoint=False,
+resume_from_checkpoint=None,
+run_name=/content/dissertation/scripts/ner/output,
+save_on_each_node=False,
+save_only_model=False,
+save_safetensors=True,
+save_steps=500,
+save_strategy=epoch,
+save_total_limit=None,
+seed=42,
+skip_memory_metrics=True,
+split_batches=None,
+tf32=None,
+torch_compile=False,
+torch_compile_backend=None,
+torch_compile_mode=None,
+torch_empty_cache_steps=None,
+torchdynamo=None,
+tpu_metrics_debug=False,
+tpu_num_cores=None,
+use_cpu=False,
+use_ipex=False,
+use_legacy_prediction_loop=False,
+use_mps_device=False,
+warmup_ratio=0.0,
+warmup_steps=0,
+weight_decay=0.0,
+)
+
+
+
+
+
+
+
+[INFO|configuration_utils.py:733] 2024-09-09 20:32:03,967 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--PlanTL-GOB-ES--bsc-bio-ehr-es/snapshots/1e543adb2d21f19d85a89305eebdbd64ab656b99/config.json
+[INFO|configuration_utils.py:800] 2024-09-09 20:32:03,971 >> Model config RobertaConfig {
+  "_name_or_path": "PlanTL-GOB-ES/bsc-bio-ehr-es",
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "ner",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "O",
+    "1": "B-FARMACO",
+    "2": "I-FARMACO"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "B-FARMACO": 1,
+    "I-FARMACO": 2,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.44.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50262
+}
+
+[INFO|configuration_utils.py:733] 2024-09-09 20:32:04,220 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--PlanTL-GOB-ES--bsc-bio-ehr-es/snapshots/1e543adb2d21f19d85a89305eebdbd64ab656b99/config.json
+[INFO|configuration_utils.py:800] 2024-09-09 20:32:04,221 >> Model config RobertaConfig {
+  "_name_or_path": "PlanTL-GOB-ES/bsc-bio-ehr-es",
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.44.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50262
+}
+
+[INFO|tokenization_utils_base.py:2269] 2024-09-09 20:32:04,231 >> loading file vocab.json from cache at /root/.cache/huggingface/hub/models--PlanTL-GOB-ES--bsc-bio-ehr-es/snapshots/1e543adb2d21f19d85a89305eebdbd64ab656b99/vocab.json
+[INFO|tokenization_utils_base.py:2269] 2024-09-09 20:32:04,231 >> loading file merges.txt from cache at /root/.cache/huggingface/hub/models--PlanTL-GOB-ES--bsc-bio-ehr-es/snapshots/1e543adb2d21f19d85a89305eebdbd64ab656b99/merges.txt
+[INFO|tokenization_utils_base.py:2269] 2024-09-09 20:32:04,231 >> loading file tokenizer.json from cache at None
+[INFO|tokenization_utils_base.py:2269] 2024-09-09 20:32:04,231 >> loading file added_tokens.json from cache at None
+[INFO|tokenization_utils_base.py:2269] 2024-09-09 20:32:04,231 >> loading file special_tokens_map.json from cache at /root/.cache/huggingface/hub/models--PlanTL-GOB-ES--bsc-bio-ehr-es/snapshots/1e543adb2d21f19d85a89305eebdbd64ab656b99/special_tokens_map.json
+[INFO|tokenization_utils_base.py:2269] 2024-09-09 20:32:04,231 >> loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--PlanTL-GOB-ES--bsc-bio-ehr-es/snapshots/1e543adb2d21f19d85a89305eebdbd64ab656b99/tokenizer_config.json
+[INFO|configuration_utils.py:733] 2024-09-09 20:32:04,231 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--PlanTL-GOB-ES--bsc-bio-ehr-es/snapshots/1e543adb2d21f19d85a89305eebdbd64ab656b99/config.json
+[INFO|configuration_utils.py:800] 2024-09-09 20:32:04,232 >> Model config RobertaConfig {
+  "_name_or_path": "PlanTL-GOB-ES/bsc-bio-ehr-es",
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.44.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50262
+}
+
+/usr/local/lib/python3.10/dist-packages/transformers/tokenization_utils_base.py:1601: FutureWarning: `clean_up_tokenization_spaces` was not set. It will be set to `True` by default. This behavior will be depracted in transformers v4.45, and will be then set to `False` by default. For more details check this issue: https://github.com/huggingface/transformers/issues/31884
+  warnings.warn(
+[INFO|configuration_utils.py:733] 2024-09-09 20:32:04,315 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--PlanTL-GOB-ES--bsc-bio-ehr-es/snapshots/1e543adb2d21f19d85a89305eebdbd64ab656b99/config.json
+[INFO|configuration_utils.py:800] 2024-09-09 20:32:04,317 >> Model config RobertaConfig {
+  "_name_or_path": "PlanTL-GOB-ES/bsc-bio-ehr-es",
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.44.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50262
+}
+
+[INFO|modeling_utils.py:3678] 2024-09-09 20:32:04,638 >> loading weights file pytorch_model.bin from cache at /root/.cache/huggingface/hub/models--PlanTL-GOB-ES--bsc-bio-ehr-es/snapshots/1e543adb2d21f19d85a89305eebdbd64ab656b99/pytorch_model.bin
+[INFO|modeling_utils.py:4497] 2024-09-09 20:32:04,717 >> Some weights of the model checkpoint at PlanTL-GOB-ES/bsc-bio-ehr-es were not used when initializing RobertaForTokenClassification: ['lm_head.bias', 'lm_head.decoder.bias', 'lm_head.decoder.weight', 'lm_head.dense.bias', 'lm_head.dense.weight', 'lm_head.layer_norm.bias', 'lm_head.layer_norm.weight']
+- This IS expected if you are initializing RobertaForTokenClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
+- This IS NOT expected if you are initializing RobertaForTokenClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
+[WARNING|modeling_utils.py:4509] 2024-09-09 20:32:04,717 >> Some weights of RobertaForTokenClassification were not initialized from the model checkpoint at PlanTL-GOB-ES/bsc-bio-ehr-es and are newly initialized: ['classifier.bias', 'classifier.weight']
+You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
+
+
+
+/content/dissertation/scripts/ner/run_ner_train.py:397: FutureWarning: load_metric is deprecated and will be removed in the next major version of datasets. Use 'evaluate.load' instead, from the new library 🤗 Evaluate: https://huggingface.co/docs/evaluate
+  metric = load_metric("seqeval", trust_remote_code=True)
+[INFO|trainer.py:811] 2024-09-09 20:32:11,813 >> The following columns in the training set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: ner_tags, tokens, id. If ner_tags, tokens, id are not expected by `RobertaForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:2134] 2024-09-09 20:32:12,365 >> ***** Running training *****
+[INFO|trainer.py:2135] 2024-09-09 20:32:12,365 >>   Num examples = 31,229
+[INFO|trainer.py:2136] 2024-09-09 20:32:12,365 >>   Num Epochs = 10
+[INFO|trainer.py:2137] 2024-09-09 20:32:12,365 >>   Instantaneous batch size per device = 32
+[INFO|trainer.py:2140] 2024-09-09 20:32:12,365 >>   Total train batch size (w. parallel, distributed & accumulation) = 64
+[INFO|trainer.py:2141] 2024-09-09 20:32:12,365 >>   Gradient Accumulation steps = 2
+[INFO|trainer.py:2142] 2024-09-09 20:32:12,365 >>   Total optimization steps = 4,880
+[INFO|trainer.py:2143] 2024-09-09 20:32:12,366 >>   Number of trainable parameters = 124,055,043
+
  0%|          | 0/4880 [00:00<?, ?it/s]
  0%|          | 1/4880 [00:01<1:27:08,  1.07s/it]
  0%|          | 2/4880 [00:01<51:50,  1.57it/s]  
  0%|          | 3/4880 [00:01<40:04,  2.03it/s]
  0%|          | 4/4880 [00:01<32:00,  2.54it/s]
  0%|          | 5/4880 [00:02<27:14,  2.98it/s]
  0%|          | 6/4880 [00:02<23:04,  3.52it/s]
  0%|          | 7/4880 [00:02<20:44,  3.92it/s]
  0%|          | 8/4880 [00:02<19:50,  4.09it/s]
  0%|          | 9/4880 [00:02<18:14,  4.45it/s]
  0%|          | 10/4880 [00:03<18:16,  4.44it/s]
  0%|          | 11/4880 [00:03<18:21,  4.42it/s]
  0%|          | 12/4880 [00:03<18:13,  4.45it/s]
  0%|          | 13/4880 [00:03<20:48,  3.90it/s]
  0%|          | 14/4880 [00:04<19:12,  4.22it/s]
  0%|          | 15/4880 [00:04<18:57,  4.28it/s]
  0%|          | 16/4880 [00:04<20:26,  3.97it/s]
  0%|          | 17/4880 [00:04<18:54,  4.29it/s]
  0%|          | 18/4880 [00:05<18:43,  4.33it/s]
  0%|          | 19/4880 [00:05<17:40,  4.58it/s]
  0%|          | 20/4880 [00:05<17:42,  4.58it/s]
  0%|          | 21/4880 [00:05<19:40,  4.12it/s]
  0%|          | 22/4880 [00:06<21:06,  3.83it/s]
  0%|          | 23/4880 [00:06<20:44,  3.90it/s]
  0%|          | 24/4880 [00:06<19:05,  4.24it/s]
  1%|          | 25/4880 [00:06<19:15,  4.20it/s]
  1%|          | 26/4880 [00:07<32:43,  2.47it/s]
  1%|          | 27/4880 [00:07<29:28,  2.74it/s]
  1%|          | 28/4880 [00:08<25:53,  3.12it/s]
  1%|          | 29/4880 [00:08<25:25,  3.18it/s]
  1%|          | 30/4880 [00:08<24:29,  3.30it/s]
  1%|          | 31/4880 [00:08<22:37,  3.57it/s]
  1%|          | 32/4880 [00:09<22:12,  3.64it/s]
  1%|          | 33/4880 [00:09<23:20,  3.46it/s]
  1%|          | 34/4880 [00:09<20:17,  3.98it/s]
  1%|          | 35/4880 [00:09<22:23,  3.61it/s]
  1%|          | 36/4880 [00:10<20:55,  3.86it/s]
  1%|          | 37/4880 [00:10<19:06,  4.22it/s]
  1%|          | 38/4880 [00:10<19:05,  4.23it/s]
  1%|          | 39/4880 [00:10<18:39,  4.33it/s]
  1%|          | 40/4880 [00:11<17:34,  4.59it/s]
  1%|          | 41/4880 [00:11<17:43,  4.55it/s]
  1%|          | 42/4880 [00:11<18:14,  4.42it/s]
  1%|          | 43/4880 [00:11<20:42,  3.89it/s]
  1%|          | 44/4880 [00:12<20:32,  3.92it/s]
  1%|          | 45/4880 [00:12<19:46,  4.07it/s]
  1%|          | 46/4880 [00:12<23:54,  3.37it/s]
  1%|          | 47/4880 [00:13<24:10,  3.33it/s]
  1%|          | 48/4880 [00:13<21:41,  3.71it/s]
  1%|          | 49/4880 [00:13<20:20,  3.96it/s]
  1%|          | 50/4880 [00:13<22:35,  3.56it/s]
  1%|          | 51/4880 [00:13<20:26,  3.94it/s]
  1%|          | 52/4880 [00:14<23:08,  3.48it/s]
  1%|          | 53/4880 [00:14<21:48,  3.69it/s]
  1%|          | 54/4880 [00:14<20:26,  3.93it/s]
  1%|          | 55/4880 [00:14<19:01,  4.23it/s]
  1%|          | 56/4880 [00:15<19:21,  4.15it/s]
  1%|          | 57/4880 [00:15<18:40,  4.30it/s]
  1%|          | 58/4880 [00:15<18:21,  4.38it/s]
  1%|          | 59/4880 [00:15<21:15,  3.78it/s]
  1%|          | 60/4880 [00:16<19:39,  4.09it/s]
  1%|▏         | 61/4880 [00:16<18:42,  4.29it/s]
  1%|▏         | 62/4880 [00:16<21:29,  3.74it/s]
  1%|▏         | 63/4880 [00:16<19:12,  4.18it/s]
  1%|▏         | 64/4880 [00:17<20:40,  3.88it/s]
  1%|▏         | 65/4880 [00:17<21:24,  3.75it/s]
  1%|▏         | 66/4880 [00:17<20:20,  3.94it/s]
  1%|▏         | 67/4880 [00:18<21:13,  3.78it/s]
  1%|▏         | 68/4880 [00:18<22:13,  3.61it/s]
  1%|▏         | 69/4880 [00:18<27:02,  2.97it/s]
  1%|▏         | 70/4880 [00:19<24:24,  3.28it/s]
  1%|▏         | 71/4880 [00:19<23:54,  3.35it/s]
  1%|▏         | 72/4880 [00:19<22:06,  3.62it/s]
  1%|▏         | 73/4880 [00:19<20:21,  3.93it/s]
  2%|▏         | 74/4880 [00:19<20:15,  3.95it/s]
  2%|▏         | 75/4880 [00:20<18:29,  4.33it/s]
  2%|▏         | 76/4880 [00:20<19:05,  4.19it/s]
  2%|▏         | 77/4880 [00:20<18:28,  4.33it/s]
  2%|▏         | 78/4880 [00:21<22:00,  3.64it/s]
  2%|▏         | 79/4880 [00:21<20:54,  3.83it/s]
  2%|▏         | 80/4880 [00:21<21:07,  3.79it/s]
  2%|▏         | 81/4880 [00:21<19:31,  4.10it/s]
  2%|▏         | 82/4880 [00:21<18:51,  4.24it/s]
  2%|▏         | 83/4880 [00:22<17:55,  4.46it/s]
  2%|▏         | 84/4880 [00:22<18:01,  4.44it/s]
  2%|▏         | 85/4880 [00:22<16:52,  4.74it/s]
  2%|▏         | 86/4880 [00:22<20:18,  3.94it/s]
  2%|▏         | 87/4880 [00:23<18:12,  4.39it/s]
  2%|▏         | 88/4880 [00:23<18:41,  4.27it/s]
  2%|▏         | 89/4880 [00:23<20:06,  3.97it/s]
  2%|▏         | 90/4880 [00:23<18:53,  4.23it/s]
  2%|▏         | 91/4880 [00:24<20:59,  3.80it/s]
  2%|▏         | 92/4880 [00:24<21:49,  3.66it/s]
  2%|▏         | 93/4880 [00:24<20:38,  3.86it/s]
  2%|▏         | 94/4880 [00:24<20:24,  3.91it/s]
  2%|▏         | 95/4880 [00:25<19:17,  4.13it/s]
  2%|▏         | 96/4880 [00:25<18:43,  4.26it/s]
  2%|▏         | 97/4880 [00:25<17:33,  4.54it/s]
  2%|▏         | 98/4880 [00:26<26:49,  2.97it/s]
  2%|▏         | 99/4880 [00:26<24:45,  3.22it/s]
  2%|▏         | 100/4880 [00:26<23:12,  3.43it/s]
  2%|▏         | 101/4880 [00:26<21:27,  3.71it/s]
  2%|▏         | 102/4880 [00:27<19:15,  4.14it/s]
  2%|▏         | 103/4880 [00:27<20:09,  3.95it/s]
  2%|▏         | 104/4880 [00:27<18:53,  4.21it/s]
  2%|▏         | 105/4880 [00:27<18:41,  4.26it/s]
  2%|▏         | 106/4880 [00:28<22:02,  3.61it/s]
  2%|▏         | 107/4880 [00:28<21:52,  3.64it/s]
  2%|▏         | 108/4880 [00:28<20:29,  3.88it/s]
  2%|▏         | 109/4880 [00:28<22:58,  3.46it/s]
  2%|▏         | 110/4880 [00:29<20:57,  3.79it/s]
  2%|▏         | 111/4880 [00:29<20:23,  3.90it/s]
  2%|▏         | 112/4880 [00:29<19:20,  4.11it/s]
  2%|▏         | 113/4880 [00:29<18:24,  4.32it/s]
  2%|▏         | 114/4880 [00:30<20:43,  3.83it/s]
  2%|▏         | 115/4880 [00:30<19:47,  4.01it/s]
  2%|▏         | 116/4880 [00:30<19:29,  4.07it/s]
  2%|▏         | 117/4880 [00:31<23:37,  3.36it/s]
  2%|▏         | 118/4880 [00:31<21:15,  3.73it/s]
  2%|▏         | 119/4880 [00:31<19:30,  4.07it/s]
  2%|▏         | 120/4880 [00:31<17:52,  4.44it/s]
  2%|▏         | 121/4880 [00:31<18:00,  4.40it/s]
  2%|▎         | 122/4880 [00:32<19:07,  4.15it/s]
  3%|▎         | 123/4880 [00:32<20:05,  3.95it/s]
  3%|▎         | 124/4880 [00:32<20:37,  3.84it/s]
  3%|▎         | 125/4880 [00:32<21:36,  3.67it/s]
  3%|▎         | 126/4880 [00:33<20:03,  3.95it/s]
  3%|▎         | 127/4880 [00:33<19:36,  4.04it/s]
  3%|▎         | 128/4880 [00:33<18:48,  4.21it/s]
  3%|▎         | 129/4880 [00:33<18:09,  4.36it/s]
  3%|▎         | 130/4880 [00:34<18:15,  4.34it/s]
  3%|▎         | 131/4880 [00:34<18:29,  4.28it/s]
  3%|▎         | 132/4880 [00:34<17:20,  4.56it/s]
  3%|▎         | 133/4880 [00:34<15:50,  4.99it/s]
  3%|▎         | 134/4880 [00:34<15:53,  4.98it/s]
  3%|▎         | 135/4880 [00:35<18:29,  4.28it/s]
  3%|▎         | 136/4880 [00:35<20:17,  3.90it/s]
  3%|▎         | 137/4880 [00:35<18:54,  4.18it/s]
  3%|▎         | 138/4880 [00:35<19:28,  4.06it/s]
  3%|▎         | 139/4880 [00:36<18:04,  4.37it/s]
  3%|▎         | 140/4880 [00:36<17:41,  4.46it/s]
  3%|▎         | 141/4880 [00:36<20:00,  3.95it/s]
  3%|▎         | 142/4880 [00:36<18:07,  4.36it/s]
  3%|▎         | 143/4880 [00:37<17:16,  4.57it/s]
  3%|▎         | 144/4880 [00:37<17:35,  4.49it/s]
  3%|▎         | 145/4880 [00:37<17:54,  4.41it/s]
  3%|▎         | 146/4880 [00:37<21:07,  3.74it/s]
  3%|▎         | 147/4880 [00:38<19:57,  3.95it/s]
  3%|▎         | 148/4880 [00:38<23:03,  3.42it/s]
  3%|▎         | 149/4880 [00:38<22:06,  3.57it/s]
  3%|▎         | 150/4880 [00:38<21:17,  3.70it/s]
  3%|▎         | 151/4880 [00:39<23:05,  3.41it/s]
  3%|▎         | 152/4880 [00:39<20:47,  3.79it/s]
  3%|▎         | 153/4880 [00:39<19:52,  3.97it/s]
  3%|▎         | 154/4880 [00:39<19:24,  4.06it/s]
  3%|▎         | 155/4880 [00:40<23:42,  3.32it/s]
  3%|▎         | 156/4880 [00:40<21:24,  3.68it/s]
  3%|▎         | 157/4880 [00:40<22:00,  3.58it/s]
  3%|▎         | 158/4880 [00:41<23:59,  3.28it/s]
  3%|▎         | 159/4880 [00:41<23:10,  3.39it/s]
  3%|▎         | 160/4880 [00:41<20:35,  3.82it/s]
  3%|▎         | 161/4880 [00:41<19:23,  4.05it/s]
  3%|▎         | 162/4880 [00:42<21:14,  3.70it/s]
  3%|▎         | 163/4880 [00:42<21:41,  3.63it/s]
  3%|▎         | 164/4880 [00:42<20:54,  3.76it/s]
  3%|▎         | 165/4880 [00:43<26:29,  2.97it/s]
  3%|▎         | 166/4880 [00:43<23:33,  3.34it/s]
  3%|▎         | 167/4880 [00:43<23:51,  3.29it/s]
  3%|▎         | 168/4880 [00:44<22:20,  3.51it/s]
  3%|▎         | 169/4880 [00:44<22:18,  3.52it/s]
  3%|▎         | 170/4880 [00:44<21:28,  3.65it/s]
  4%|▎         | 171/4880 [00:44<20:39,  3.80it/s]
  4%|▎         | 172/4880 [00:45<19:56,  3.93it/s]
  4%|▎         | 173/4880 [00:45<18:21,  4.27it/s]
  4%|▎         | 174/4880 [00:45<18:18,  4.28it/s]
  4%|▎         | 175/4880 [00:45<17:23,  4.51it/s]
  4%|▎         | 176/4880 [00:45<16:41,  4.70it/s]
  4%|▎         | 177/4880 [00:46<16:14,  4.82it/s]
  4%|▎         | 178/4880 [00:46<16:21,  4.79it/s]
  4%|▎         | 179/4880 [00:46<17:10,  4.56it/s]
  4%|▎         | 180/4880 [00:46<17:32,  4.47it/s]
  4%|▎         | 181/4880 [00:46<18:11,  4.30it/s]
  4%|▎         | 182/4880 [00:47<21:09,  3.70it/s]
  4%|▍         | 183/4880 [00:47<20:04,  3.90it/s]
  4%|▍         | 184/4880 [00:47<20:54,  3.74it/s]
  4%|▍         | 185/4880 [00:48<20:14,  3.87it/s]
  4%|▍         | 186/4880 [00:48<19:08,  4.09it/s]
  4%|▍         | 187/4880 [00:48<17:28,  4.48it/s]
  4%|▍         | 188/4880 [00:48<19:06,  4.09it/s]
  4%|▍         | 189/4880 [00:49<18:58,  4.12it/s]
  4%|▍         | 190/4880 [00:49<18:33,  4.21it/s]
  4%|▍         | 191/4880 [00:49<19:33,  4.00it/s]
  4%|▍         | 192/4880 [00:49<18:21,  4.25it/s]
  4%|▍         | 193/4880 [00:50<21:23,  3.65it/s]
  4%|▍         | 194/4880 [00:50<20:18,  3.85it/s]
  4%|▍         | 195/4880 [00:50<18:47,  4.15it/s]
  4%|▍         | 196/4880 [00:50<18:15,  4.28it/s]
  4%|▍         | 197/4880 [00:51<29:10,  2.68it/s]
  4%|▍         | 198/4880 [00:51<25:33,  3.05it/s]
  4%|▍         | 199/4880 [00:51<23:17,  3.35it/s]
  4%|▍         | 200/4880 [00:52<21:36,  3.61it/s]
  4%|▍         | 201/4880 [00:52<19:17,  4.04it/s]
  4%|▍         | 202/4880 [00:52<22:08,  3.52it/s]
  4%|▍         | 203/4880 [00:52<20:03,  3.89it/s]
  4%|▍         | 204/4880 [00:53<17:55,  4.35it/s]
  4%|▍         | 205/4880 [00:53<17:16,  4.51it/s]
  4%|▍         | 206/4880 [00:53<17:24,  4.47it/s]
  4%|▍         | 207/4880 [00:53<18:48,  4.14it/s]
  4%|▍         | 208/4880 [00:53<18:10,  4.28it/s]
  4%|▍         | 209/4880 [00:54<19:10,  4.06it/s]
  4%|▍         | 210/4880 [00:54<17:59,  4.32it/s]
  4%|▍         | 211/4880 [00:54<17:59,  4.32it/s]
  4%|▍         | 212/4880 [00:54<17:19,  4.49it/s]
  4%|▍         | 213/4880 [00:55<16:57,  4.59it/s]
  4%|▍         | 214/4880 [00:55<19:35,  3.97it/s]
  4%|▍         | 215/4880 [00:55<18:48,  4.13it/s]
  4%|▍         | 216/4880 [00:55<18:20,  4.24it/s]
  4%|▍         | 217/4880 [00:56<17:47,  4.37it/s]
  4%|▍         | 218/4880 [00:56<19:04,  4.07it/s]
  4%|▍         | 219/4880 [00:56<18:38,  4.17it/s]
  5%|▍         | 220/4880 [00:56<19:40,  3.95it/s]
  5%|▍         | 221/4880 [00:57<20:54,  3.71it/s]
  5%|▍         | 222/4880 [00:57<19:41,  3.94it/s]
  5%|▍         | 223/4880 [00:57<21:12,  3.66it/s]
  5%|▍         | 224/4880 [00:57<20:18,  3.82it/s]
  5%|▍         | 225/4880 [00:58<20:56,  3.70it/s]
  5%|▍         | 226/4880 [00:58<18:45,  4.14it/s]
  5%|▍         | 227/4880 [00:58<18:32,  4.18it/s]
  5%|▍         | 228/4880 [00:58<20:35,  3.76it/s]
  5%|▍         | 229/4880 [00:59<29:58,  2.59it/s]
  5%|▍         | 230/4880 [00:59<27:43,  2.80it/s]
  5%|▍         | 231/4880 [01:00<24:54,  3.11it/s]
  5%|▍         | 232/4880 [01:00<25:56,  2.99it/s]
  5%|▍         | 233/4880 [01:00<26:18,  2.94it/s]
  5%|▍         | 234/4880 [01:01<25:14,  3.07it/s]
  5%|▍         | 235/4880 [01:01<22:07,  3.50it/s]
  5%|▍         | 236/4880 [01:01<20:46,  3.72it/s]
  5%|▍         | 237/4880 [01:01<19:57,  3.88it/s]
  5%|▍         | 238/4880 [01:01<18:35,  4.16it/s]
  5%|▍         | 239/4880 [01:02<20:47,  3.72it/s]
  5%|▍         | 240/4880 [01:02<20:05,  3.85it/s]
  5%|▍         | 241/4880 [01:03<30:01,  2.57it/s]
  5%|▍         | 242/4880 [01:03<28:00,  2.76it/s]
  5%|▍         | 243/4880 [01:04<36:16,  2.13it/s]
  5%|▌         | 244/4880 [01:04<32:19,  2.39it/s]
  5%|▌         | 245/4880 [01:04<27:58,  2.76it/s]
  5%|▌         | 246/4880 [01:05<25:08,  3.07it/s]
  5%|▌         | 247/4880 [01:05<22:57,  3.36it/s]
  5%|▌         | 248/4880 [01:05<20:15,  3.81it/s]
  5%|▌         | 249/4880 [01:05<19:44,  3.91it/s]
  5%|▌         | 250/4880 [01:06<21:45,  3.55it/s]
  5%|▌         | 251/4880 [01:06<21:55,  3.52it/s]
  5%|▌         | 252/4880 [01:06<22:48,  3.38it/s]
  5%|▌         | 253/4880 [01:06<21:55,  3.52it/s]
  5%|▌         | 254/4880 [01:07<20:59,  3.67it/s]
  5%|▌         | 255/4880 [01:07<19:14,  4.01it/s]
  5%|▌         | 256/4880 [01:07<19:15,  4.00it/s]
  5%|▌         | 257/4880 [01:07<18:57,  4.07it/s]
  5%|▌         | 258/4880 [01:08<20:10,  3.82it/s]
  5%|▌         | 259/4880 [01:08<22:10,  3.47it/s]
  5%|▌         | 260/4880 [01:08<20:45,  3.71it/s]
  5%|▌         | 261/4880 [01:09<22:42,  3.39it/s]
  5%|▌         | 262/4880 [01:09<22:25,  3.43it/s]
  5%|▌         | 263/4880 [01:09<22:13,  3.46it/s]
  5%|▌         | 264/4880 [01:09<20:42,  3.72it/s]
  5%|▌         | 265/4880 [01:10<19:39,  3.91it/s]
  5%|▌         | 266/4880 [01:10<18:21,  4.19it/s]
  5%|▌         | 267/4880 [01:10<29:19,  2.62it/s]
  5%|▌         | 268/4880 [01:11<25:09,  3.06it/s]
  6%|▌         | 269/4880 [01:11<21:44,  3.53it/s]
  6%|▌         | 270/4880 [01:11<20:29,  3.75it/s]
  6%|▌         | 271/4880 [01:11<19:14,  3.99it/s]
  6%|▌         | 272/4880 [01:11<17:20,  4.43it/s]
  6%|▌         | 273/4880 [01:12<17:19,  4.43it/s]
  6%|▌         | 274/4880 [01:12<18:06,  4.24it/s]
  6%|▌         | 275/4880 [01:12<18:07,  4.24it/s]
  6%|▌         | 276/4880 [01:12<17:05,  4.49it/s]
  6%|▌         | 277/4880 [01:13<17:08,  4.48it/s]
  6%|▌         | 278/4880 [01:13<18:37,  4.12it/s]
  6%|▌         | 279/4880 [01:13<17:25,  4.40it/s]
  6%|▌         | 280/4880 [01:13<16:56,  4.52it/s]
  6%|▌         | 281/4880 [01:14<18:06,  4.23it/s]
  6%|▌         | 282/4880 [01:14<19:03,  4.02it/s]
  6%|▌         | 283/4880 [01:14<18:20,  4.18it/s]
  6%|▌         | 284/4880 [01:14<20:47,  3.69it/s]
  6%|▌         | 285/4880 [01:15<27:14,  2.81it/s]
  6%|▌         | 286/4880 [01:15<24:54,  3.07it/s]
  6%|▌         | 287/4880 [01:15<21:53,  3.50it/s]
  6%|▌         | 288/4880 [01:16<21:35,  3.54it/s]
  6%|▌         | 289/4880 [01:16<20:00,  3.82it/s]
  6%|▌         | 290/4880 [01:16<18:43,  4.09it/s]
  6%|▌         | 291/4880 [01:16<18:31,  4.13it/s]
  6%|▌         | 292/4880 [01:17<19:01,  4.02it/s]
  6%|▌         | 293/4880 [01:17<19:44,  3.87it/s]
  6%|▌         | 294/4880 [01:17<18:35,  4.11it/s]
  6%|▌         | 295/4880 [01:18<22:10,  3.45it/s]
  6%|▌         | 296/4880 [01:18<22:03,  3.46it/s]
  6%|▌         | 297/4880 [01:18<21:03,  3.63it/s]
  6%|▌         | 298/4880 [01:18<20:06,  3.80it/s]
  6%|▌         | 299/4880 [01:18<18:38,  4.09it/s]
  6%|▌         | 300/4880 [01:19<20:05,  3.80it/s]
  6%|▌         | 301/4880 [01:19<19:06,  3.99it/s]
  6%|▌         | 302/4880 [01:19<18:59,  4.02it/s]
  6%|▌         | 303/4880 [01:20<19:47,  3.85it/s]
  6%|▌         | 304/4880 [01:20<18:31,  4.12it/s]
  6%|▋         | 305/4880 [01:20<19:39,  3.88it/s]
  6%|▋         | 306/4880 [01:20<19:49,  3.85it/s]
  6%|▋         | 307/4880 [01:21<21:08,  3.61it/s]
  6%|▋         | 308/4880 [01:21<24:27,  3.12it/s]
  6%|▋         | 309/4880 [01:21<23:29,  3.24it/s]
  6%|▋         | 310/4880 [01:22<21:31,  3.54it/s]
  6%|▋         | 311/4880 [01:22<22:59,  3.31it/s]
  6%|▋         | 312/4880 [01:22<21:31,  3.54it/s]
  6%|▋         | 313/4880 [01:22<19:46,  3.85it/s]
  6%|▋         | 314/4880 [01:23<20:07,  3.78it/s]
  6%|▋         | 315/4880 [01:23<18:53,  4.03it/s]
  6%|▋         | 316/4880 [01:23<17:05,  4.45it/s]
  6%|▋         | 317/4880 [01:23<17:06,  4.45it/s]
  7%|▋         | 318/4880 [01:23<17:40,  4.30it/s]
  7%|▋         | 319/4880 [01:24<17:50,  4.26it/s]
  7%|▋         | 320/4880 [01:24<17:07,  4.44it/s]
  7%|▋         | 321/4880 [01:24<16:20,  4.65it/s]
  7%|▋         | 322/4880 [01:24<15:45,  4.82it/s]
  7%|▋         | 323/4880 [01:25<18:53,  4.02it/s]
  7%|▋         | 324/4880 [01:25<20:46,  3.66it/s]
  7%|▋         | 325/4880 [01:25<19:36,  3.87it/s]
  7%|▋         | 326/4880 [01:25<17:54,  4.24it/s]
  7%|▋         | 327/4880 [01:26<17:58,  4.22it/s]
  7%|▋         | 328/4880 [01:26<20:19,  3.73it/s]
  7%|▋         | 329/4880 [01:26<19:48,  3.83it/s]
  7%|▋         | 330/4880 [01:26<19:04,  3.98it/s]
  7%|▋         | 331/4880 [01:27<17:18,  4.38it/s]
  7%|▋         | 332/4880 [01:27<16:50,  4.50it/s]
  7%|▋         | 333/4880 [01:27<17:46,  4.26it/s]
  7%|▋         | 334/4880 [01:27<20:28,  3.70it/s]
  7%|▋         | 335/4880 [01:28<21:43,  3.49it/s]
  7%|▋         | 336/4880 [01:28<20:54,  3.62it/s]
  7%|▋         | 337/4880 [01:28<20:25,  3.71it/s]
  7%|▋         | 338/4880 [01:29<20:40,  3.66it/s]
  7%|▋         | 339/4880 [01:29<20:11,  3.75it/s]
  7%|▋         | 340/4880 [01:29<20:01,  3.78it/s]
  7%|▋         | 341/4880 [01:29<19:43,  3.83it/s]
  7%|▋         | 342/4880 [01:30<19:45,  3.83it/s]
  7%|▋         | 343/4880 [01:30<18:33,  4.08it/s]
  7%|▋         | 344/4880 [01:30<17:39,  4.28it/s]
  7%|▋         | 345/4880 [01:30<17:27,  4.33it/s]
  7%|▋         | 346/4880 [01:30<18:11,  4.15it/s]
  7%|▋         | 347/4880 [01:31<17:37,  4.29it/s]
  7%|▋         | 348/4880 [01:31<16:00,  4.72it/s]
  7%|▋         | 349/4880 [01:31<16:36,  4.55it/s]
  7%|▋         | 350/4880 [01:31<17:36,  4.29it/s]
  7%|▋         | 351/4880 [01:32<17:30,  4.31it/s]
  7%|▋         | 352/4880 [01:32<17:19,  4.35it/s]
  7%|▋         | 353/4880 [01:32<18:27,  4.09it/s]
  7%|▋         | 354/4880 [01:32<18:16,  4.13it/s]
  7%|▋         | 355/4880 [01:33<19:37,  3.84it/s]
  7%|▋         | 356/4880 [01:33<18:40,  4.04it/s]
  7%|▋         | 357/4880 [01:33<17:29,  4.31it/s]
  7%|▋         | 358/4880 [01:33<16:25,  4.59it/s]
  7%|▋         | 359/4880 [01:33<17:48,  4.23it/s]
  7%|▋         | 360/4880 [01:34<20:25,  3.69it/s]
  7%|▋         | 361/4880 [01:34<19:07,  3.94it/s]
  7%|▋         | 362/4880 [01:34<18:07,  4.15it/s]
  7%|▋         | 363/4880 [01:35<18:28,  4.08it/s]
  7%|▋         | 364/4880 [01:35<17:45,  4.24it/s]
  7%|▋         | 365/4880 [01:35<17:06,  4.40it/s]
  8%|▊         | 366/4880 [01:35<17:25,  4.32it/s]
  8%|▊         | 367/4880 [01:35<16:40,  4.51it/s]
  8%|▊         | 368/4880 [01:36<17:05,  4.40it/s]
  8%|▊         | 369/4880 [01:36<18:25,  4.08it/s]
  8%|▊         | 370/4880 [01:36<17:22,  4.33it/s]
  8%|▊         | 371/4880 [01:36<16:00,  4.69it/s]
  8%|▊         | 372/4880 [01:37<17:24,  4.32it/s]
  8%|▊         | 373/4880 [01:37<17:22,  4.32it/s]
  8%|▊         | 374/4880 [01:37<17:59,  4.17it/s]
  8%|▊         | 375/4880 [01:37<18:57,  3.96it/s]
  8%|▊         | 376/4880 [01:37<17:13,  4.36it/s]
  8%|▊         | 377/4880 [01:38<18:29,  4.06it/s]
  8%|▊         | 378/4880 [01:38<20:46,  3.61it/s]
  8%|▊         | 379/4880 [01:38<19:21,  3.88it/s]
  8%|▊         | 380/4880 [01:39<18:44,  4.00it/s]
  8%|▊         | 381/4880 [01:39<19:06,  3.93it/s]
  8%|▊         | 382/4880 [01:39<17:53,  4.19it/s]
  8%|▊         | 383/4880 [01:39<17:10,  4.36it/s]
  8%|▊         | 384/4880 [01:40<18:17,  4.10it/s]
  8%|▊         | 385/4880 [01:40<17:22,  4.31it/s]
  8%|▊         | 386/4880 [01:40<17:48,  4.21it/s]
  8%|▊         | 387/4880 [01:40<18:39,  4.01it/s]
  8%|▊         | 388/4880 [01:40<17:23,  4.31it/s]
  8%|▊         | 389/4880 [01:41<17:41,  4.23it/s]
  8%|▊         | 390/4880 [01:41<16:44,  4.47it/s]
  8%|▊         | 391/4880 [01:41<15:36,  4.79it/s]
  8%|▊         | 392/4880 [01:41<15:45,  4.75it/s]
  8%|▊         | 393/4880 [01:42<16:28,  4.54it/s]
  8%|▊         | 394/4880 [01:42<15:57,  4.69it/s]
  8%|▊         | 395/4880 [01:42<15:47,  4.73it/s]
  8%|▊         | 396/4880 [01:42<14:58,  4.99it/s]
  8%|▊         | 397/4880 [01:42<14:27,  5.17it/s]
  8%|▊         | 398/4880 [01:43<15:28,  4.83it/s]
  8%|▊         | 399/4880 [01:43<17:24,  4.29it/s]
  8%|▊         | 400/4880 [01:43<19:01,  3.92it/s]
  8%|▊         | 401/4880 [01:43<19:28,  3.83it/s]
  8%|▊         | 402/4880 [01:44<18:12,  4.10it/s]
  8%|▊         | 403/4880 [01:44<18:03,  4.13it/s]
  8%|▊         | 404/4880 [01:44<20:43,  3.60it/s]
  8%|▊         | 405/4880 [01:44<20:33,  3.63it/s]
  8%|▊         | 406/4880 [01:45<20:00,  3.73it/s]
  8%|▊         | 407/4880 [01:45<18:32,  4.02it/s]
  8%|▊         | 408/4880 [01:45<17:42,  4.21it/s]
  8%|▊         | 409/4880 [01:45<17:36,  4.23it/s]
  8%|▊         | 410/4880 [01:46<17:49,  4.18it/s]
  8%|▊         | 411/4880 [01:46<23:07,  3.22it/s]
  8%|▊         | 412/4880 [01:46<20:55,  3.56it/s]
  8%|▊         | 413/4880 [01:46<19:01,  3.91it/s]
  8%|▊         | 414/4880 [01:47<19:54,  3.74it/s]
  9%|▊         | 415/4880 [01:47<19:32,  3.81it/s]
  9%|▊         | 416/4880 [01:47<18:28,  4.03it/s]
  9%|▊         | 417/4880 [01:47<17:32,  4.24it/s]
  9%|▊         | 418/4880 [01:48<18:10,  4.09it/s]
  9%|▊         | 419/4880 [01:48<17:53,  4.15it/s]
  9%|▊         | 420/4880 [01:48<16:12,  4.59it/s]
  9%|▊         | 421/4880 [01:48<15:53,  4.68it/s]
  9%|▊         | 422/4880 [01:49<18:30,  4.02it/s]
  9%|▊         | 423/4880 [01:49<18:39,  3.98it/s]
  9%|▊         | 424/4880 [01:49<19:06,  3.89it/s]
  9%|▊         | 425/4880 [01:49<17:52,  4.15it/s]
  9%|▊         | 426/4880 [01:50<19:16,  3.85it/s]
  9%|▉         | 427/4880 [01:50<18:09,  4.09it/s]
  9%|▉         | 428/4880 [01:50<17:46,  4.18it/s]
  9%|▉         | 429/4880 [01:51<20:36,  3.60it/s]
  9%|▉         | 430/4880 [01:51<19:26,  3.81it/s]
  9%|▉         | 431/4880 [01:51<19:11,  3.86it/s]
  9%|▉         | 432/4880 [01:51<18:48,  3.94it/s]
  9%|▉         | 433/4880 [01:51<18:50,  3.93it/s]
  9%|▉         | 434/4880 [01:52<19:43,  3.76it/s]
  9%|▉         | 435/4880 [01:52<17:27,  4.24it/s]
  9%|▉         | 436/4880 [01:52<16:47,  4.41it/s]
  9%|▉         | 437/4880 [01:52<16:02,  4.62it/s]
  9%|▉         | 438/4880 [01:53<16:47,  4.41it/s]
  9%|▉         | 439/4880 [01:53<18:35,  3.98it/s]
  9%|▉         | 440/4880 [01:53<17:37,  4.20it/s]
  9%|▉         | 441/4880 [01:53<18:43,  3.95it/s]
  9%|▉         | 442/4880 [01:54<18:25,  4.01it/s]
  9%|▉         | 443/4880 [01:54<17:47,  4.16it/s]
  9%|▉         | 444/4880 [01:54<18:59,  3.89it/s]
  9%|▉         | 445/4880 [01:54<17:10,  4.30it/s]
  9%|▉         | 446/4880 [01:55<16:48,  4.40it/s]
  9%|▉         | 447/4880 [01:55<15:28,  4.77it/s]
  9%|▉         | 448/4880 [01:55<15:04,  4.90it/s]
  9%|▉         | 449/4880 [01:55<15:11,  4.86it/s]
  9%|▉         | 450/4880 [01:55<14:54,  4.95it/s]
  9%|▉         | 451/4880 [01:56<15:07,  4.88it/s]
  9%|▉         | 452/4880 [01:56<15:59,  4.61it/s]
  9%|▉         | 453/4880 [01:56<17:40,  4.17it/s]
  9%|▉         | 454/4880 [01:56<16:52,  4.37it/s]
  9%|▉         | 455/4880 [01:57<17:33,  4.20it/s]
  9%|▉         | 456/4880 [01:57<17:25,  4.23it/s]
  9%|▉         | 457/4880 [01:57<16:26,  4.48it/s]
  9%|▉         | 458/4880 [01:57<15:19,  4.81it/s]
  9%|▉         | 459/4880 [01:57<14:45,  4.99it/s]
  9%|▉         | 460/4880 [01:58<15:33,  4.73it/s]
  9%|▉         | 461/4880 [01:58<16:51,  4.37it/s]
  9%|▉         | 462/4880 [01:58<16:24,  4.49it/s]
  9%|▉         | 463/4880 [01:58<16:46,  4.39it/s]
 10%|▉         | 464/4880 [01:58<17:21,  4.24it/s]
 10%|▉         | 465/4880 [01:59<21:30,  3.42it/s]
 10%|▉         | 466/4880 [01:59<22:22,  3.29it/s]
 10%|▉         | 467/4880 [02:00<21:53,  3.36it/s]
 10%|▉         | 468/4880 [02:00<24:02,  3.06it/s]
 10%|▉         | 469/4880 [02:00<24:19,  3.02it/s]
 10%|▉         | 470/4880 [02:01<22:17,  3.30it/s]
 10%|▉         | 471/4880 [02:01<21:14,  3.46it/s]
 10%|▉         | 472/4880 [02:01<19:40,  3.73it/s]
 10%|▉         | 473/4880 [02:01<18:23,  3.99it/s]
 10%|▉         | 474/4880 [02:01<16:40,  4.40it/s]
 10%|▉         | 475/4880 [02:02<15:14,  4.82it/s]
 10%|▉         | 476/4880 [02:02<15:35,  4.71it/s]
 10%|▉         | 477/4880 [02:02<19:30,  3.76it/s]
 10%|▉         | 478/4880 [02:02<18:43,  3.92it/s]
 10%|▉         | 479/4880 [02:03<19:32,  3.76it/s]
 10%|▉         | 480/4880 [02:03<18:19,  4.00it/s]
 10%|▉         | 481/4880 [02:03<16:35,  4.42it/s]
 10%|▉         | 482/4880 [02:03<16:39,  4.40it/s]
 10%|▉         | 483/4880 [02:04<18:17,  4.00it/s]
 10%|▉         | 484/4880 [02:04<17:41,  4.14it/s]
 10%|▉         | 485/4880 [02:04<19:41,  3.72it/s]
 10%|▉         | 486/4880 [02:04<17:24,  4.21it/s]
 10%|▉         | 487/4880 [02:05<17:49,  4.11it/s]
 10%|█         | 488/4880 [02:05<17:13,  4.25it/s][INFO|trainer.py:811] 2024-09-09 20:34:17,645 >> The following columns in the evaluation set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: ner_tags, tokens, id. If ner_tags, tokens, id are not expected by `RobertaForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3819] 2024-09-09 20:34:17,647 >> 
+***** Running Evaluation *****
+[INFO|trainer.py:3821] 2024-09-09 20:34:17,647 >>   Num examples = 6810
+[INFO|trainer.py:3824] 2024-09-09 20:34:17,647 >>   Batch size = 8
+
+
  0%|          | 0/852 [00:00<?, ?it/s]
+
  1%|          | 10/852 [00:00<00:08, 95.08it/s]
+
  2%|▏         | 20/852 [00:00<00:09, 84.15it/s]
+
  3%|▎         | 29/852 [00:00<00:09, 82.37it/s]
+
  4%|▍         | 38/852 [00:00<00:09, 81.48it/s]
+
  6%|▌         | 47/852 [00:00<00:09, 81.77it/s]
+
  7%|▋         | 56/852 [00:00<00:09, 82.90it/s]
+
  8%|▊         | 65/852 [00:00<00:09, 81.44it/s]
+
  9%|▊         | 74/852 [00:00<00:09, 78.40it/s]
+
 10%|▉         | 82/852 [00:01<00:09, 77.16it/s]
+
 11%|█         | 91/852 [00:01<00:09, 78.96it/s]
+
 12%|█▏        | 100/852 [00:01<00:09, 79.19it/s]
+
 13%|█▎        | 109/852 [00:01<00:09, 79.85it/s]
+
 14%|█▍        | 118/852 [00:01<00:09, 81.27it/s]
+
 15%|█▍        | 127/852 [00:01<00:09, 78.40it/s]
+
 16%|█▌        | 136/852 [00:01<00:08, 79.81it/s]
+
 17%|█▋        | 145/852 [00:01<00:08, 79.57it/s]
+
 18%|█▊        | 154/852 [00:01<00:08, 79.99it/s]
+
 19%|█▉        | 163/852 [00:02<00:08, 81.18it/s]
+
 20%|██        | 172/852 [00:02<00:08, 82.07it/s]
+
 21%|██        | 181/852 [00:02<00:08, 81.93it/s]
+
 22%|██▏       | 190/852 [00:02<00:07, 82.93it/s]
+
 23%|██▎       | 199/852 [00:02<00:07, 82.66it/s]
+
 24%|██▍       | 208/852 [00:02<00:07, 81.16it/s]
+
 25%|██▌       | 217/852 [00:02<00:07, 81.34it/s]
+
 27%|██▋       | 226/852 [00:02<00:07, 81.72it/s]
+
 28%|██▊       | 235/852 [00:02<00:07, 80.54it/s]
+
 29%|██▊       | 244/852 [00:03<00:07, 77.84it/s]
+
 30%|██▉       | 253/852 [00:03<00:07, 80.37it/s]
+
 31%|███       | 262/852 [00:03<00:07, 82.06it/s]
+
 32%|███▏      | 271/852 [00:03<00:07, 81.57it/s]
+
 33%|███▎      | 280/852 [00:03<00:06, 83.40it/s]
+
 34%|███▍      | 289/852 [00:03<00:06, 82.65it/s]
+
 35%|███▍      | 298/852 [00:03<00:06, 83.45it/s]
+
 36%|███▌      | 307/852 [00:03<00:06, 84.22it/s]
+
 37%|███▋      | 316/852 [00:03<00:06, 82.13it/s]
+
 38%|███▊      | 325/852 [00:03<00:06, 82.32it/s]
+
 39%|███▉      | 334/852 [00:04<00:06, 82.25it/s]
+
 40%|████      | 343/852 [00:04<00:06, 83.35it/s]
+
 41%|████▏     | 352/852 [00:04<00:06, 83.19it/s]
+
 42%|████▏     | 361/852 [00:04<00:06, 80.97it/s]
+
 43%|████▎     | 370/852 [00:04<00:05, 81.66it/s]
+
 44%|████▍     | 379/852 [00:04<00:05, 81.74it/s]
+
 46%|████▌     | 388/852 [00:04<00:05, 81.25it/s]
+
 47%|████▋     | 397/852 [00:04<00:05, 81.47it/s]
+
 48%|████▊     | 406/852 [00:05<00:05, 78.66it/s]
+
 49%|████▊     | 415/852 [00:05<00:05, 79.89it/s]
+
 50%|████▉     | 424/852 [00:05<00:05, 80.80it/s]
+
 51%|█████     | 433/852 [00:05<00:05, 80.84it/s]
+
 52%|█████▏    | 442/852 [00:05<00:04, 82.40it/s]
+
 53%|█████▎    | 451/852 [00:05<00:04, 82.75it/s]
+
 54%|█████▍    | 460/852 [00:05<00:04, 82.33it/s]
+
 55%|█████▌    | 469/852 [00:05<00:04, 80.45it/s]
+
 56%|█████▌    | 478/852 [00:05<00:04, 77.02it/s]
+
 57%|█████▋    | 486/852 [00:05<00:04, 77.70it/s]
+
 58%|█████▊    | 495/852 [00:06<00:04, 79.98it/s]
+
 59%|█████▉    | 504/852 [00:06<00:04, 81.79it/s]
+
 60%|██████    | 513/852 [00:06<00:04, 82.10it/s]
+
 61%|██████▏   | 522/852 [00:06<00:04, 81.21it/s]
+
 62%|██████▏   | 531/852 [00:06<00:03, 81.14it/s]
+
 63%|██████▎   | 540/852 [00:06<00:03, 82.49it/s]
+
 64%|██████▍   | 549/852 [00:06<00:03, 80.76it/s]
+
 65%|██████▌   | 558/852 [00:06<00:03, 81.05it/s]
+
 67%|██████▋   | 567/852 [00:06<00:03, 81.88it/s]
+
 68%|██████▊   | 576/852 [00:07<00:03, 82.38it/s]
+
 69%|██████▊   | 585/852 [00:07<00:03, 80.85it/s]
+
 70%|██████▉   | 594/852 [00:07<00:03, 81.90it/s]
+
 71%|███████   | 603/852 [00:07<00:03, 81.87it/s]
+
 72%|███████▏  | 612/852 [00:07<00:02, 80.25it/s]
+
 73%|███████▎  | 621/852 [00:07<00:02, 80.24it/s]
+
 74%|███████▍  | 630/852 [00:07<00:02, 78.83it/s]
+
 75%|███████▌  | 639/852 [00:07<00:02, 79.78it/s]
+
 76%|███████▌  | 647/852 [00:07<00:02, 77.63it/s]
+
 77%|███████▋  | 656/852 [00:08<00:02, 79.59it/s]
+
 78%|███████▊  | 665/852 [00:08<00:02, 80.43it/s]
+
 79%|███████▉  | 674/852 [00:08<00:02, 80.09it/s]
+
 80%|████████  | 683/852 [00:08<00:02, 81.05it/s]
+
 81%|████████  | 692/852 [00:08<00:01, 82.71it/s]
+
 82%|████████▏ | 701/852 [00:08<00:01, 83.34it/s]
+
 83%|████████▎ | 710/852 [00:08<00:01, 84.26it/s]
+
 84%|████████▍ | 719/852 [00:08<00:01, 83.19it/s]
+
 85%|████████▌ | 728/852 [00:08<00:01, 83.95it/s]
+
 87%|████████▋ | 737/852 [00:09<00:01, 83.62it/s]
+
 88%|████████▊ | 746/852 [00:09<00:01, 84.17it/s]
+
 89%|████████▊ | 755/852 [00:09<00:01, 84.42it/s]
+
 90%|████████▉ | 764/852 [00:09<00:01, 84.99it/s]
+
 91%|█████████ | 773/852 [00:09<00:00, 83.65it/s]
+
 92%|█████████▏| 782/852 [00:09<00:00, 81.82it/s]
+
 93%|█████████▎| 791/852 [00:09<00:00, 82.51it/s]
+
 94%|█████████▍| 800/852 [00:09<00:00, 82.35it/s]
+
 95%|█████████▍| 809/852 [00:09<00:00, 83.95it/s]
+
 96%|█████████▌| 818/852 [00:10<00:00, 82.34it/s]
+
 97%|█████████▋| 827/852 [00:10<00:00, 84.00it/s]
+
 98%|█████████▊| 836/852 [00:10<00:00, 84.05it/s]
+
 99%|█████████▉| 845/852 [00:10<00:00, 82.65it/s]
                                                  
+
                                                 
 10%|█         | 488/4880 [02:19<17:13,  4.25it/s]
+
+
                                                 [INFO|trainer.py:3503] 2024-09-09 20:34:31,586 >> Saving model checkpoint to /content/dissertation/scripts/ner/output/checkpoint-488
+[INFO|configuration_utils.py:472] 2024-09-09 20:34:31,588 >> Configuration saved in /content/dissertation/scripts/ner/output/checkpoint-488/config.json
+[INFO|modeling_utils.py:2799] 2024-09-09 20:34:32,614 >> Model weights saved in /content/dissertation/scripts/ner/output/checkpoint-488/model.safetensors
+[INFO|tokenization_utils_base.py:2684] 2024-09-09 20:34:32,615 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/checkpoint-488/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2693] 2024-09-09 20:34:32,615 >> Special tokens file saved in /content/dissertation/scripts/ner/output/checkpoint-488/special_tokens_map.json
+[INFO|tokenization_utils_base.py:2684] 2024-09-09 20:34:37,941 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2693] 2024-09-09 20:34:37,941 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+
 10%|█         | 489/4880 [02:26<7:47:21,  6.39s/it]
 10%|█         | 490/4880 [02:26<5:36:26,  4.60s/it]
 10%|█         | 491/4880 [02:26<4:03:41,  3.33s/it]
 10%|█         | 492/4880 [02:27<2:55:16,  2.40s/it]
 10%|█         | 493/4880 [02:27<2:08:23,  1.76s/it]
 10%|█         | 494/4880 [02:27<1:34:36,  1.29s/it]
 10%|█         | 495/4880 [02:27<1:10:33,  1.04it/s]
 10%|█         | 496/4880 [02:28<1:00:59,  1.20it/s]
 10%|█         | 497/4880 [02:28<47:22,  1.54it/s]  
 10%|█         | 498/4880 [02:28<38:06,  1.92it/s]
 10%|█         | 499/4880 [02:28<32:59,  2.21it/s]
 10%|█         | 500/4880 [02:29<29:11,  2.50it/s]
                                                  
 10%|█         | 500/4880 [02:29<29:11,  2.50it/s]
 10%|█         | 501/4880 [02:29<28:38,  2.55it/s]
 10%|█         | 502/4880 [02:29<24:03,  3.03it/s]
 10%|█         | 503/4880 [02:30<22:13,  3.28it/s]
 10%|█         | 504/4880 [02:30<21:04,  3.46it/s]
 10%|█         | 505/4880 [02:30<19:33,  3.73it/s]
 10%|█         | 506/4880 [02:30<19:55,  3.66it/s]
 10%|█         | 507/4880 [02:30<17:58,  4.06it/s]
 10%|█         | 508/4880 [02:31<17:15,  4.22it/s]
 10%|█         | 509/4880 [02:31<16:30,  4.41it/s]
 10%|█         | 510/4880 [02:31<16:42,  4.36it/s]
 10%|█         | 511/4880 [02:31<17:07,  4.25it/s]
 10%|█         | 512/4880 [02:32<16:53,  4.31it/s]
 11%|█         | 513/4880 [02:32<17:01,  4.28it/s]
 11%|█         | 514/4880 [02:32<16:01,  4.54it/s]
 11%|█         | 515/4880 [02:32<17:45,  4.10it/s]
 11%|█         | 516/4880 [02:33<17:49,  4.08it/s]
 11%|█         | 517/4880 [02:33<17:33,  4.14it/s]
 11%|█         | 518/4880 [02:33<16:13,  4.48it/s]
 11%|█         | 519/4880 [02:33<16:38,  4.37it/s]
 11%|█         | 520/4880 [02:33<16:44,  4.34it/s]
 11%|█         | 521/4880 [02:34<16:44,  4.34it/s]
 11%|█         | 522/4880 [02:34<16:13,  4.48it/s]
 11%|█         | 523/4880 [02:34<15:30,  4.68it/s]
 11%|█         | 524/4880 [02:34<14:40,  4.95it/s]
 11%|█         | 525/4880 [02:35<19:15,  3.77it/s]
 11%|█         | 526/4880 [02:35<20:08,  3.60it/s]
 11%|█         | 527/4880 [02:35<19:06,  3.80it/s]
 11%|█         | 528/4880 [02:35<17:44,  4.09it/s]
 11%|█         | 529/4880 [02:36<16:12,  4.48it/s]
 11%|█         | 530/4880 [02:36<17:48,  4.07it/s]
 11%|█         | 531/4880 [02:36<16:58,  4.27it/s]
 11%|█         | 532/4880 [02:36<17:30,  4.14it/s]
 11%|█         | 533/4880 [02:37<16:54,  4.29it/s]
 11%|█         | 534/4880 [02:37<16:35,  4.37it/s]
 11%|█         | 535/4880 [02:37<16:47,  4.31it/s]
 11%|█         | 536/4880 [02:37<16:16,  4.45it/s]
 11%|█         | 537/4880 [02:37<16:33,  4.37it/s]
 11%|█         | 538/4880 [02:38<19:00,  3.81it/s]
 11%|█         | 539/4880 [02:38<17:58,  4.02it/s]
 11%|█         | 540/4880 [02:38<16:09,  4.48it/s]
 11%|█         | 541/4880 [02:39<17:51,  4.05it/s]
 11%|█         | 542/4880 [02:39<17:42,  4.08it/s]
 11%|█         | 543/4880 [02:39<17:00,  4.25it/s]
 11%|█         | 544/4880 [02:39<20:08,  3.59it/s]
 11%|█         | 545/4880 [02:40<20:14,  3.57it/s]
 11%|█         | 546/4880 [02:40<20:19,  3.55it/s]
 11%|█         | 547/4880 [02:40<19:18,  3.74it/s]
 11%|█         | 548/4880 [02:40<18:11,  3.97it/s]
 11%|█▏        | 549/4880 [02:41<17:21,  4.16it/s]
 11%|█▏        | 550/4880 [02:41<16:37,  4.34it/s]
 11%|█▏        | 551/4880 [02:41<16:13,  4.44it/s]
 11%|█▏        | 552/4880 [02:41<14:50,  4.86it/s]
 11%|█▏        | 553/4880 [02:41<15:00,  4.81it/s]
 11%|█▏        | 554/4880 [02:42<16:26,  4.38it/s]
 11%|█▏        | 555/4880 [02:42<16:19,  4.41it/s]
 11%|█▏        | 556/4880 [02:42<15:54,  4.53it/s]
 11%|█▏        | 557/4880 [02:42<16:46,  4.29it/s]
 11%|█▏        | 558/4880 [02:43<17:45,  4.05it/s]
 11%|█▏        | 559/4880 [02:43<16:21,  4.40it/s]
 11%|█▏        | 560/4880 [02:43<15:21,  4.69it/s]
 11%|█▏        | 561/4880 [02:43<14:52,  4.84it/s]
 12%|█▏        | 562/4880 [02:43<15:23,  4.67it/s]
 12%|█▏        | 563/4880 [02:44<14:32,  4.95it/s]
 12%|█▏        | 564/4880 [02:44<14:45,  4.87it/s]
 12%|█▏        | 565/4880 [02:44<15:10,  4.74it/s]
 12%|█▏        | 566/4880 [02:44<16:14,  4.43it/s]
 12%|█▏        | 567/4880 [02:45<16:32,  4.35it/s]
 12%|█▏        | 568/4880 [02:45<17:01,  4.22it/s]
 12%|█▏        | 569/4880 [02:45<16:14,  4.42it/s]
 12%|█▏        | 570/4880 [02:45<16:17,  4.41it/s]
 12%|█▏        | 571/4880 [02:45<16:50,  4.27it/s]
 12%|█▏        | 572/4880 [02:46<17:16,  4.15it/s]
 12%|█▏        | 573/4880 [02:46<18:22,  3.91it/s]
 12%|█▏        | 574/4880 [02:46<19:40,  3.65it/s]
 12%|█▏        | 575/4880 [02:47<20:31,  3.50it/s]
 12%|█▏        | 576/4880 [02:47<27:07,  2.65it/s]
 12%|█▏        | 577/4880 [02:47<23:46,  3.02it/s]
 12%|█▏        | 578/4880 [02:48<20:54,  3.43it/s]
 12%|█▏        | 579/4880 [02:48<18:55,  3.79it/s]
 12%|█▏        | 580/4880 [02:48<16:38,  4.31it/s]
 12%|█▏        | 581/4880 [02:48<16:51,  4.25it/s]
 12%|█▏        | 582/4880 [02:49<17:48,  4.02it/s]
 12%|█▏        | 583/4880 [02:49<16:15,  4.41it/s]
 12%|█▏        | 584/4880 [02:49<15:56,  4.49it/s]
 12%|█▏        | 585/4880 [02:49<14:34,  4.91it/s]
 12%|█▏        | 586/4880 [02:49<15:24,  4.65it/s]
 12%|█▏        | 587/4880 [02:49<14:50,  4.82it/s]
 12%|█▏        | 588/4880 [02:50<15:37,  4.58it/s]
 12%|█▏        | 589/4880 [02:50<17:36,  4.06it/s]
 12%|█▏        | 590/4880 [02:50<20:27,  3.50it/s]
 12%|█▏        | 591/4880 [02:51<19:07,  3.74it/s]
 12%|█▏        | 592/4880 [02:51<19:00,  3.76it/s]
 12%|█▏        | 593/4880 [02:51<17:40,  4.04it/s]
 12%|█▏        | 594/4880 [02:51<16:42,  4.28it/s]
 12%|█▏        | 595/4880 [02:52<16:38,  4.29it/s]
 12%|█▏        | 596/4880 [02:52<17:36,  4.05it/s]
 12%|█▏        | 597/4880 [02:52<16:15,  4.39it/s]
 12%|█▏        | 598/4880 [02:52<20:08,  3.54it/s]
 12%|█▏        | 599/4880 [02:53<18:32,  3.85it/s]
 12%|█▏        | 600/4880 [02:53<20:22,  3.50it/s]
 12%|█▏        | 601/4880 [02:54<25:31,  2.79it/s]
 12%|█▏        | 602/4880 [02:54<22:15,  3.20it/s]
 12%|█▏        | 603/4880 [02:54<23:11,  3.07it/s]
 12%|█▏        | 604/4880 [02:54<21:13,  3.36it/s]
 12%|█▏        | 605/4880 [02:54<18:30,  3.85it/s]
 12%|█▏        | 606/4880 [02:55<16:50,  4.23it/s]
 12%|█▏        | 607/4880 [02:55<15:30,  4.59it/s]
 12%|█▏        | 608/4880 [02:55<15:26,  4.61it/s]
 12%|█▏        | 609/4880 [02:55<15:21,  4.63it/s]
 12%|█▎        | 610/4880 [02:56<16:03,  4.43it/s]
 13%|█▎        | 611/4880 [02:56<16:35,  4.29it/s]
 13%|█▎        | 612/4880 [02:56<15:54,  4.47it/s]
 13%|█▎        | 613/4880 [02:56<15:12,  4.68it/s]
 13%|█▎        | 614/4880 [02:56<15:18,  4.64it/s]
 13%|█▎        | 615/4880 [02:57<17:12,  4.13it/s]
 13%|█▎        | 616/4880 [02:57<17:14,  4.12it/s]
 13%|█▎        | 617/4880 [02:57<19:42,  3.61it/s]
 13%|█▎        | 618/4880 [02:57<18:16,  3.89it/s]

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.99288256227758,
+    "total_flos": 2.130317416831723e+16,
+    "train_loss": 0.017706579814779112,
+    "train_runtime": 1901.5665,
+    "train_samples": 44938,
+    "train_samples_per_second": 236.321,
+    "train_steps_per_second": 3.692
+}

trainer_state.json

@@ -0,0 +1,260 @@
+{
+  "best_metric": 0.8053395240858967,
+  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-7020",
+  "epoch": 9.99288256227758,
+  "eval_steps": 500,
+  "global_step": 7020,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.7117437722419929,
+      "grad_norm": 0.739061713218689,
+      "learning_rate": 4.643874643874644e-05,
+      "loss": 0.1165,
+      "step": 500
+    },
+    {
+      "epoch": 0.999288256227758,
+      "eval_accuracy": 0.9735468988149999,
+      "eval_f1": 0.7736125182933693,
+      "eval_loss": 0.08085697889328003,
+      "eval_precision": 0.7454979388153613,
+      "eval_recall": 0.8039307440336921,
+      "eval_runtime": 14.5471,
+      "eval_samples_per_second": 468.136,
+      "eval_steps_per_second": 58.569,
+      "step": 702
+    },
+    {
+      "epoch": 1.4234875444839858,
+      "grad_norm": 0.5600239038467407,
+      "learning_rate": 4.287749287749288e-05,
+      "loss": 0.0461,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9746797204333557,
+      "eval_f1": 0.7832803271240344,
+      "eval_loss": 0.09561743587255478,
+      "eval_precision": 0.7611479028697572,
+      "eval_recall": 0.8067384183434722,
+      "eval_runtime": 14.2524,
+      "eval_samples_per_second": 477.815,
+      "eval_steps_per_second": 59.779,
+      "step": 1405
+    },
+    {
+      "epoch": 2.135231316725979,
+      "grad_norm": 0.4771190881729126,
+      "learning_rate": 3.931623931623932e-05,
+      "loss": 0.0273,
+      "step": 1500
+    },
+    {
+      "epoch": 2.8469750889679717,
+      "grad_norm": 0.29212868213653564,
+      "learning_rate": 3.575498575498576e-05,
+      "loss": 0.0165,
+      "step": 2000
+    },
+    {
+      "epoch": 2.999288256227758,
+      "eval_accuracy": 0.974439424938553,
+      "eval_f1": 0.7853282741175118,
+      "eval_loss": 0.10571284592151642,
+      "eval_precision": 0.7721003843545106,
+      "eval_recall": 0.7990173139915769,
+      "eval_runtime": 14.2529,
+      "eval_samples_per_second": 477.797,
+      "eval_steps_per_second": 59.777,
+      "step": 2107
+    },
+    {
+      "epoch": 3.5587188612099645,
+      "grad_norm": 0.08157803863286972,
+      "learning_rate": 3.2193732193732194e-05,
+      "loss": 0.011,
+      "step": 2500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9750504620539086,
+      "eval_f1": 0.7971327796108773,
+      "eval_loss": 0.12744221091270447,
+      "eval_precision": 0.775858250276855,
+      "eval_recall": 0.8196069255966308,
+      "eval_runtime": 14.6113,
+      "eval_samples_per_second": 466.079,
+      "eval_steps_per_second": 58.311,
+      "step": 2810
+    },
+    {
+      "epoch": 4.270462633451958,
+      "grad_norm": 0.45329615473747253,
+      "learning_rate": 2.863247863247863e-05,
+      "loss": 0.0079,
+      "step": 3000
+    },
+    {
+      "epoch": 4.98220640569395,
+      "grad_norm": 0.4696062207221985,
+      "learning_rate": 2.5071225071225073e-05,
+      "loss": 0.006,
+      "step": 3500
+    },
+    {
+      "epoch": 4.999288256227758,
+      "eval_accuracy": 0.9744943496230793,
+      "eval_f1": 0.7975886853698122,
+      "eval_loss": 0.13583102822303772,
+      "eval_precision": 0.7904411764705882,
+      "eval_recall": 0.8048666354702855,
+      "eval_runtime": 14.3515,
+      "eval_samples_per_second": 474.516,
+      "eval_steps_per_second": 59.367,
+      "step": 3512
+    },
+    {
+      "epoch": 5.693950177935943,
+      "grad_norm": 0.6243860721588135,
+      "learning_rate": 2.150997150997151e-05,
+      "loss": 0.0045,
+      "step": 4000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9746179301632636,
+      "eval_f1": 0.7948299953423382,
+      "eval_loss": 0.14197705686092377,
+      "eval_precision": 0.7911451089476125,
+      "eval_recall": 0.7985493682732803,
+      "eval_runtime": 14.3392,
+      "eval_samples_per_second": 474.921,
+      "eval_steps_per_second": 59.417,
+      "step": 4215
+    },
+    {
+      "epoch": 6.405693950177936,
+      "grad_norm": 0.2963380813598633,
+      "learning_rate": 1.794871794871795e-05,
+      "loss": 0.0037,
+      "step": 4500
+    },
+    {
+      "epoch": 6.999288256227758,
+      "eval_accuracy": 0.9748925535858953,
+      "eval_f1": 0.796227293898463,
+      "eval_loss": 0.160146102309227,
+      "eval_precision": 0.7925359295317571,
+      "eval_recall": 0.7999532054281703,
+      "eval_runtime": 14.175,
+      "eval_samples_per_second": 480.423,
+      "eval_steps_per_second": 60.106,
+      "step": 4917
+    },
+    {
+      "epoch": 7.117437722419929,
+      "grad_norm": 0.4029097557067871,
+      "learning_rate": 1.4387464387464389e-05,
+      "loss": 0.0028,
+      "step": 5000
+    },
+    {
+      "epoch": 7.829181494661921,
+      "grad_norm": 0.04551521688699722,
+      "learning_rate": 1.0826210826210826e-05,
+      "loss": 0.0022,
+      "step": 5500
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9758331388084089,
+      "eval_f1": 0.8050679995350458,
+      "eval_loss": 0.1620824635028839,
+      "eval_precision": 0.7999537999538,
+      "eval_recall": 0.8102480112306972,
+      "eval_runtime": 14.4773,
+      "eval_samples_per_second": 470.391,
+      "eval_steps_per_second": 58.851,
+      "step": 5620
+    },
+    {
+      "epoch": 8.540925266903916,
+      "grad_norm": 0.3445014953613281,
+      "learning_rate": 7.264957264957266e-06,
+      "loss": 0.0016,
+      "step": 6000
+    },
+    {
+      "epoch": 8.999288256227757,
+      "eval_accuracy": 0.9757713485383168,
+      "eval_f1": 0.8028807062376583,
+      "eval_loss": 0.16807720065116882,
+      "eval_precision": 0.7972318339100346,
+      "eval_recall": 0.8086102012166588,
+      "eval_runtime": 14.1973,
+      "eval_samples_per_second": 479.67,
+      "eval_steps_per_second": 60.012,
+      "step": 6322
+    },
+    {
+      "epoch": 9.252669039145907,
+      "grad_norm": 0.12732785940170288,
+      "learning_rate": 3.7037037037037037e-06,
+      "loss": 0.0013,
+      "step": 6500
+    },
+    {
+      "epoch": 9.9644128113879,
+      "grad_norm": 0.05971187725663185,
+      "learning_rate": 1.4245014245014247e-07,
+      "loss": 0.0013,
+      "step": 7000
+    },
+    {
+      "epoch": 9.99288256227758,
+      "eval_accuracy": 0.9758743323218038,
+      "eval_f1": 0.8053395240858967,
+      "eval_loss": 0.17113561928272247,
+      "eval_precision": 0.7991246256622898,
+      "eval_recall": 0.8116518483855872,
+      "eval_runtime": 14.5918,
+      "eval_samples_per_second": 466.7,
+      "eval_steps_per_second": 58.389,
+      "step": 7020
+    },
+    {
+      "epoch": 9.99288256227758,
+      "step": 7020,
+      "total_flos": 2.130317416831723e+16,
+      "train_loss": 0.017706579814779112,
+      "train_runtime": 1901.5665,
+      "train_samples_per_second": 236.321,
+      "train_steps_per_second": 3.692
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7020,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.130317416831723e+16,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13556e6c97b2f39e25d5830ab0bc61ce81f807bcf643d150d23dd97c2f606c57
+size 5240