End of training

README.md

@@ -3,9 +3,10 @@ library_name: transformers
 license: apache-2.0
 base_model: PlanTL-GOB-ES/bsc-bio-ehr-es
 tags:
+- token-classification
 - generated_from_trainer
 datasets:
-- combined-train-drugtemist-dev-85-ner
+- Rodrigo1771/combined-train-drugtemist-dev-85-ner
 metrics:
 - precision
 - recall
@@ -18,24 +19,24 @@ model-index:
       name: Token Classification
       type: token-classification
     dataset:
-      name: combined-train-drugtemist-dev-85-ner
-      type: combined-train-drugtemist-dev-85-ner
+      name: Rodrigo1771/combined-train-drugtemist-dev-85-ner
+      type: Rodrigo1771/combined-train-drugtemist-dev-85-ner
       config: CombinedTrainDrugTEMISTDevNER
       split: validation
       args: CombinedTrainDrugTEMISTDevNER
     metrics:
     - name: Precision
       type: precision
-      value: 0.09223170184104176
+      value: 0.09400470929179497
     - name: Recall
       type: recall
-      value: 0.9439338235294118
+      value: 0.9540441176470589
     - name: F1
       type: f1
-      value: 0.16804385175488834
+      value: 0.17114591920857378
     - name: Accuracy
       type: accuracy
-      value: 0.7862743213368668
+      value: 0.7890274211487498
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -43,13 +44,13 @@ 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 combined-train-drugtemist-dev-85-ner dataset.
+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/combined-train-drugtemist-dev-85-ner dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.6103
-- Precision: 0.0922
-- Recall: 0.9439
-- F1: 0.1680
-- Accuracy: 0.7863
+- Loss: 1.1806
+- Precision: 0.0940
+- Recall: 0.9540
+- F1: 0.1711
+- Accuracy: 0.7890
 
 ## Model description
 

all_results.json

@@ -1,26 +1,26 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.8564886649182308,
-    "eval_f1": 0.45455732567249935,
-    "eval_loss": 0.7005925178527832,
-    "eval_precision": 0.3152508603513856,
-    "eval_recall": 0.8144595226953674,
-    "eval_runtime": 14.2934,
+    "eval_accuracy": 0.7890274211487498,
+    "eval_f1": 0.17114591920857378,
+    "eval_loss": 1.1805996894836426,
+    "eval_precision": 0.09400470929179497,
+    "eval_recall": 0.9540441176470589,
+    "eval_runtime": 14.5104,
     "eval_samples": 6810,
-    "eval_samples_per_second": 476.445,
-    "eval_steps_per_second": 59.608,
-    "predict_accuracy": 0.9437946603149774,
-    "predict_f1": 0.6365870441364396,
-    "predict_loss": 0.22766011953353882,
-    "predict_precision": 0.519576379974326,
-    "predict_recall": 0.8216188784572444,
-    "predict_runtime": 29.1056,
-    "predict_samples_per_second": 502.103,
-    "predict_steps_per_second": 62.771,
+    "eval_samples_per_second": 469.318,
+    "eval_steps_per_second": 58.716,
+    "predict_accuracy": 0.8712439909780733,
+    "predict_f1": 0.22974402369367464,
+    "predict_loss": 0.7416015267372131,
+    "predict_precision": 0.130696405648267,
+    "predict_recall": 0.9487478159580663,
+    "predict_runtime": 29.7849,
+    "predict_samples_per_second": 490.651,
+    "predict_steps_per_second": 61.34,
     "total_flos": 1.7176580067661056e+16,
     "train_loss": 0.0812657987344287,
-    "train_runtime": 1549.44,
+    "train_runtime": 1542.5562,
     "train_samples": 34604,
-    "train_samples_per_second": 223.332,
-    "train_steps_per_second": 3.492
+    "train_samples_per_second": 224.329,
+    "train_steps_per_second": 3.507
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.8564886649182308,
-    "eval_f1": 0.45455732567249935,
-    "eval_loss": 0.7005925178527832,
-    "eval_precision": 0.3152508603513856,
-    "eval_recall": 0.8144595226953674,
-    "eval_runtime": 14.2934,
+    "eval_accuracy": 0.7890274211487498,
+    "eval_f1": 0.17114591920857378,
+    "eval_loss": 1.1805996894836426,
+    "eval_precision": 0.09400470929179497,
+    "eval_recall": 0.9540441176470589,
+    "eval_runtime": 14.5104,
     "eval_samples": 6810,
-    "eval_samples_per_second": 476.445,
-    "eval_steps_per_second": 59.608
+    "eval_samples_per_second": 469.318,
+    "eval_steps_per_second": 58.716
 }

predict_results.json

@@ -1,10 +1,10 @@
 {
-    "predict_accuracy": 0.9437946603149774,
-    "predict_f1": 0.6365870441364396,
-    "predict_loss": 0.22766011953353882,
-    "predict_precision": 0.519576379974326,
-    "predict_recall": 0.8216188784572444,
-    "predict_runtime": 29.1056,
-    "predict_samples_per_second": 502.103,
-    "predict_steps_per_second": 62.771
+    "predict_accuracy": 0.8712439909780733,
+    "predict_f1": 0.22974402369367464,
+    "predict_loss": 0.7416015267372131,
+    "predict_precision": 0.130696405648267,
+    "predict_recall": 0.9487478159580663,
+    "predict_runtime": 29.7849,
+    "predict_samples_per_second": 490.651,
+    "predict_steps_per_second": 61.34
 }

tb/events.out.tfevents.1725582923.2a66098fac87.23851.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec2dc4fe260f4b1afdfb552850a0337582e07d058a8812069b094e4c853b8cac
+size 560

train.log

@@ -1501,3 +1501,53 @@ Training completed. Do not forget to share your model on huggingface.co/models =
 {'eval_loss': 1.6103088855743408, 'eval_precision': 0.09223170184104176, 'eval_recall': 0.9439338235294118, 'eval_f1': 0.16804385175488834, 'eval_accuracy': 0.7862743213368668, 'eval_runtime': 14.5612, 'eval_samples_per_second': 467.681, 'eval_steps_per_second': 58.512, 'epoch': 10.0}
 {'train_runtime': 1542.5562, 'train_samples_per_second': 224.329, 'train_steps_per_second': 3.507, 'train_loss': 0.0812657987344287, 'epoch': 10.0}
 
+***** train metrics *****
+  epoch                    =       10.0
+  total_flos               = 15996936GF
+  train_loss               =     0.0813
+  train_runtime            = 0:25:42.55
+  train_samples            =      34604
+  train_samples_per_second =    224.329
+  train_steps_per_second   =      3.507
+09/06/2024 00:35:09 - INFO - __main__ -   *** Evaluate ***
+[INFO|trainer.py:811] 2024-09-06 00:35:09,352 >> The following columns in the evaluation set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: tokens, ner_tags, id. If tokens, ner_tags, id are not expected by `RobertaForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3819] 2024-09-06 00:35:09,354 >> 
+***** Running Evaluation *****
+[INFO|trainer.py:3821] 2024-09-06 00:35:09,354 >>   Num examples = 6810
+[INFO|trainer.py:3824] 2024-09-06 00:35:09,354 >>   Batch size = 8
+
  0%|          | 0/852 [00:00<?, ?it/s]
  1%|          | 10/852 [00:00<00:09, 89.32it/s]
  2%|▏         | 19/852 [00:00<00:10, 77.68it/s]
  3%|▎         | 27/852 [00:00<00:10, 77.19it/s]
  4%|▍         | 35/852 [00:00<00:10, 77.86it/s]
  5%|▌         | 44/852 [00:00<00:10, 79.22it/s]
  6%|▌         | 53/852 [00:00<00:09, 80.49it/s]
  7%|▋         | 62/852 [00:00<00:09, 81.06it/s]
  8%|▊         | 71/852 [00:00<00:09, 78.26it/s]
  9%|▉         | 80/852 [00:01<00:09, 78.24it/s]
 10%|█         | 89/852 [00:01<00:09, 78.87it/s]
 11%|█▏        | 97/852 [00:01<00:09, 78.68it/s]
 12%|█▏        | 106/852 [00:01<00:09, 79.83it/s]
 13%|█▎        | 114/852 [00:01<00:09, 79.66it/s]
 14%|█▍        | 123/852 [00:01<00:09, 80.84it/s]
 15%|█▌        | 132/852 [00:01<00:09, 76.65it/s]
 16%|█▋        | 140/852 [00:01<00:09, 76.74it/s]
 17%|█▋        | 148/852 [00:01<00:09, 76.37it/s]
 18%|█▊        | 156/852 [00:01<00:09, 77.14it/s]
 19%|█▉        | 164/852 [00:02<00:08, 77.82it/s]
 20%|██        | 172/852 [00:02<00:08, 77.95it/s]
 21%|██        | 180/852 [00:02<00:08, 76.88it/s]
 22%|██▏       | 188/852 [00:02<00:08, 77.69it/s]
 23%|██▎       | 196/852 [00:02<00:08, 77.62it/s]
 24%|██▍       | 205/852 [00:02<00:08, 78.72it/s]
 25%|██▌       | 213/852 [00:02<00:08, 77.59it/s]
 26%|██▌       | 221/852 [00:02<00:08, 76.68it/s]
 27%|██▋       | 230/852 [00:02<00:07, 77.89it/s]
 28%|██▊       | 238/852 [00:03<00:07, 77.24it/s]
 29%|██▉       | 246/852 [00:03<00:07, 75.97it/s]
 30%|██▉       | 255/852 [00:03<00:07, 78.31it/s]
 31%|███       | 264/852 [00:03<00:07, 79.83it/s]
 32%|███▏      | 272/852 [00:03<00:07, 79.08it/s]
 33%|███▎      | 281/852 [00:03<00:07, 80.99it/s]
 34%|███▍      | 290/852 [00:03<00:07, 79.72it/s]
 35%|███▍      | 298/852 [00:03<00:06, 79.60it/s]
 36%|███▌      | 307/852 [00:03<00:06, 80.29it/s]
 37%|███▋      | 316/852 [00:04<00:06, 78.10it/s]
 38%|███▊      | 324/852 [00:04<00:06, 78.24it/s]
 39%|███▉      | 332/852 [00:04<00:06, 77.89it/s]
 40%|████      | 341/852 [00:04<00:06, 79.08it/s]
 41%|████      | 350/852 [00:04<00:06, 79.99it/s]
 42%|████▏     | 359/852 [00:04<00:06, 78.18it/s]
 43%|████▎     | 368/852 [00:04<00:06, 79.04it/s]
 44%|████▍     | 377/852 [00:04<00:05, 79.49it/s]
 45%|████▌     | 385/852 [00:04<00:05, 78.76it/s]
 46%|████▌     | 394/852 [00:05<00:05, 79.43it/s]
 47%|████▋     | 402/852 [00:05<00:05, 79.23it/s]
 48%|████▊     | 410/852 [00:05<00:05, 77.00it/s]
 49%|████▉     | 418/852 [00:05<00:05, 77.80it/s]
 50%|█████     | 426/852 [00:05<00:05, 76.10it/s]
 51%|█████     | 435/852 [00:05<00:05, 77.97it/s]
 52%|█████▏    | 444/852 [00:05<00:05, 79.13it/s]
 53%|█████▎    | 453/852 [00:05<00:04, 79.97it/s]
 54%|█████▍    | 461/852 [00:05<00:05, 77.94it/s]
 55%|█████▌    | 469/852 [00:05<00:04, 77.33it/s]
 56%|█████▌    | 477/852 [00:06<00:05, 74.29it/s]
 57%|█████▋    | 485/852 [00:06<00:04, 74.32it/s]
 58%|█████▊    | 494/852 [00:06<00:04, 76.20it/s]
 59%|█████▉    | 503/852 [00:06<00:04, 77.78it/s]
 60%|█████▉    | 511/852 [00:06<00:04, 77.92it/s]
 61%|██████    | 520/852 [00:06<00:04, 79.03it/s]
 62%|██████▏   | 528/852 [00:06<00:04, 77.15it/s]
 63%|██████▎   | 537/852 [00:06<00:03, 79.37it/s]
 64%|██████▍   | 546/852 [00:06<00:03, 80.04it/s]
 65%|██████▌   | 555/852 [00:07<00:03, 77.62it/s]
 66%|██████▌   | 564/852 [00:07<00:03, 78.59it/s]
 67%|██████▋   | 572/852 [00:07<00:03, 78.53it/s]
 68%|██████▊   | 580/852 [00:07<00:03, 78.15it/s]
 69%|██████▉   | 588/852 [00:07<00:03, 77.41it/s]
 70%|███████   | 597/852 [00:07<00:03, 78.12it/s]
 71%|███████   | 605/852 [00:07<00:03, 73.52it/s]
 72%|███████▏  | 613/852 [00:07<00:03, 74.71it/s]
 73%|███████▎  | 621/852 [00:07<00:03, 75.59it/s]
 74%|███████▍  | 629/852 [00:08<00:02, 74.68it/s]
 75%|███████▍  | 638/852 [00:08<00:02, 76.62it/s]
 76%|███████▌  | 646/852 [00:08<00:02, 74.33it/s]
 77%|███████▋  | 655/852 [00:08<00:02, 77.01it/s]
 78%|███████▊  | 663/852 [00:08<00:02, 77.50it/s]
 79%|███████▉  | 671/852 [00:08<00:02, 77.95it/s]
 80%|███████▉  | 680/852 [00:08<00:02, 78.57it/s]
 81%|████████  | 689/852 [00:08<00:02, 80.07it/s]
 82%|████████▏ | 698/852 [00:08<00:01, 80.51it/s]
 83%|████████▎ | 707/852 [00:09<00:01, 81.55it/s]
 84%|████████▍ | 716/852 [00:09<00:01, 81.02it/s]
 85%|████████▌ | 725/852 [00:09<00:01, 80.67it/s]
 86%|████████▌ | 734/852 [00:09<00:01, 81.53it/s]
 87%|████████▋ | 743/852 [00:09<00:01, 81.13it/s]
 88%|████████▊ | 752/852 [00:09<00:01, 80.94it/s]
 89%|████████▉ | 761/852 [00:09<00:01, 82.18it/s]
 90%|█████████ | 770/852 [00:09<00:01, 80.71it/s]
 91%|█████████▏| 779/852 [00:09<00:00, 80.08it/s]
 92%|█████████▏| 788/852 [00:10<00:00, 79.00it/s]
 93%|█████████▎| 796/852 [00:10<00:00, 79.25it/s]
 94%|█████████▍| 805/852 [00:10<00:00, 79.61it/s]
 95%|█████████▌| 813/852 [00:10<00:00, 77.98it/s]
 96%|█████████▋| 822/852 [00:10<00:00, 79.28it/s]
 98%|█████████▊| 831/852 [00:10<00:00, 80.44it/s]
 99%|█████████▊| 840/852 [00:10<00:00, 79.45it/s]
+  _warn_prf(average, modifier, msg_start, len(result))
+
+***** eval metrics *****
+  epoch                   =       10.0
+  eval_accuracy           =      0.789
+  eval_f1                 =     0.1711
+  eval_loss               =     1.1806
+  eval_precision          =      0.094
+  eval_recall             =      0.954
+  eval_runtime            = 0:00:14.51
+  eval_samples            =       6810
+  eval_samples_per_second =    469.318
+  eval_steps_per_second   =     58.716
+09/06/2024 00:35:23 - INFO - __main__ -   *** Predict ***
+[INFO|trainer.py:811] 2024-09-06 00:35:23,872 >> The following columns in the test set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: tokens, ner_tags, id. If tokens, ner_tags, id are not expected by `RobertaForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3819] 2024-09-06 00:35:23,875 >> 
+***** Running Prediction *****
+[INFO|trainer.py:3821] 2024-09-06 00:35:23,875 >>   Num examples = 14614
+[INFO|trainer.py:3824] 2024-09-06 00:35:23,875 >>   Batch size = 8
+
  0%|          | 0/1827 [00:00<?, ?it/s]
  1%|          | 10/1827 [00:00<00:21, 86.39it/s]
  1%|          | 19/1827 [00:00<00:23, 76.76it/s]
  1%|▏         | 27/1827 [00:00<00:23, 77.35it/s]
  2%|▏         | 35/1827 [00:00<00:23, 76.72it/s]
  2%|▏         | 44/1827 [00:00<00:22, 78.12it/s]
  3%|▎         | 52/1827 [00:00<00:22, 77.86it/s]
  3%|▎         | 60/1827 [00:00<00:23, 75.65it/s]
  4%|▍         | 69/1827 [00:00<00:22, 77.33it/s]
  4%|▍         | 78/1827 [00:00<00:22, 79.22it/s]
  5%|▍         | 86/1827 [00:01<00:22, 77.72it/s]
  5%|▌         | 95/1827 [00:01<00:21, 78.74it/s]
  6%|▌         | 104/1827 [00:01<00:21, 80.10it/s]
  6%|▌         | 113/1827 [00:01<00:21, 80.17it/s]
  7%|▋         | 122/1827 [00:01<00:21, 78.09it/s]
  7%|▋         | 130/1827 [00:01<00:21, 77.56it/s]
  8%|▊         | 139/1827 [00:01<00:21, 79.53it/s]
  8%|▊         | 148/1827 [00:01<00:20, 80.02it/s]
  9%|▊         | 157/1827 [00:02<00:21, 77.61it/s]
  9%|▉         | 166/1827 [00:02<00:21, 78.42it/s]
 10%|▉         | 174/1827 [00:02<00:21, 78.59it/s]
 10%|█         | 183/1827 [00:02<00:20, 80.08it/s]
 11%|█         | 192/1827 [00:02<00:20, 80.24it/s]
 11%|█         | 201/1827 [00:02<00:20, 78.80it/s]
 11%|█▏        | 210/1827 [00:02<00:20, 79.79it/s]
 12%|█▏        | 218/1827 [00:02<00:20, 79.07it/s]
 12%|█▏        | 227/1827 [00:02<00:19, 80.02it/s]
 13%|█▎        | 236/1827 [00:03<00:20, 77.94it/s]
 13%|█▎        | 245/1827 [00:03<00:19, 79.46it/s]
 14%|█▍        | 253/1827 [00:03<00:19, 79.43it/s]
 14%|█▍        | 261/1827 [00:03<00:19, 79.16it/s]
 15%|█▍        | 270/1827 [00:03<00:19, 80.39it/s]
 15%|█▌        | 279/1827 [00:03<00:18, 81.75it/s]
 16%|█▌        | 288/1827 [00:03<00:18, 81.38it/s]
 16%|█▋        | 297/1827 [00:03<00:18, 82.03it/s]
 17%|█▋        | 306/1827 [00:03<00:18, 80.70it/s]
 17%|█▋        | 315/1827 [00:03<00:18, 81.67it/s]
 18%|█▊        | 324/1827 [00:04<00:18, 81.89it/s]
 18%|█▊        | 333/1827 [00:04<00:18, 81.13it/s]
 19%|█▊        | 342/1827 [00:04<00:18, 82.33it/s]
 19%|█▉        | 351/1827 [00:04<00:18, 77.88it/s]
 20%|█▉        | 360/1827 [00:04<00:18, 79.12it/s]
 20%|██        | 369/1827 [00:04<00:18, 80.42it/s]
 21%|██        | 378/1827 [00:04<00:17, 81.01it/s]
 21%|██        | 387/1827 [00:04<00:17, 80.82it/s]
 22%|██▏       | 396/1827 [00:04<00:17, 80.20it/s]
 22%|██▏       | 405/1827 [00:05<00:17, 81.70it/s]
 23%|██▎       | 414/1827 [00:05<00:17, 79.66it/s]
 23%|██▎       | 423/1827 [00:05<00:17, 81.14it/s]
 24%|██▎       | 432/1827 [00:05<00:17, 79.99it/s]
 24%|██▍       | 441/1827 [00:05<00:17, 79.76it/s]
 25%|██▍       | 449/1827 [00:05<00:17, 79.62it/s]
 25%|██▌       | 458/1827 [00:05<00:17, 80.43it/s]
 26%|██▌       | 467/1827 [00:05<00:16, 81.29it/s]
 26%|██▌       | 476/1827 [00:05<00:17, 78.86it/s]
 26%|██▋       | 484/1827 [00:06<00:17, 78.63it/s]
 27%|██▋       | 492/1827 [00:06<00:17, 77.21it/s]
 27%|██▋       | 500/1827 [00:06<00:17, 77.67it/s]
 28%|██▊       | 508/1827 [00:06<00:16, 77.96it/s]
 28%|██▊       | 516/1827 [00:06<00:16, 78.33it/s]
 29%|██▊       | 525/1827 [00:06<00:16, 79.91it/s]
 29%|██▉       | 533/1827 [00:06<00:16, 78.68it/s]
 30%|██▉       | 542/1827 [00:06<00:16, 79.19it/s]
 30%|███       | 550/1827 [00:06<00:16, 78.85it/s]
 31%|███       | 558/1827 [00:07<00:16, 78.22it/s]
 31%|███       | 566/1827 [00:07<00:16, 78.72it/s]
 31%|███▏      | 575/1827 [00:07<00:15, 80.21it/s]
 32%|███▏      | 584/1827 [00:07<00:15, 78.95it/s]
 32%|███▏      | 593/1827 [00:07<00:15, 79.34it/s]
 33%|███▎      | 601/1827 [00:07<00:15, 79.41it/s]
 33%|███▎      | 609/1827 [00:07<00:15, 77.56it/s]
 34%|███▍      | 617/1827 [00:07<00:15, 77.74it/s]
 34%|███▍      | 626/1827 [00:07<00:15, 78.70it/s]
 35%|███▍      | 634/1827 [00:07<00:15, 78.98it/s]
 35%|███▌      | 643/1827 [00:08<00:14, 79.92it/s]
 36%|███▌      | 651/1827 [00:08<00:15, 75.14it/s]
 36%|███▌      | 659/1827 [00:08<00:15, 75.62it/s]
 37%|███▋      | 668/1827 [00:08<00:14, 77.65it/s]
 37%|███▋      | 676/1827 [00:08<00:14, 77.24it/s]
 37%|███▋      | 684/1827 [00:08<00:14, 77.46it/s]
 38%|███▊      | 693/1827 [00:08<00:14, 79.15it/s]
 38%|███▊      | 701/1827 [00:08<00:14, 78.98it/s]
 39%|███▉      | 710/1827 [00:08<00:13, 80.68it/s]
 39%|███▉      | 719/1827 [00:09<00:13, 82.04it/s]
 40%|███▉      | 728/1827 [00:09<00:13, 82.63it/s]
 40%|████      | 737/1827 [00:09<00:13, 82.07it/s]
 41%|████      | 746/1827 [00:09<00:13, 82.50it/s]
 41%|████▏     | 755/1827 [00:09<00:12, 83.02it/s]
 42%|████▏     | 764/1827 [00:09<00:12, 82.22it/s]
 42%|████▏     | 773/1827 [00:09<00:12, 81.23it/s]
 43%|████▎     | 782/1827 [00:09<00:13, 79.31it/s]
 43%|████▎     | 791/1827 [00:09<00:12, 80.00it/s]
 44%|████▍     | 800/1827 [00:10<00:12, 79.01it/s]
 44%|████▍     | 809/1827 [00:10<00:12, 80.15it/s]
 45%|████▍     | 818/1827 [00:10<00:12, 79.71it/s]
 45%|████▌     | 827/1827 [00:10<00:12, 80.88it/s]
 46%|████▌     | 836/1827 [00:10<00:12, 79.73it/s]
 46%|████▋     | 845/1827 [00:10<00:12, 80.84it/s]
 47%|████▋     | 854/1827 [00:10<00:11, 81.23it/s]
 47%|████▋     | 863/1827 [00:10<00:11, 81.73it/s]
 48%|████▊     | 872/1827 [00:10<00:11, 81.75it/s]
 48%|████▊     | 881/1827 [00:11<00:11, 81.62it/s]
 49%|████▊     | 890/1827 [00:11<00:11, 81.91it/s]
 49%|████▉     | 899/1827 [00:11<00:11, 81.83it/s]
 50%|████▉     | 908/1827 [00:11<00:11, 81.18it/s]
 50%|█████     | 917/1827 [00:11<00:11, 81.62it/s]
 51%|█████     | 926/1827 [00:11<00:11, 80.27it/s]
 51%|█████     | 935/1827 [00:11<00:11, 77.20it/s]
 52%|█████▏    | 943/1827 [00:11<00:11, 77.54it/s]
 52%|█████▏    | 951/1827 [00:11<00:11, 76.33it/s]
 52%|█████▏    | 959/1827 [00:12<00:11, 76.78it/s]
 53%|█████▎    | 967/1827 [00:12<00:11, 77.61it/s]
 53%|█████▎    | 975/1827 [00:12<00:11, 76.48it/s]
 54%|█████▍    | 983/1827 [00:12<00:11, 76.46it/s]
 54%|█████▍    | 991/1827 [00:12<00:10, 76.57it/s]
 55%|█████▍    | 999/1827 [00:12<00:10, 77.28it/s]
 55%|█████▌    | 1007/1827 [00:12<00:10, 76.62it/s]
 56%|█████▌    | 1015/1827 [00:12<00:10, 77.11it/s]
 56%|█████▌    | 1023/1827 [00:12<00:10, 77.82it/s]
 56%|█████▋    | 1032/1827 [00:13<00:10, 78.52it/s]
 57%|█████▋    | 1040/1827 [00:13<00:10, 77.02it/s]
 57%|█████▋    | 1048/1827 [00:13<00:10, 77.36it/s]
 58%|█████▊    | 1056/1827 [00:13<00:09, 77.94it/s]
 58%|█████▊    | 1065/1827 [00:13<00:09, 79.46it/s]
 59%|█████▉    | 1074/1827 [00:13<00:09, 80.78it/s]
 59%|█████▉    | 1083/1827 [00:13<00:09, 81.47it/s]
 60%|█████▉    | 1092/1827 [00:13<00:08, 81.99it/s]
 60%|██████    | 1101/1827 [00:13<00:08, 82.35it/s]
 61%|██████    | 1110/1827 [00:13<00:08, 81.93it/s]
 61%|██████    | 1119/1827 [00:14<00:08, 81.44it/s]
 62%|██████▏   | 1128/1827 [00:14<00:08, 81.89it/s]
 62%|██████▏   | 1137/1827 [00:14<00:08, 82.95it/s]
 63%|██████▎   | 1146/1827 [00:14<00:08, 83.37it/s]
 63%|██████▎   | 1155/1827 [00:14<00:08, 83.36it/s]
 64%|██████▎   | 1164/1827 [00:14<00:08, 80.93it/s]
 64%|██████▍   | 1173/1827 [00:14<00:08, 81.26it/s]
 65%|██████▍   | 1182/1827 [00:14<00:08, 79.25it/s]
 65%|██████▌   | 1191/1827 [00:14<00:07, 79.98it/s]
 66%|██████▌   | 1200/1827 [00:15<00:07, 80.51it/s]
 66%|██████▌   | 1209/1827 [00:15<00:07, 80.69it/s]
 67%|██████▋   | 1218/1827 [00:15<00:07, 79.18it/s]
 67%|██████▋   | 1227/1827 [00:15<00:07, 80.18it/s]
 68%|██████▊   | 1236/1827 [00:15<00:07, 80.99it/s]
 68%|██████▊   | 1245/1827 [00:15<00:07, 81.49it/s]
 69%|██████▊   | 1254/1827 [00:15<00:07, 78.96it/s]
 69%|██████▉   | 1262/1827 [00:15<00:07, 78.04it/s]
 70%|██████▉   | 1271/1827 [00:15<00:06, 79.53it/s]
 70%|███████   | 1280/1827 [00:16<00:06, 80.75it/s]
 71%|███████   | 1289/1827 [00:16<00:06, 81.53it/s]
 71%|███████   | 1298/1827 [00:16<00:06, 82.69it/s]
 72%|███████▏  | 1307/1827 [00:16<00:06, 82.39it/s]
 72%|███████▏  | 1316/1827 [00:16<00:06, 81.93it/s]
 73%|███████▎  | 1325/1827 [00:16<00:06, 79.75it/s]
 73%|███████▎  | 1334/1827 [00:16<00:06, 80.32it/s]
 74%|███████▎  | 1343/1827 [00:16<00:06, 79.81it/s]
 74%|███████▍  | 1351/1827 [00:16<00:05, 79.79it/s]
 74%|███████▍  | 1359/1827 [00:17<00:05, 79.73it/s]
 75%|███████▍  | 1368/1827 [00:17<00:05, 80.26it/s]
 75%|███████▌  | 1377/1827 [00:17<00:05, 79.57it/s]
 76%|███████▌  | 1386/1827 [00:17<00:05, 80.10it/s]
 76%|███████▋  | 1395/1827 [00:17<00:05, 79.37it/s]
 77%|███████▋  | 1404/1827 [00:17<00:05, 80.47it/s]
 77%|███████▋  | 1413/1827 [00:17<00:05, 80.08it/s]
 78%|███████▊  | 1422/1827 [00:17<00:05, 79.11it/s]
 78%|███████▊  | 1430/1827 [00:17<00:05, 77.31it/s]
 79%|███████▊  | 1438/1827 [00:18<00:05, 75.53it/s]
 79%|███████▉  | 1446/1827 [00:18<00:04, 76.28it/s]
 80%|███████▉  | 1455/1827 [00:18<00:04, 77.69it/s]
 80%|████████  | 1463/1827 [00:18<00:04, 77.56it/s]
 81%|████████  | 1471/1827 [00:18<00:04, 75.50it/s]
 81%|████████  | 1479/1827 [00:18<00:04, 75.70it/s]
 81%|████████▏ | 1487/1827 [00:18<00:04, 76.40it/s]
 82%|████████▏ | 1495/1827 [00:18<00:04, 74.59it/s]
 82%|████████▏ | 1503/1827 [00:18<00:04, 74.46it/s]
 83%|████████▎ | 1511/1827 [00:19<00:04, 76.02it/s]
 83%|████████▎ | 1520/1827 [00:19<00:03, 77.31it/s]
 84%|████████▎ | 1528/1827 [00:19<00:03, 77.21it/s]
 84%|████████▍ | 1536/1827 [00:19<00:03, 77.82it/s]
 85%|████████▍ | 1544/1827 [00:19<00:03, 78.06it/s]
 85%|████████▍ | 1552/1827 [00:19<00:03, 78.36it/s]
 85%|████████▌ | 1560/1827 [00:19<00:03, 78.42it/s]
 86%|████████▌ | 1568/1827 [00:19<00:03, 78.31it/s]
 86%|████████▋ | 1576/1827 [00:19<00:03, 77.83it/s]
 87%|████████▋ | 1584/1827 [00:19<00:03, 78.43it/s]
 87%|████████▋ | 1592/1827 [00:20<00:03, 78.12it/s]
 88%|████████▊ | 1600/1827 [00:20<00:02, 78.44it/s]
 88%|████████▊ | 1609/1827 [00:20<00:02, 79.26it/s]
 89%|████████▊ | 1617/1827 [00:20<00:02, 78.96it/s]
 89%|████████▉ | 1625/1827 [00:20<00:02, 73.35it/s]
 89%|████████▉ | 1633/1827 [00:20<00:02, 74.00it/s]
 90%|████████▉ | 1641/1827 [00:20<00:02, 75.00it/s]
 90%|█████████ | 1650/1827 [00:20<00:02, 76.76it/s]
 91%|█████████ | 1658/1827 [00:20<00:02, 76.99it/s]
 91%|█████████ | 1666/1827 [00:21<00:02, 76.77it/s]
 92%|█████████▏| 1674/1827 [00:21<00:01, 77.34it/s]
 92%|█████████▏| 1683/1827 [00:21<00:01, 78.69it/s]
 93%|█████████▎| 1692/1827 [00:21<00:01, 77.53it/s]
 93%|█████████▎| 1700/1827 [00:21<00:01, 77.18it/s]
 93%|█████████▎| 1708/1827 [00:21<00:01, 77.66it/s]
 94%|█████████▍| 1717/1827 [00:21<00:01, 78.89it/s]
 94%|█████████▍| 1725/1827 [00:21<00:01, 75.55it/s]
 95%|█████████▍| 1733/1827 [00:21<00:01, 76.47it/s]
 95%|█████████▌| 1741/1827 [00:21<00:01, 77.39it/s]
 96%|█████████▌| 1750/1827 [00:22<00:00, 79.17it/s]
 96%|█████████▌| 1758/1827 [00:22<00:00, 79.36it/s]
 97%|█████████▋| 1767/1827 [00:22<00:00, 80.84it/s]
 97%|█████████▋| 1776/1827 [00:22<00:00, 81.41it/s]
 98%|█████████▊| 1785/1827 [00:22<00:00, 82.37it/s]
 98%|█████████▊| 1794/1827 [00:22<00:00, 79.49it/s]
 99%|█████████▊| 1802/1827 [00:22<00:00, 79.48it/s]
 99%|█████████▉| 1810/1827 [00:22<00:00, 79.52it/s]
+[INFO|trainer.py:3503] 2024-09-06 00:35:54,295 >> Saving model checkpoint to /content/dissertation/scripts/ner/output
+[INFO|configuration_utils.py:472] 2024-09-06 00:35:54,297 >> Configuration saved in /content/dissertation/scripts/ner/output/config.json
+[INFO|modeling_utils.py:2799] 2024-09-06 00:35:55,671 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
+[INFO|tokenization_utils_base.py:2684] 2024-09-06 00:35:55,672 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2693] 2024-09-06 00:35:55,672 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+***** predict metrics *****
+  predict_accuracy           =     0.8712
+  predict_f1                 =     0.2297
+  predict_loss               =     0.7416
+  predict_precision          =     0.1307
+  predict_recall             =     0.9487
+  predict_runtime            = 0:00:29.78
+  predict_samples_per_second =    490.651
+  predict_steps_per_second   =      61.34
+

train_results.json

@@ -2,8 +2,8 @@
     "epoch": 10.0,
     "total_flos": 1.7176580067661056e+16,
     "train_loss": 0.0812657987344287,
-    "train_runtime": 1549.44,
+    "train_runtime": 1542.5562,
     "train_samples": 34604,
-    "train_samples_per_second": 223.332,
-    "train_steps_per_second": 3.492
+    "train_samples_per_second": 224.329,
+    "train_steps_per_second": 3.507
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.45455732567249935,
+  "best_metric": 0.17114591920857378,
   "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-2705",
   "epoch": 10.0,
   "eval_steps": 500,
@@ -17,14 +17,14 @@
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8443022505389485,
-      "eval_f1": 0.407486125870823,
-      "eval_loss": 0.47723615169525146,
-      "eval_precision": 0.27250473783954515,
-      "eval_recall": 0.8074403369209172,
-      "eval_runtime": 14.3908,
-      "eval_samples_per_second": 473.218,
-      "eval_steps_per_second": 59.204,
+      "eval_accuracy": 0.7762986255097697,
+      "eval_f1": 0.15197789412449098,
+      "eval_loss": 0.815065860748291,
+      "eval_precision": 0.08251737207833228,
+      "eval_recall": 0.9604779411764706,
+      "eval_runtime": 14.2875,
+      "eval_samples_per_second": 476.64,
+      "eval_steps_per_second": 59.632,
       "step": 541
     },
     {
@@ -36,14 +36,14 @@
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8552871874442172,
-      "eval_f1": 0.43975123088883133,
-      "eval_loss": 0.4583655595779419,
-      "eval_precision": 0.30406737143881024,
-      "eval_recall": 0.7941038839494619,
-      "eval_runtime": 14.212,
-      "eval_samples_per_second": 479.173,
-      "eval_steps_per_second": 59.949,
+      "eval_accuracy": 0.7900778557403161,
+      "eval_f1": 0.168,
+      "eval_loss": 0.833220362663269,
+      "eval_precision": 0.09218777996774771,
+      "eval_recall": 0.9457720588235294,
+      "eval_runtime": 14.245,
+      "eval_samples_per_second": 478.064,
+      "eval_steps_per_second": 59.811,
       "step": 1082
     },
     {
@@ -55,14 +55,14 @@
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8435470361267112,
-      "eval_f1": 0.4338006724608259,
-      "eval_loss": 0.6447410583496094,
-      "eval_precision": 0.29758899817216466,
-      "eval_recall": 0.7999532054281703,
-      "eval_runtime": 14.3233,
-      "eval_samples_per_second": 475.451,
-      "eval_steps_per_second": 59.484,
+      "eval_accuracy": 0.7738201491205184,
+      "eval_f1": 0.16426810845193607,
+      "eval_loss": 1.1093989610671997,
+      "eval_precision": 0.08991208982505004,
+      "eval_recall": 0.9494485294117647,
+      "eval_runtime": 14.3321,
+      "eval_samples_per_second": 475.159,
+      "eval_steps_per_second": 59.447,
       "step": 1623
     },
     {
@@ -74,14 +74,14 @@
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8398602166778805,
-      "eval_f1": 0.42338430558177587,
-      "eval_loss": 0.7413247227668762,
-      "eval_precision": 0.2895756219333735,
-      "eval_recall": 0.7870846981750117,
-      "eval_runtime": 14.5396,
-      "eval_samples_per_second": 468.375,
-      "eval_steps_per_second": 58.598,
+      "eval_accuracy": 0.7740055199307949,
+      "eval_f1": 0.16182605273514364,
+      "eval_loss": 1.1206157207489014,
+      "eval_precision": 0.08849100456227942,
+      "eval_recall": 0.9448529411764706,
+      "eval_runtime": 14.4251,
+      "eval_samples_per_second": 472.092,
+      "eval_steps_per_second": 59.064,
       "step": 2164
     },
     {
@@ -93,14 +93,14 @@
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8564886649182308,
-      "eval_f1": 0.45455732567249935,
-      "eval_loss": 0.7005925178527832,
-      "eval_precision": 0.3152508603513856,
-      "eval_recall": 0.8144595226953674,
-      "eval_runtime": 14.2723,
-      "eval_samples_per_second": 477.148,
-      "eval_steps_per_second": 59.696,
+      "eval_accuracy": 0.7890274211487498,
+      "eval_f1": 0.17114591920857378,
+      "eval_loss": 1.1805996894836426,
+      "eval_precision": 0.09400470929179497,
+      "eval_recall": 0.9540441176470589,
+      "eval_runtime": 14.1983,
+      "eval_samples_per_second": 479.634,
+      "eval_steps_per_second": 60.007,
       "step": 2705
     },
     {
@@ -112,14 +112,14 @@
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8504057561069384,
-      "eval_f1": 0.44700636942675154,
-      "eval_loss": 0.8111857175827026,
-      "eval_precision": 0.30710659898477155,
-      "eval_recall": 0.8210107627515209,
-      "eval_runtime": 14.3991,
-      "eval_samples_per_second": 472.946,
-      "eval_steps_per_second": 59.17,
+      "eval_accuracy": 0.7827454103560493,
+      "eval_f1": 0.1644558094933674,
+      "eval_loss": 1.31381094455719,
+      "eval_precision": 0.09005776299667426,
+      "eval_recall": 0.9457720588235294,
+      "eval_runtime": 14.1863,
+      "eval_samples_per_second": 480.042,
+      "eval_steps_per_second": 60.058,
       "step": 3246
     },
     {
@@ -131,14 +131,14 @@
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.8532961676301372,
-      "eval_f1": 0.4493518337567586,
-      "eval_loss": 0.904643714427948,
-      "eval_precision": 0.3113658932924077,
-      "eval_recall": 0.8069723912026205,
-      "eval_runtime": 14.3036,
-      "eval_samples_per_second": 476.105,
-      "eval_steps_per_second": 59.566,
+      "eval_accuracy": 0.7874071429552226,
+      "eval_f1": 0.16785861076859843,
+      "eval_loss": 1.4008890390396118,
+      "eval_precision": 0.09217297102103457,
+      "eval_recall": 0.9384191176470589,
+      "eval_runtime": 14.2845,
+      "eval_samples_per_second": 476.742,
+      "eval_steps_per_second": 59.645,
       "step": 3787
     },
     {
@@ -150,14 +150,14 @@
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.8481538440413583,
-      "eval_f1": 0.44435897435897437,
-      "eval_loss": 0.9722912907600403,
-      "eval_precision": 0.30602154335158044,
-      "eval_recall": 0.8109499298081423,
-      "eval_runtime": 14.4222,
-      "eval_samples_per_second": 472.19,
-      "eval_steps_per_second": 59.076,
+      "eval_accuracy": 0.7790997844206132,
+      "eval_f1": 0.16497502819397453,
+      "eval_loss": 1.5610512495040894,
+      "eval_precision": 0.0904114426982165,
+      "eval_recall": 0.9411764705882353,
+      "eval_runtime": 14.5465,
+      "eval_samples_per_second": 468.153,
+      "eval_steps_per_second": 58.571,
       "step": 4328
     },
     {
@@ -169,14 +169,14 @@
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.850186057368833,
-      "eval_f1": 0.44668737060041414,
-      "eval_loss": 1.002764105796814,
-      "eval_precision": 0.30871042747272404,
-      "eval_recall": 0.8076743097800655,
-      "eval_runtime": 14.2485,
-      "eval_samples_per_second": 477.944,
-      "eval_steps_per_second": 59.796,
+      "eval_accuracy": 0.7835212215249838,
+      "eval_f1": 0.16788916055419725,
+      "eval_loss": 1.5933887958526611,
+      "eval_precision": 0.09211232337685567,
+      "eval_recall": 0.9466911764705882,
+      "eval_runtime": 14.1983,
+      "eval_samples_per_second": 479.635,
+      "eval_steps_per_second": 60.007,
       "step": 4869
     },
     {
@@ -188,14 +188,14 @@
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.8533304955579661,
-      "eval_f1": 0.45038613797131544,
-      "eval_loss": 1.0022608041763306,
-      "eval_precision": 0.31162999550965426,
-      "eval_recall": 0.8118858212447356,
-      "eval_runtime": 14.4559,
-      "eval_samples_per_second": 471.089,
-      "eval_steps_per_second": 58.938,
+      "eval_accuracy": 0.7862743213368668,
+      "eval_f1": 0.16804385175488834,
+      "eval_loss": 1.6103088855743408,
+      "eval_precision": 0.09223170184104176,
+      "eval_recall": 0.9439338235294118,
+      "eval_runtime": 14.5612,
+      "eval_samples_per_second": 467.681,
+      "eval_steps_per_second": 58.512,
       "step": 5410
     },
     {
@@ -203,9 +203,9 @@
       "step": 5410,
       "total_flos": 1.7176580067661056e+16,
       "train_loss": 0.0812657987344287,
-      "train_runtime": 1549.44,
-      "train_samples_per_second": 223.332,
-      "train_steps_per_second": 3.492
+      "train_runtime": 1542.5562,
+      "train_samples_per_second": 224.329,
+      "train_steps_per_second": 3.507
     }
   ],
   "logging_steps": 500,