End of training

README.md

@@ -2,9 +2,10 @@
 license: apache-2.0
 base_model: PlanTL-GOB-ES/bsc-bio-ehr-es
 tags:
+- token-classification
 - generated_from_trainer
 datasets:
-- drugtemist-ner
+- Rodrigo1771/drugtemist-ner
 metrics:
 - precision
 - recall
@@ -17,24 +18,24 @@ model-index:
       name: Token Classification
       type: token-classification
     dataset:
-      name: drugtemist-ner
-      type: drugtemist-ner
+      name: Rodrigo1771/drugtemist-ner
+      type: Rodrigo1771/drugtemist-ner
       config: DrugTEMIST NER
       split: validation
       args: DrugTEMIST NER
     metrics:
     - name: Precision
       type: precision
-      value: 0.9437386569872959
+      value: 0.9430379746835443
     - name: Recall
       type: recall
-      value: 0.9558823529411765
+      value: 0.9586397058823529
     - name: F1
       type: f1
-      value: 0.949771689497717
+      value: 0.95077484047402
     - name: Accuracy
       type: accuracy
-      value: 0.9990250667691949
+      value: 0.9990319324961724
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -42,12 +43,12 @@ 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 drugtemist-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/drugtemist-ner dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0065
-- Precision: 0.9437
-- Recall: 0.9559
-- F1: 0.9498
+- Loss: 0.0058
+- Precision: 0.9430
+- Recall: 0.9586
+- F1: 0.9508
 - Accuracy: 0.9990
 
 ## Model description

all_results.json

@@ -1,19 +1,26 @@
 {
-    "eval_accuracy": 0.028794858943639246,
-    "eval_f1": 0.00559299062744574,
-    "eval_loss": 1.2929713726043701,
-    "eval_precision": 0.0028258395540381536,
-    "eval_recall": 0.2693014705882353,
-    "eval_runtime": 16.5731,
+    "epoch": 9.997061416397296,
+    "eval_accuracy": 0.9990319324961724,
+    "eval_f1": 0.95077484047402,
+    "eval_loss": 0.0058256350457668304,
+    "eval_precision": 0.9430379746835443,
+    "eval_recall": 0.9586397058823529,
+    "eval_runtime": 15.6142,
     "eval_samples": 6807,
-    "eval_samples_per_second": 410.727,
-    "eval_steps_per_second": 51.348,
-    "predict_accuracy": 0.028794858943639246,
-    "predict_f1": 0.00559299062744574,
-    "predict_loss": 1.2929713726043701,
-    "predict_precision": 0.0028258395540381536,
-    "predict_recall": 0.2693014705882353,
-    "predict_runtime": 15.7694,
-    "predict_samples_per_second": 431.659,
-    "predict_steps_per_second": 53.965
+    "eval_samples_per_second": 435.949,
+    "eval_steps_per_second": 54.502,
+    "predict_accuracy": 0.9990319324961724,
+    "predict_f1": 0.95077484047402,
+    "predict_loss": 0.0058256350457668304,
+    "predict_precision": 0.9430379746835443,
+    "predict_recall": 0.9586397058823529,
+    "predict_runtime": 15.6972,
+    "predict_samples_per_second": 433.645,
+    "predict_steps_per_second": 54.214,
+    "total_flos": 6700358578825584.0,
+    "train_loss": 0.0023016003105682196,
+    "train_runtime": 3387.0101,
+    "train_samples": 27224,
+    "train_samples_per_second": 80.378,
+    "train_steps_per_second": 5.022
 }

eval_results.json

@@ -1,11 +1,12 @@
 {
-    "eval_accuracy": 0.028794858943639246,
-    "eval_f1": 0.00559299062744574,
-    "eval_loss": 1.2929713726043701,
-    "eval_precision": 0.0028258395540381536,
-    "eval_recall": 0.2693014705882353,
-    "eval_runtime": 16.5731,
+    "epoch": 9.997061416397296,
+    "eval_accuracy": 0.9990319324961724,
+    "eval_f1": 0.95077484047402,
+    "eval_loss": 0.0058256350457668304,
+    "eval_precision": 0.9430379746835443,
+    "eval_recall": 0.9586397058823529,
+    "eval_runtime": 15.6142,
     "eval_samples": 6807,
-    "eval_samples_per_second": 410.727,
-    "eval_steps_per_second": 51.348
+    "eval_samples_per_second": 435.949,
+    "eval_steps_per_second": 54.502
 }

predict_results.json

@@ -1,10 +1,10 @@
 {
-    "predict_accuracy": 0.028794858943639246,
-    "predict_f1": 0.00559299062744574,
-    "predict_loss": 1.2929713726043701,
-    "predict_precision": 0.0028258395540381536,
-    "predict_recall": 0.2693014705882353,
-    "predict_runtime": 15.7694,
-    "predict_samples_per_second": 431.659,
-    "predict_steps_per_second": 53.965
+    "predict_accuracy": 0.9990319324961724,
+    "predict_f1": 0.95077484047402,
+    "predict_loss": 0.0058256350457668304,
+    "predict_precision": 0.9430379746835443,
+    "predict_recall": 0.9586397058823529,
+    "predict_runtime": 15.6972,
+    "predict_samples_per_second": 433.645,
+    "predict_steps_per_second": 54.214
 }

tb/events.out.tfevents.1715786919.61af03e56d14.5022.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a27f137b8259d61cfd2526684bc8f5a7589382dd45db907e2e5fc45dc56b2e14
+size 569

train.log

@@ -1590,3 +1590,51 @@ Training completed. Do not forget to share your model on huggingface.co/models =
 [INFO|modeling_utils.py:2590] 2024-05-15 15:28:19,154 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
 [INFO|tokenization_utils_base.py:2488] 2024-05-15 15:28:19,155 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
 [INFO|tokenization_utils_base.py:2497] 2024-05-15 15:28:19,156 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+{'eval_loss': 0.006454338785260916, 'eval_precision': 0.9437386569872959, 'eval_recall': 0.9558823529411765, 'eval_f1': 0.949771689497717, 'eval_accuracy': 0.9990250667691949, 'eval_runtime': 15.7656, 'eval_samples_per_second': 431.764, 'eval_steps_per_second': 53.978, 'epoch': 10.0}
+{'train_runtime': 3387.0101, 'train_samples_per_second': 80.378, 'train_steps_per_second': 5.022, 'train_loss': 0.0023016003105682196, 'epoch': 10.0}
+***** train metrics *****
+  epoch                    =     9.9971
+  total_flos               =  6240195GF
+  train_loss               =     0.0023
+  train_runtime            = 0:56:27.01
+  train_samples            =      27224
+  train_samples_per_second =     80.378
+  train_steps_per_second   =      5.022
+05/15/2024 15:28:23 - INFO - __main__ -   *** Evaluate ***
+[INFO|trainer.py:786] 2024-05-15 15:28:23,892 >> 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:3614] 2024-05-15 15:28:23,895 >> ***** Running Evaluation *****
+[INFO|trainer.py:3616] 2024-05-15 15:28:23,895 >>   Num examples = 6807
+[INFO|trainer.py:3619] 2024-05-15 15:28:23,895 >>   Batch size = 8
+
  0%|          | 0/851 [00:00<?, ?it/s]
  1%|          | 10/851 [00:00<00:09, 90.51it/s]
  2%|▏         | 20/851 [00:00<00:10, 77.83it/s]
  3%|▎         | 28/851 [00:00<00:11, 74.40it/s]
  4%|▍         | 36/851 [00:00<00:11, 73.24it/s]
  5%|▌         | 44/851 [00:00<00:10, 73.71it/s]
  6%|▌         | 52/851 [00:00<00:10, 74.57it/s]
  7%|▋         | 60/851 [00:00<00:10, 75.32it/s]
  8%|▊         | 68/851 [00:00<00:11, 69.52it/s]
  9%|▉         | 76/851 [00:01<00:11, 69.83it/s]
 10%|▉         | 84/851 [00:01<00:10, 70.07it/s]
 11%|█         | 92/851 [00:01<00:10, 69.97it/s]
 12%|█▏        | 100/851 [00:01<00:10, 69.61it/s]
 13%|█▎        | 107/851 [00:01<00:10, 68.65it/s]
 14%|█▎        | 115/851 [00:01<00:10, 68.72it/s]
 15%|█▍        | 124/851 [00:01<00:10, 72.49it/s]
 16%|█▌        | 132/851 [00:01<00:10, 65.39it/s]
 16%|█▋        | 139/851 [00:01<00:10, 65.90it/s]
 17%|█▋        | 147/851 [00:02<00:10, 67.50it/s]
 18%|█▊        | 154/851 [00:02<00:10, 67.04it/s]
 19%|█▉        | 161/851 [00:02<00:10, 66.91it/s]
 20%|█▉        | 169/851 [00:02<00:10, 67.78it/s]
 21%|██        | 177/851 [00:02<00:09, 69.96it/s]
 22%|██▏       | 185/851 [00:02<00:09, 71.55it/s]
 23%|██▎       | 193/851 [00:02<00:09, 72.55it/s]
 24%|██▎       | 201/851 [00:02<00:08, 74.28it/s]
 25%|██▍       | 209/851 [00:02<00:09, 70.56it/s]
 25%|██▌       | 217/851 [00:03<00:09, 68.99it/s]
 26%|██▋       | 224/851 [00:03<00:09, 69.14it/s]
 27%|██▋       | 232/851 [00:03<00:08, 71.37it/s]
 28%|██▊       | 240/851 [00:03<00:08, 68.79it/s]
 29%|██▉       | 247/851 [00:03<00:09, 66.73it/s]
 30%|██▉       | 255/851 [00:03<00:08, 69.91it/s]
 31%|███       | 263/851 [00:03<00:08, 72.69it/s]
 32%|███▏      | 271/851 [00:03<00:08, 71.09it/s]
 33%|███▎      | 279/851 [00:03<00:07, 73.46it/s]
 34%|███▎      | 287/851 [00:04<00:07, 72.07it/s]
 35%|███▍      | 295/851 [00:04<00:07, 72.11it/s]
 36%|███▌      | 303/851 [00:04<00:07, 73.81it/s]
 37%|███▋      | 311/851 [00:04<00:07, 68.28it/s]
 38%|███▊      | 320/851 [00:04<00:07, 71.75it/s]
 39%|███▊      | 328/851 [00:04<00:07, 72.02it/s]
 39%|███▉      | 336/851 [00:04<00:07, 71.36it/s]
 40%|████      | 344/851 [00:04<00:06, 73.18it/s]
 41%|████▏     | 352/851 [00:04<00:06, 72.55it/s]
 42%|████▏     | 360/851 [00:05<00:07, 66.67it/s]
 43%|████▎     | 368/851 [00:05<00:07, 68.29it/s]
 44%|████▍     | 375/851 [00:05<00:06, 68.41it/s]
 45%|████▍     | 382/851 [00:05<00:07, 66.01it/s]
 46%|████▌     | 390/851 [00:05<00:06, 69.04it/s]
 47%|████▋     | 398/851 [00:05<00:06, 70.03it/s]
 48%|████▊     | 406/851 [00:05<00:06, 65.84it/s]
 49%|████▊     | 414/851 [00:05<00:06, 66.93it/s]
 50%|████▉     | 422/851 [00:06<00:06, 69.24it/s]
 50%|█████     | 429/851 [00:06<00:06, 67.99it/s]
 51%|█████▏    | 437/851 [00:06<00:05, 70.32it/s]
 52%|█████▏    | 445/851 [00:06<00:05, 71.78it/s]
 53%|█████▎    | 453/851 [00:06<00:05, 71.61it/s]
 54%|█████▍    | 461/851 [00:06<00:05, 69.59it/s]
 55%|█████▍    | 468/851 [00:06<00:05, 67.35it/s]
 56%|█████▌    | 475/851 [00:06<00:06, 62.52it/s]
 57%|█████▋    | 482/851 [00:06<00:05, 63.14it/s]
 57%|█████▋    | 489/851 [00:07<00:05, 63.97it/s]
 59%|█████▊    | 498/851 [00:07<00:05, 69.11it/s]
 59%|█████▉    | 506/851 [00:07<00:04, 71.64it/s]
 60%|██████    | 514/851 [00:07<00:04, 70.85it/s]
 61%|██████▏   | 522/851 [00:07<00:04, 66.68it/s]
 62%|██████▏   | 529/851 [00:07<00:04, 65.77it/s]
 63%|██████▎   | 537/851 [00:07<00:04, 68.38it/s]
 64%|██████▍   | 545/851 [00:07<00:04, 69.93it/s]
 65%|██████▍   | 553/851 [00:07<00:04, 67.90it/s]
 66%|██████▌   | 561/851 [00:08<00:04, 70.98it/s]
 67%|██████▋   | 569/851 [00:08<00:03, 73.27it/s]
 68%|██████▊   | 577/851 [00:08<00:03, 73.20it/s]
 69%|██████▊   | 585/851 [00:08<00:03, 68.88it/s]
 70%|██████▉   | 592/851 [00:08<00:03, 67.31it/s]
 70%|███████   | 599/851 [00:08<00:03, 67.73it/s]
 71%|███████▏  | 607/851 [00:08<00:03, 68.80it/s]
 72%|███████▏  | 614/851 [00:08<00:03, 66.33it/s]
 73%|███████▎  | 622/851 [00:08<00:03, 67.71it/s]
 74%|███████▍  | 629/851 [00:09<00:03, 63.90it/s]
 75%|███████▍  | 637/851 [00:09<00:03, 66.80it/s]
 76%|███████▌  | 644/851 [00:09<00:03, 62.23it/s]
 77%|███████▋  | 652/851 [00:09<00:03, 65.75it/s]
 78%|███████▊  | 660/851 [00:09<00:02, 68.73it/s]
 78%|███████▊  | 668/851 [00:09<00:02, 70.08it/s]
 79%|███████▉  | 676/851 [00:09<00:02, 69.11it/s]
 80%|████████  | 684/851 [00:09<00:02, 70.02it/s]
 81%|████████▏ | 693/851 [00:09<00:02, 73.25it/s]
 82%|████████▏ | 701/851 [00:10<00:02, 73.96it/s]
 83%|████████▎ | 710/851 [00:10<00:01, 75.96it/s]
 84%|████████▍ | 718/851 [00:10<00:01, 73.07it/s]
 85%|████████▌ | 726/851 [00:10<00:01, 74.63it/s]
 86%|████████▋ | 734/851 [00:10<00:01, 75.56it/s]
 87%|████████▋ | 742/851 [00:10<00:01, 75.66it/s]
 88%|████████▊ | 750/851 [00:10<00:01, 74.81it/s]
 89%|████████▉ | 758/851 [00:10<00:01, 75.25it/s]
 90%|█████████ | 766/851 [00:10<00:01, 71.07it/s]
 91%|█████████ | 774/851 [00:11<00:01, 71.24it/s]
 92%|█████████▏| 782/851 [00:11<00:00, 69.20it/s]
 93%|█████████▎| 790/851 [00:11<00:00, 70.20it/s]
 94%|█████████▍| 798/851 [00:11<00:00, 70.94it/s]
 95%|█████████▍| 806/851 [00:11<00:00, 72.86it/s]
 96%|█████████▌| 814/851 [00:11<00:00, 69.90it/s]
 97%|█████████▋| 822/851 [00:11<00:00, 70.42it/s]
 98%|█████████▊| 830/851 [00:11<00:00, 70.64it/s]
 98%|█████████▊| 838/851 [00:11<00:00, 71.14it/s]
 99%|█████████▉| 846/851 [00:12<00:00, 67.21it/s]
+***** eval metrics *****
+  epoch                   =     9.9971
+  eval_accuracy           =      0.999
+  eval_f1                 =     0.9508
+  eval_loss               =     0.0058
+  eval_precision          =      0.943
+  eval_recall             =     0.9586
+  eval_runtime            = 0:00:15.61
+  eval_samples            =       6807
+  eval_samples_per_second =    435.949
+  eval_steps_per_second   =     54.502
+05/15/2024 15:28:39 - INFO - __main__ -   *** Predict ***
+[INFO|trainer.py:786] 2024-05-15 15:28:39,512 >> The following columns in the test 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:3614] 2024-05-15 15:28:39,514 >> ***** Running Prediction *****
+[INFO|trainer.py:3616] 2024-05-15 15:28:39,514 >>   Num examples = 6807
+[INFO|trainer.py:3619] 2024-05-15 15:28:39,514 >>   Batch size = 8
+
  0%|          | 0/851 [00:00<?, ?it/s]
  1%|          | 9/851 [00:00<00:09, 89.28it/s]
  2%|▏         | 18/851 [00:00<00:10, 76.32it/s]
  3%|▎         | 26/851 [00:00<00:11, 74.84it/s]
  4%|▍         | 34/851 [00:00<00:11, 73.37it/s]
  5%|▍         | 42/851 [00:00<00:11, 72.40it/s]
  6%|▌         | 50/851 [00:00<00:10, 73.86it/s]
  7%|▋         | 58/851 [00:00<00:10, 73.78it/s]
  8%|▊         | 66/851 [00:00<00:10, 71.71it/s]
  9%|▊         | 74/851 [00:01<00:11, 67.99it/s]
 10%|▉         | 82/851 [00:01<00:11, 69.81it/s]
 11%|█         | 90/851 [00:01<00:10, 71.29it/s]
 12%|█▏        | 98/851 [00:01<00:10, 69.84it/s]
 12%|█▏        | 106/851 [00:01<00:10, 70.25it/s]
 13%|█▎        | 114/851 [00:01<00:10, 67.45it/s]
 14%|█▍        | 122/851 [00:01<00:10, 70.51it/s]
 15%|█▌        | 130/851 [00:01<00:11, 65.54it/s]
 16%|█▌        | 137/851 [00:01<00:11, 64.79it/s]
 17%|█▋        | 145/851 [00:02<00:10, 66.82it/s]
 18%|█▊        | 152/851 [00:02<00:10, 65.49it/s]
 19%|█▉        | 160/851 [00:02<00:09, 69.23it/s]
 20%|█▉        | 168/851 [00:02<00:09, 70.13it/s]
 21%|██        | 176/851 [00:02<00:09, 71.41it/s]
 22%|██▏       | 184/851 [00:02<00:09, 71.13it/s]
 23%|██▎       | 192/851 [00:02<00:09, 71.88it/s]
 24%|██▎       | 200/851 [00:02<00:08, 73.17it/s]
 24%|██▍       | 208/851 [00:02<00:09, 69.02it/s]
 25%|██▌       | 215/851 [00:03<00:09, 67.30it/s]
 26%|██▌       | 223/851 [00:03<00:09, 68.83it/s]
 27%|██▋       | 231/851 [00:03<00:08, 69.85it/s]
 28%|██▊       | 239/851 [00:03<00:09, 67.99it/s]
 29%|██▉       | 246/851 [00:03<00:09, 65.91it/s]
 30%|██▉       | 254/851 [00:03<00:08, 69.57it/s]
 31%|███       | 262/851 [00:03<00:08, 71.81it/s]
 32%|███▏      | 270/851 [00:03<00:08, 70.28it/s]
 33%|███▎      | 278/851 [00:03<00:07, 72.94it/s]
 34%|███▎      | 286/851 [00:04<00:07, 73.83it/s]
 35%|███▍      | 294/851 [00:04<00:07, 70.51it/s]
 36%|███▌      | 303/851 [00:04<00:07, 73.46it/s]
 37%|███▋      | 311/851 [00:04<00:07, 70.10it/s]
 37%|███▋      | 319/851 [00:04<00:07, 72.11it/s]
 38%|███▊      | 327/851 [00:04<00:07, 71.11it/s]
 39%|███▉      | 335/851 [00:04<00:07, 70.55it/s]
 40%|████      | 343/851 [00:04<00:07, 72.12it/s]
 41%|████      | 351/851 [00:04<00:06, 72.42it/s]
 42%|████▏     | 359/851 [00:05<00:07, 66.81it/s]
 43%|████▎     | 367/851 [00:05<00:07, 68.19it/s]
 44%|████▍     | 374/851 [00:05<00:07, 67.00it/s]
 45%|████▍     | 381/851 [00:05<00:07, 64.83it/s]
 46%|████▌     | 389/851 [00:05<00:06, 68.13it/s]
 47%|████▋     | 397/851 [00:05<00:06, 69.09it/s]
 47%|████▋     | 404/851 [00:05<00:06, 64.36it/s]
 48%|████▊     | 412/851 [00:05<00:06, 65.60it/s]
 49%|████▉     | 420/851 [00:06<00:06, 68.49it/s]
 50%|█████     | 427/851 [00:06<00:06, 65.71it/s]
 51%|█████     | 435/851 [00:06<00:06, 68.37it/s]
 52%|█████▏    | 443/851 [00:06<00:05, 70.54it/s]
 53%|█████▎    | 451/851 [00:06<00:05, 71.45it/s]
 54%|█████▍    | 459/851 [00:06<00:05, 72.02it/s]
 55%|█████▍    | 467/851 [00:06<00:05, 66.48it/s]
 56%|█████▌    | 474/851 [00:06<00:06, 62.19it/s]
 57%|█████▋    | 481/851 [00:06<00:05, 63.17it/s]
 57%|█████▋    | 488/851 [00:07<00:05, 64.90it/s]
 58%|█████▊    | 496/851 [00:07<00:05, 67.97it/s]
 59%|█████▉    | 504/851 [00:07<00:04, 71.24it/s]
 60%|██████    | 512/851 [00:07<00:04, 70.23it/s]
 61%|██████    | 520/851 [00:07<00:04, 71.31it/s]
 62%|██████▏   | 528/851 [00:07<00:04, 65.51it/s]
 63%|██████▎   | 536/851 [00:07<00:04, 67.75it/s]
 64%|██████▍   | 544/851 [00:07<00:04, 69.64it/s]
 65%|██████▍   | 552/851 [00:07<00:04, 67.32it/s]
 66%|██████▌   | 560/851 [00:08<00:04, 69.83it/s]
 67%|██████▋   | 568/851 [00:08<00:03, 72.10it/s]
 68%|██████▊   | 576/851 [00:08<00:03, 72.39it/s]
 69%|██████▊   | 584/851 [00:08<00:03, 68.89it/s]
 69%|██████▉   | 591/851 [00:08<00:03, 66.82it/s]
 70%|███████   | 598/851 [00:08<00:03, 67.43it/s]
 71%|███████   | 605/851 [00:08<00:03, 67.25it/s]
 72%|███████▏  | 612/851 [00:08<00:03, 65.87it/s]
 73%|███████▎  | 619/851 [00:08<00:03, 65.34it/s]
 74%|███████▎  | 626/851 [00:09<00:03, 64.84it/s]
 74%|███████▍  | 633/851 [00:09<00:03, 65.27it/s]
 75%|███████▌  | 640/851 [00:09<00:03, 64.95it/s]
 76%|███████▌  | 647/851 [00:09<00:03, 63.18it/s]
 77%|███████▋  | 655/851 [00:09<00:02, 66.34it/s]
 78%|███████▊  | 663/851 [00:09<00:02, 67.96it/s]
 79%|███████▊  | 670/851 [00:09<00:02, 68.44it/s]
 80%|███████▉  | 677/851 [00:09<00:02, 68.31it/s]
 80%|████████  | 685/851 [00:09<00:02, 69.35it/s]
 81%|████████▏ | 693/851 [00:10<00:02, 72.25it/s]
 82%|████████▏ | 701/851 [00:10<00:02, 72.73it/s]
 83%|████████▎ | 709/851 [00:10<00:01, 74.24it/s]
 84%|████████▍ | 717/851 [00:10<00:01, 73.24it/s]
 85%|████████▌ | 725/851 [00:10<00:01, 73.55it/s]
 86%|████████▌ | 733/851 [00:10<00:01, 74.45it/s]
 87%|████████▋ | 741/851 [00:10<00:01, 75.51it/s]
 88%|████████▊ | 749/851 [00:10<00:01, 74.86it/s]
 89%|████████▉ | 757/851 [00:10<00:01, 74.62it/s]
 90%|████████▉ | 765/851 [00:11<00:01, 73.45it/s]
 91%|█████████ | 773/851 [00:11<00:01, 70.77it/s]
 92%|█████████▏| 781/851 [00:11<00:01, 67.45it/s]
 93%|█████████▎| 789/851 [00:11<00:00, 69.14it/s]
 94%|█████████▎| 797/851 [00:11<00:00, 70.53it/s]
 95%|█████████▍| 805/851 [00:11<00:00, 73.00it/s]
 96%|█████████▌| 813/851 [00:11<00:00, 69.66it/s]
 96%|█████████▋| 821/851 [00:11<00:00, 70.69it/s]
 97%|█████████▋| 829/851 [00:11<00:00, 71.60it/s]
 98%|█████████▊| 837/851 [00:12<00:00, 72.30it/s]
 99%|█████████▉| 845/851 [00:12<00:00, 66.40it/s]
+[INFO|trainer.py:3305] 2024-05-15 15:28:55,509 >> Saving model checkpoint to /content/dissertation/scripts/ner/output
+[INFO|configuration_utils.py:471] 2024-05-15 15:28:55,510 >> Configuration saved in /content/dissertation/scripts/ner/output/config.json
+[INFO|modeling_utils.py:2590] 2024-05-15 15:28:56,772 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
+[INFO|tokenization_utils_base.py:2488] 2024-05-15 15:28:56,773 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2497] 2024-05-15 15:28:56,774 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+***** predict metrics *****
+  predict_accuracy           =      0.999
+  predict_f1                 =     0.9508
+  predict_loss               =     0.0058
+  predict_precision          =      0.943
+  predict_recall             =     0.9586
+  predict_runtime            = 0:00:15.69
+  predict_samples_per_second =    433.645
+  predict_steps_per_second   =     54.214
+

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.997061416397296,
+    "total_flos": 6700358578825584.0,
+    "train_loss": 0.0023016003105682196,
+    "train_runtime": 3387.0101,
+    "train_samples": 27224,
+    "train_samples_per_second": 80.378,
+    "train_steps_per_second": 5.022
+}

trainer_state.json

@@ -0,0 +1,388 @@
+{
+  "best_metric": 0.95077484047402,
+  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-13612",
+  "epoch": 9.997061416397296,
+  "eval_steps": 500,
+  "global_step": 17010,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.29385836027034967,
+      "grad_norm": 0.008281867019832134,
+      "learning_rate": 4.853027630805409e-05,
+      "loss": 0.0278,
+      "step": 500
+    },
+    {
+      "epoch": 0.5877167205406993,
+      "grad_norm": 0.7454735040664673,
+      "learning_rate": 4.7060552616108174e-05,
+      "loss": 0.0082,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8815750808110491,
+      "grad_norm": 0.028717944398522377,
+      "learning_rate": 4.559082892416226e-05,
+      "loss": 0.0055,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9997061416397296,
+      "eval_accuracy": 0.9984002856142422,
+      "eval_f1": 0.9022907900888265,
+      "eval_loss": 0.003938445821404457,
+      "eval_precision": 0.9181731684110371,
+      "eval_recall": 0.8869485294117647,
+      "eval_runtime": 15.693,
+      "eval_samples_per_second": 433.761,
+      "eval_steps_per_second": 54.228,
+      "step": 1701
+    },
+    {
+      "epoch": 1.1754334410813987,
+      "grad_norm": 0.007521205581724644,
+      "learning_rate": 4.4121105232216346e-05,
+      "loss": 0.0042,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4692918013517484,
+      "grad_norm": 0.0010779600124806166,
+      "learning_rate": 4.265138154027043e-05,
+      "loss": 0.0036,
+      "step": 2500
+    },
+    {
+      "epoch": 1.7631501616220981,
+      "grad_norm": 0.0016530421562492847,
+      "learning_rate": 4.118165784832452e-05,
+      "loss": 0.0037,
+      "step": 3000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9981599851700297,
+      "eval_f1": 0.9000469263256687,
+      "eval_loss": 0.005709913093596697,
+      "eval_precision": 0.9194630872483222,
+      "eval_recall": 0.8814338235294118,
+      "eval_runtime": 15.7138,
+      "eval_samples_per_second": 433.187,
+      "eval_steps_per_second": 54.156,
+      "step": 3403
+    },
+    {
+      "epoch": 2.0570085218924477,
+      "grad_norm": 0.7208561897277832,
+      "learning_rate": 3.971193415637861e-05,
+      "loss": 0.0032,
+      "step": 3500
+    },
+    {
+      "epoch": 2.3508668821627974,
+      "grad_norm": 0.010068068280816078,
+      "learning_rate": 3.824221046443269e-05,
+      "loss": 0.0023,
+      "step": 4000
+    },
+    {
+      "epoch": 2.644725242433147,
+      "grad_norm": 0.017267288640141487,
+      "learning_rate": 3.677248677248677e-05,
+      "loss": 0.0029,
+      "step": 4500
+    },
+    {
+      "epoch": 2.938583602703497,
+      "grad_norm": 0.14932678639888763,
+      "learning_rate": 3.530276308054086e-05,
+      "loss": 0.0024,
+      "step": 5000
+    },
+    {
+      "epoch": 2.9997061416397295,
+      "eval_accuracy": 0.9990113353152399,
+      "eval_f1": 0.9414965986394559,
+      "eval_loss": 0.0036316141486167908,
+      "eval_precision": 0.9292748433303492,
+      "eval_recall": 0.9540441176470589,
+      "eval_runtime": 15.718,
+      "eval_samples_per_second": 433.069,
+      "eval_steps_per_second": 54.142,
+      "step": 5104
+    },
+    {
+      "epoch": 3.2324419629738466,
+      "grad_norm": 0.0006910681258887053,
+      "learning_rate": 3.3833039388594945e-05,
+      "loss": 0.0019,
+      "step": 5500
+    },
+    {
+      "epoch": 3.5263003232441963,
+      "grad_norm": 0.07551395893096924,
+      "learning_rate": 3.2363315696649034e-05,
+      "loss": 0.0017,
+      "step": 6000
+    },
+    {
+      "epoch": 3.820158683514546,
+      "grad_norm": 0.035475652664899826,
+      "learning_rate": 3.0893592004703116e-05,
+      "loss": 0.0013,
+      "step": 6500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9989358123184874,
+      "eval_f1": 0.9402501157943493,
+      "eval_loss": 0.005335611291229725,
+      "eval_precision": 0.9477124183006536,
+      "eval_recall": 0.9329044117647058,
+      "eval_runtime": 15.7659,
+      "eval_samples_per_second": 431.754,
+      "eval_steps_per_second": 53.977,
+      "step": 6806
+    },
+    {
+      "epoch": 4.114017043784895,
+      "grad_norm": 0.0003611127322074026,
+      "learning_rate": 2.9423868312757202e-05,
+      "loss": 0.0011,
+      "step": 7000
+    },
+    {
+      "epoch": 4.407875404055245,
+      "grad_norm": 0.000422955141402781,
+      "learning_rate": 2.795414462081129e-05,
+      "loss": 0.0007,
+      "step": 7500
+    },
+    {
+      "epoch": 4.701733764325595,
+      "grad_norm": 0.00019804929615929723,
+      "learning_rate": 2.648442092886537e-05,
+      "loss": 0.0009,
+      "step": 8000
+    },
+    {
+      "epoch": 4.9955921245959445,
+      "grad_norm": 0.00024289640714414418,
+      "learning_rate": 2.501469723691946e-05,
+      "loss": 0.0012,
+      "step": 8500
+    },
+    {
+      "epoch": 4.99970614163973,
+      "eval_accuracy": 0.9988808865026674,
+      "eval_f1": 0.9406350667280257,
+      "eval_loss": 0.006314857862889767,
+      "eval_precision": 0.9419354838709677,
+      "eval_recall": 0.9393382352941176,
+      "eval_runtime": 15.9896,
+      "eval_samples_per_second": 425.713,
+      "eval_steps_per_second": 53.222,
+      "step": 8507
+    },
+    {
+      "epoch": 5.289450484866294,
+      "grad_norm": 0.0022946298122406006,
+      "learning_rate": 2.3544973544973546e-05,
+      "loss": 0.0014,
+      "step": 9000
+    },
+    {
+      "epoch": 5.583308845136644,
+      "grad_norm": 0.0001970727025764063,
+      "learning_rate": 2.2075249853027632e-05,
+      "loss": 0.0008,
+      "step": 9500
+    },
+    {
+      "epoch": 5.877167205406994,
+      "grad_norm": 1.1225332021713257,
+      "learning_rate": 2.0605526161081718e-05,
+      "loss": 0.0007,
+      "step": 10000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9989426780454649,
+      "eval_f1": 0.9480401093892434,
+      "eval_loss": 0.0047547053545713425,
+      "eval_precision": 0.9403254972875226,
+      "eval_recall": 0.9558823529411765,
+      "eval_runtime": 15.7168,
+      "eval_samples_per_second": 433.104,
+      "eval_steps_per_second": 54.146,
+      "step": 10209
+    },
+    {
+      "epoch": 6.171025565677343,
+      "grad_norm": 0.0004107556596864015,
+      "learning_rate": 1.91358024691358e-05,
+      "loss": 0.0005,
+      "step": 10500
+    },
+    {
+      "epoch": 6.464883925947693,
+      "grad_norm": 1.6069244146347046,
+      "learning_rate": 1.766607877718989e-05,
+      "loss": 0.0005,
+      "step": 11000
+    },
+    {
+      "epoch": 6.758742286218043,
+      "grad_norm": 0.00012789235915988684,
+      "learning_rate": 1.6196355085243976e-05,
+      "loss": 0.0005,
+      "step": 11500
+    },
+    {
+      "epoch": 6.99970614163973,
+      "eval_accuracy": 0.9989220808645324,
+      "eval_f1": 0.9445983379501385,
+      "eval_loss": 0.00655418960377574,
+      "eval_precision": 0.9489795918367347,
+      "eval_recall": 0.9402573529411765,
+      "eval_runtime": 15.8136,
+      "eval_samples_per_second": 430.452,
+      "eval_steps_per_second": 53.814,
+      "step": 11910
+    },
+    {
+      "epoch": 7.052600646488393,
+      "grad_norm": 0.00015387983876280487,
+      "learning_rate": 1.472663139329806e-05,
+      "loss": 0.0002,
+      "step": 12000
+    },
+    {
+      "epoch": 7.346459006758742,
+      "grad_norm": 0.00010029759141616523,
+      "learning_rate": 1.3256907701352148e-05,
+      "loss": 0.0002,
+      "step": 12500
+    },
+    {
+      "epoch": 7.640317367029092,
+      "grad_norm": 9.027074702316895e-05,
+      "learning_rate": 1.1787184009406232e-05,
+      "loss": 0.0002,
+      "step": 13000
+    },
+    {
+      "epoch": 7.934175727299442,
+      "grad_norm": 8.488987077726051e-05,
+      "learning_rate": 1.0317460317460318e-05,
+      "loss": 0.0002,
+      "step": 13500
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9990319324961724,
+      "eval_f1": 0.95077484047402,
+      "eval_loss": 0.0058256350457668304,
+      "eval_precision": 0.9430379746835443,
+      "eval_recall": 0.9586397058823529,
+      "eval_runtime": 15.7444,
+      "eval_samples_per_second": 432.344,
+      "eval_steps_per_second": 54.051,
+      "step": 13612
+    },
+    {
+      "epoch": 8.22803408756979,
+      "grad_norm": 0.0003942732000723481,
+      "learning_rate": 8.847736625514404e-06,
+      "loss": 0.0002,
+      "step": 14000
+    },
+    {
+      "epoch": 8.521892447840141,
+      "grad_norm": 0.006602809764444828,
+      "learning_rate": 7.37801293356849e-06,
+      "loss": 0.0001,
+      "step": 14500
+    },
+    {
+      "epoch": 8.81575080811049,
+      "grad_norm": 8.021829853532836e-05,
+      "learning_rate": 5.908289241622575e-06,
+      "loss": 0.0001,
+      "step": 15000
+    },
+    {
+      "epoch": 8.999706141639729,
+      "eval_accuracy": 0.9990250667691949,
+      "eval_f1": 0.9494305239179955,
+      "eval_loss": 0.006341648753732443,
+      "eval_precision": 0.941282746160795,
+      "eval_recall": 0.9577205882352942,
+      "eval_runtime": 16.0762,
+      "eval_samples_per_second": 423.422,
+      "eval_steps_per_second": 52.936,
+      "step": 15313
+    },
+    {
+      "epoch": 9.10960916838084,
+      "grad_norm": 6.85813938616775e-05,
+      "learning_rate": 4.438565549676661e-06,
+      "loss": 0.0001,
+      "step": 15500
+    },
+    {
+      "epoch": 9.40346752865119,
+      "grad_norm": 0.0018187027890235186,
+      "learning_rate": 2.9688418577307467e-06,
+      "loss": 0.0001,
+      "step": 16000
+    },
+    {
+      "epoch": 9.69732588892154,
+      "grad_norm": 0.000292234995868057,
+      "learning_rate": 1.4991181657848325e-06,
+      "loss": 0.0,
+      "step": 16500
+    },
+    {
+      "epoch": 9.991184249191889,
+      "grad_norm": 6.29909336566925e-05,
+      "learning_rate": 2.9394473838918286e-08,
+      "loss": 0.0,
+      "step": 17000
+    },
+    {
+      "epoch": 9.997061416397296,
+      "eval_accuracy": 0.9990250667691949,
+      "eval_f1": 0.949771689497717,
+      "eval_loss": 0.006454338785260916,
+      "eval_precision": 0.9437386569872959,
+      "eval_recall": 0.9558823529411765,
+      "eval_runtime": 15.7656,
+      "eval_samples_per_second": 431.764,
+      "eval_steps_per_second": 53.978,
+      "step": 17010
+    },
+    {
+      "epoch": 9.997061416397296,
+      "step": 17010,
+      "total_flos": 6700358578825584.0,
+      "train_loss": 0.0023016003105682196,
+      "train_runtime": 3387.0101,
+      "train_samples_per_second": 80.378,
+      "train_steps_per_second": 5.022
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 17010,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 6700358578825584.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}