End of training

README.md

@@ -1,9 +1,10 @@
 ---
 base_model: IVN-RIN/bioBIT
 tags:
+- token-classification
 - generated_from_trainer
 datasets:
-- drugtemist-it-ner
+- Rodrigo1771/drugtemist-it-ner
 metrics:
 - precision
 - recall
@@ -16,8 +17,8 @@ model-index:
       name: Token Classification
       type: token-classification
     dataset:
-      name: drugtemist-it-ner
-      type: drugtemist-it-ner
+      name: Rodrigo1771/drugtemist-it-ner
+      type: Rodrigo1771/drugtemist-it-ner
       config: DrugTEMIST Italian NER
       split: validation
       args: DrugTEMIST Italian NER
@@ -41,7 +42,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # output
 
-This model is a fine-tuned version of [IVN-RIN/bioBIT](https://huggingface.co/IVN-RIN/bioBIT) on the drugtemist-it-ner dataset.
+This model is a fine-tuned version of [IVN-RIN/bioBIT](https://huggingface.co/IVN-RIN/bioBIT) on the Rodrigo1771/drugtemist-it-ner dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.0067
 - Precision: 0.9328

all_results.json

@@ -1,26 +1,26 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.9986953367008066,
-    "eval_f1": 0.9314045730284647,
-    "eval_loss": 0.005624314770102501,
-    "eval_precision": 0.9327102803738317,
-    "eval_recall": 0.9301025163094129,
-    "eval_runtime": 13.3976,
-    "eval_samples": 6946,
-    "eval_samples_per_second": 518.45,
-    "eval_steps_per_second": 64.862,
-    "predict_accuracy": 0.9986842934577083,
-    "predict_f1": 0.9213546039742514,
-    "predict_loss": 0.005766334943473339,
-    "predict_precision": 0.8892490545651,
-    "predict_recall": 0.9558652729384437,
-    "predict_runtime": 26.2719,
-    "predict_samples_per_second": 560.104,
-    "predict_steps_per_second": 70.037,
-    "total_flos": 1.0996932656642544e+16,
-    "train_loss": 0.003382195293697344,
-    "train_runtime": 1039.0596,
-    "train_samples": 27768,
-    "train_samples_per_second": 267.242,
-    "train_steps_per_second": 4.177
+    "eval_accuracy": 0.9988184887042326,
+    "eval_f1": 0.936867469879518,
+    "eval_loss": 0.006724909413605928,
+    "eval_precision": 0.9328214971209213,
+    "eval_recall": 0.9409486931268151,
+    "eval_runtime": 14.1827,
+    "eval_samples": 6798,
+    "eval_samples_per_second": 479.317,
+    "eval_steps_per_second": 59.932,
+    "predict_accuracy": 0.9981367644802958,
+    "predict_f1": 0.8965517241379309,
+    "predict_loss": 0.010722821578383446,
+    "predict_precision": 0.8768736616702355,
+    "predict_recall": 0.9171332586786114,
+    "predict_runtime": 27.7085,
+    "predict_samples_per_second": 527.095,
+    "predict_steps_per_second": 65.9,
+    "total_flos": 1.4262694978690116e+16,
+    "train_loss": 0.0022696754537961062,
+    "train_runtime": 1261.5031,
+    "train_samples": 27198,
+    "train_samples_per_second": 215.6,
+    "train_steps_per_second": 3.369
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.9986953367008066,
-    "eval_f1": 0.9314045730284647,
-    "eval_loss": 0.005624314770102501,
-    "eval_precision": 0.9327102803738317,
-    "eval_recall": 0.9301025163094129,
-    "eval_runtime": 13.3976,
-    "eval_samples": 6946,
-    "eval_samples_per_second": 518.45,
-    "eval_steps_per_second": 64.862
+    "eval_accuracy": 0.9988184887042326,
+    "eval_f1": 0.936867469879518,
+    "eval_loss": 0.006724909413605928,
+    "eval_precision": 0.9328214971209213,
+    "eval_recall": 0.9409486931268151,
+    "eval_runtime": 14.1827,
+    "eval_samples": 6798,
+    "eval_samples_per_second": 479.317,
+    "eval_steps_per_second": 59.932
 }

predict_results.json

@@ -1,10 +1,10 @@
 {
-    "predict_accuracy": 0.9986842934577083,
-    "predict_f1": 0.9213546039742514,
-    "predict_loss": 0.005766334943473339,
-    "predict_precision": 0.8892490545651,
-    "predict_recall": 0.9558652729384437,
-    "predict_runtime": 26.2719,
-    "predict_samples_per_second": 560.104,
-    "predict_steps_per_second": 70.037
+    "predict_accuracy": 0.9981367644802958,
+    "predict_f1": 0.8965517241379309,
+    "predict_loss": 0.010722821578383446,
+    "predict_precision": 0.8768736616702355,
+    "predict_recall": 0.9171332586786114,
+    "predict_runtime": 27.7085,
+    "predict_samples_per_second": 527.095,
+    "predict_steps_per_second": 65.9
 }

tb/events.out.tfevents.1725056175.6b97e535edda.43233.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80aa2e40ff46b611632e40117a4acd4e5f3d41a9a035d30842811f76e0b6b31b
+size 560

train.log

@@ -1336,3 +1336,51 @@ Training completed. Do not forget to share your model on huggingface.co/models =
 {'eval_loss': 0.006724909413605928, 'eval_precision': 0.9328214971209213, 'eval_recall': 0.9409486931268151, 'eval_f1': 0.936867469879518, 'eval_accuracy': 0.9988184887042326, 'eval_runtime': 14.3451, 'eval_samples_per_second': 473.891, 'eval_steps_per_second': 59.254, 'epoch': 10.0}
 {'train_runtime': 1261.5031, 'train_samples_per_second': 215.6, 'train_steps_per_second': 3.369, 'train_loss': 0.0022696754537961062, 'epoch': 10.0}
 
+***** train metrics *****
+  epoch                    =       10.0
+  total_flos               = 13283169GF
+  train_loss               =     0.0023
+  train_runtime            = 0:21:01.50
+  train_samples            =      27198
+  train_samples_per_second =      215.6
+  train_steps_per_second   =      3.369
+08/30/2024 22:16:00 - INFO - __main__ -   *** Evaluate ***
+[INFO|trainer.py:805] 2024-08-30 22:16:00,920 >> The following columns in the evaluation set don't have a corresponding argument in `BertForTokenClassification.forward` and have been ignored: id, tokens, ner_tags. If id, tokens, ner_tags are not expected by `BertForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3788] 2024-08-30 22:16:00,922 >> 
+***** Running Evaluation *****
+[INFO|trainer.py:3790] 2024-08-30 22:16:00,923 >>   Num examples = 6798
+[INFO|trainer.py:3793] 2024-08-30 22:16:00,923 >>   Batch size = 8
+
  0%|          | 0/850 [00:00<?, ?it/s]
  1%|          | 9/850 [00:00<00:09, 88.19it/s]
  2%|▏         | 18/850 [00:00<00:10, 81.56it/s]
  3%|▎         | 27/850 [00:00<00:10, 80.93it/s]
  4%|▍         | 36/850 [00:00<00:09, 84.26it/s]
  5%|▌         | 45/850 [00:00<00:09, 85.17it/s]
  6%|▋         | 54/850 [00:00<00:10, 77.34it/s]
  7%|▋         | 63/850 [00:00<00:10, 78.32it/s]
  8%|▊         | 71/850 [00:00<00:10, 74.58it/s]
  9%|▉         | 79/850 [00:01<00:10, 75.01it/s]
 10%|█         | 88/850 [00:01<00:09, 76.98it/s]
 11%|█▏        | 97/850 [00:01<00:09, 79.71it/s]
 13%|█▎        | 107/850 [00:01<00:08, 83.23it/s]
 14%|█▎        | 116/850 [00:01<00:08, 83.47it/s]
 15%|█▍        | 125/850 [00:01<00:08, 83.94it/s]
 16%|█▌        | 134/850 [00:01<00:08, 81.91it/s]
 17%|█▋        | 144/850 [00:01<00:08, 84.36it/s]
 18%|█▊        | 153/850 [00:01<00:08, 79.26it/s]
 19%|█▉        | 162/850 [00:02<00:08, 81.90it/s]
 20%|██        | 171/850 [00:02<00:08, 83.12it/s]
 21%|██        | 180/850 [00:02<00:07, 84.98it/s]
 22%|██▏       | 189/850 [00:02<00:08, 82.59it/s]
 23%|██▎       | 198/850 [00:02<00:07, 83.32it/s]
 24%|██▍       | 207/850 [00:02<00:07, 82.06it/s]
 25%|██▌       | 216/850 [00:02<00:07, 80.40it/s]
 27%|██▋       | 226/850 [00:02<00:07, 83.55it/s]
 28%|██▊       | 236/850 [00:02<00:07, 85.79it/s]
 29%|██▉       | 245/850 [00:02<00:07, 84.04it/s]
 30%|██▉       | 254/850 [00:03<00:06, 85.38it/s]
 31%|███       | 263/850 [00:03<00:06, 85.08it/s]
 32%|███▏      | 272/850 [00:03<00:06, 86.46it/s]
 33%|███▎      | 281/850 [00:03<00:06, 87.20it/s]
 34%|███▍      | 290/850 [00:03<00:06, 87.65it/s]
 35%|███▌      | 299/850 [00:03<00:06, 85.40it/s]
 36%|███▌      | 308/850 [00:03<00:06, 85.40it/s]
 37%|███▋      | 318/850 [00:03<00:06, 87.55it/s]
 38%|███▊      | 327/850 [00:03<00:06, 85.57it/s]
 40%|███▉      | 336/850 [00:04<00:05, 86.18it/s]
 41%|████      | 345/850 [00:04<00:05, 85.61it/s]
 42%|████▏     | 355/850 [00:04<00:05, 87.78it/s]
 43%|████▎     | 364/850 [00:04<00:05, 84.15it/s]
 44%|████▍     | 373/850 [00:04<00:05, 84.84it/s]
 45%|████▍     | 382/850 [00:04<00:05, 80.74it/s]
 46%|████▌     | 391/850 [00:04<00:05, 80.97it/s]
 47%|████▋     | 400/850 [00:04<00:05, 78.77it/s]
 48%|████▊     | 410/850 [00:04<00:05, 83.43it/s]
 49%|████▉     | 420/850 [00:05<00:05, 85.76it/s]
 51%|█████     | 430/850 [00:05<00:04, 86.88it/s]
 52%|█████▏    | 439/850 [00:05<00:04, 85.98it/s]
 53%|█████▎    | 448/850 [00:05<00:04, 81.98it/s]
 54%|█████▍    | 457/850 [00:05<00:04, 83.43it/s]
 55%|█████▍    | 467/850 [00:05<00:04, 86.56it/s]
 56%|█████▌    | 477/850 [00:05<00:04, 88.16it/s]
 57%|█████▋    | 486/850 [00:05<00:04, 85.36it/s]
 58%|█████▊    | 496/850 [00:05<00:04, 87.52it/s]
 60%|█████▉    | 506/850 [00:06<00:03, 90.01it/s]
 61%|██████    | 516/850 [00:06<00:03, 89.02it/s]
 62%|██████▏   | 525/850 [00:06<00:03, 88.20it/s]
 63%|██████▎   | 534/850 [00:06<00:03, 85.08it/s]
 64%|██████▍   | 543/850 [00:06<00:03, 83.29it/s]
 65%|████���█▍   | 552/850 [00:06<00:03, 81.94it/s]
 66%|██████▌   | 561/850 [00:06<00:03, 80.66it/s]
 67%|██████▋   | 570/850 [00:06<00:03, 79.12it/s]
 68%|██████▊   | 578/850 [00:06<00:03, 78.66it/s]
 69%|██████▉   | 587/850 [00:07<00:03, 80.62it/s]
 70%|███████   | 596/850 [00:07<00:03, 81.35it/s]
 71%|███████   | 605/850 [00:07<00:03, 81.20it/s]
 72%|███████▏  | 615/850 [00:07<00:02, 82.94it/s]
 74%|███████▎  | 625/850 [00:07<00:02, 84.80it/s]
 75%|███████▍  | 634/850 [00:07<00:02, 84.22it/s]
 76%|███████▌  | 643/850 [00:07<00:02, 84.53it/s]
 77%|███████▋  | 653/850 [00:07<00:02, 87.76it/s]
 78%|███████▊  | 662/850 [00:07<00:02, 87.53it/s]
 79%|███████▉  | 671/850 [00:08<00:02, 87.19it/s]
 80%|████████  | 680/850 [00:08<00:01, 86.07it/s]
 81%|████████  | 689/850 [00:08<00:01, 84.91it/s]
 82%|████████▏ | 698/850 [00:08<00:01, 84.44it/s]
 83%|████████▎ | 707/850 [00:08<00:01, 81.34it/s]
 84%|████████▍ | 716/850 [00:08<00:01, 81.52it/s]
 85%|████████▌ | 725/850 [00:08<00:01, 78.53it/s]
 86%|████████▋ | 734/850 [00:08<00:01, 81.34it/s]
 87%|████████▋ | 743/850 [00:08<00:01, 79.59it/s]
 88%|████████▊ | 752/850 [00:09<00:01, 78.93it/s]
 90%|████████▉ | 761/850 [00:09<00:01, 81.25it/s]
 91%|█████████ | 770/850 [00:09<00:00, 81.25it/s]
 92%|█████████▏| 779/850 [00:09<00:00, 80.11it/s]
 93%|█████████▎| 788/850 [00:09<00:00, 78.07it/s]
 94%|█████████▍| 798/850 [00:09<00:00, 82.72it/s]
 95%|█████████▍| 807/850 [00:09<00:00, 81.30it/s]
 96%|█████████▌| 816/850 [00:09<00:00, 80.69it/s]
 97%|█████████▋| 826/850 [00:09<00:00, 82.05it/s]
 98%|█████████▊| 835/850 [00:10<00:00, 82.91it/s]
 99%|█████████▉| 844/850 [00:10<00:00, 83.44it/s]
+***** eval metrics *****
+  epoch                   =       10.0
+  eval_accuracy           =     0.9988
+  eval_f1                 =     0.9369
+  eval_loss               =     0.0067
+  eval_precision          =     0.9328
+  eval_recall             =     0.9409
+  eval_runtime            = 0:00:14.18
+  eval_samples            =       6798
+  eval_samples_per_second =    479.317
+  eval_steps_per_second   =     59.932
+08/30/2024 22:16:15 - INFO - __main__ -   *** Predict ***
+[INFO|trainer.py:805] 2024-08-30 22:16:15,108 >> The following columns in the test set don't have a corresponding argument in `BertForTokenClassification.forward` and have been ignored: id, tokens, ner_tags. If id, tokens, ner_tags are not expected by `BertForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3788] 2024-08-30 22:16:15,110 >> 
+***** Running Prediction *****
+[INFO|trainer.py:3790] 2024-08-30 22:16:15,110 >>   Num examples = 14605
+[INFO|trainer.py:3793] 2024-08-30 22:16:15,110 >>   Batch size = 8
+
  0%|          | 0/1826 [00:00<?, ?it/s]
  1%|          | 10/1826 [00:00<00:18, 99.46it/s]
  1%|          | 20/1826 [00:00<00:22, 80.10it/s]
  2%|▏         | 29/1826 [00:00<00:21, 82.82it/s]
  2%|▏         | 38/1826 [00:00<00:21, 82.91it/s]
  3%|▎         | 47/1826 [00:00<00:20, 85.19it/s]
  3%|▎         | 56/1826 [00:00<00:20, 85.50it/s]
  4%|▎         | 65/1826 [00:00<00:21, 82.13it/s]
  4%|▍         | 75/1826 [00:00<00:20, 84.79it/s]
  5%|▍         | 84/1826 [00:00<00:20, 86.10it/s]
  5%|▌         | 93/1826 [00:01<00:19, 87.04it/s]
  6%|▌         | 103/1826 [00:01<00:19, 89.92it/s]
  6%|▌         | 113/1826 [00:01<00:19, 89.70it/s]
  7%|▋         | 122/1826 [00:01<00:19, 86.11it/s]
  7%|▋         | 131/1826 [00:01<00:19, 84.81it/s]
  8%|▊         | 140/1826 [00:01<00:19, 85.70it/s]
  8%|▊         | 150/1826 [00:01<00:19, 87.70it/s]
  9%|▊         | 159/1826 [00:01<00:20, 82.63it/s]
  9%|▉         | 168/1826 [00:01<00:20, 82.89it/s]
 10%|▉         | 178/1826 [00:02<00:19, 86.63it/s]
 10%|█         | 187/1826 [00:02<00:19, 86.22it/s]
 11%|█         | 197/1826 [00:02<00:18, 86.83it/s]
 11%|█▏        | 206/1826 [00:02<00:18, 85.81it/s]
 12%|█▏        | 215/1826 [00:02<00:19, 84.24it/s]
 12%|█▏        | 224/1826 [00:02<00:18, 85.72it/s]
 13%|█▎        | 233/1826 [00:02<00:19, 82.32it/s]
 13%|█▎        | 242/1826 [00:02<00:19, 81.78it/s]
 14%|█▎        | 251/1826 [00:02<00:19, 81.73it/s]
 14%|█▍        | 260/1826 [00:03<00:19, 81.82it/s]
 15%|█▍        | 270/1826 [00:03<00:18, 84.77it/s]
 15%|█▌        | 280/1826 [00:03<00:17, 88.25it/s]
 16%|█▌        | 289/1826 [00:03<00:17, 86.31it/s]
 16%|█▋        | 298/1826 [00:03<00:17, 86.81it/s]
 17%|█▋        | 307/1826 [00:03<00:17, 86.16it/s]
 17%|█▋        | 316/1826 [00:03<00:17, 85.47it/s]
 18%|█▊        | 326/1826 [00:03<00:17, 87.31it/s]
 18%|█▊        | 335/1826 [00:03<00:17, 86.45it/s]
 19%|█▉        | 345/1826 [00:04<00:16, 88.96it/s]
 19%|█▉        | 354/1826 [00:04<00:17, 84.12it/s]
 20%|█▉        | 364/1826 [00:04<00:16, 86.21it/s]
 20%|██        | 374/1826 [00:04<00:16, 87.29it/s]
 21%|██        | 383/1826 [00:04<00:16, 87.07it/s]
 21%|██▏       | 392/1826 [00:04<00:17, 82.64it/s]
 22%|██▏       | 401/1826 [00:04<00:16, 84.60it/s]
 22%|██▏       | 410/1826 [00:04<00:16, 85.41it/s]
 23%|██▎       | 419/1826 [00:04<00:16, 84.65it/s]
 23%|██▎       | 428/1826 [00:05<00:16, 83.56it/s]
 24%|██▍       | 437/1826 [00:05<00:16, 84.51it/s]
 24%|██▍       | 446/1826 [00:05<00:16, 84.92it/s]
 25%|██▍       | 455/1826 [00:05<00:15, 85.69it/s]
 25%|██▌       | 465/1826 [00:05<00:15, 88.89it/s]
 26%|██▌       | 474/1826 [00:05<00:16, 84.46it/s]
 26%|██▋       | 483/1826 [00:05<00:15, 84.49it/s]
 27%|██▋       | 492/1826 [00:05<00:16, 82.42it/s]
 27%|██▋       | 502/1826 [00:05<00:15, 84.22it/s]
 28%|██▊       | 512/1826 [00:05<00:15, 86.43it/s]
 29%|██▊       | 522/1826 [00:06<00:14, 88.67it/s]
 29%|██▉       | 531/1826 [00:06<00:14, 86.95it/s]
 30%|██▉       | 540/1826 [00:06<00:14, 87.11it/s]
 30%|███       | 549/1826 [00:06<00:14, 85.25it/s]
 31%|███       | 558/1826 [00:06<00:15, 83.01it/s]
 31%|███       | 568/1826 [00:06<00:14, 85.14it/s]
 32%|███▏      | 577/1826 [00:06<00:14, 84.98it/s]
 32%|███▏      | 586/1826 [00:06<00:15, 78.67it/s]
 33%|███▎      | 595/1826 [00:06<00:15, 80.71it/s]
 33%|███▎      | 604/1826 [00:07<00:14, 82.23it/s]
 34%|███▎      | 613/1826 [00:07<00:14, 81.40it/s]
 34%|███▍      | 622/1826 [00:07<00:14, 82.83it/s]
 35%|███▍      | 632/1826 [00:07<00:13, 86.62it/s]
 35%|███▌      | 642/1826 [00:07<00:13, 87.79it/s]
 36%|███▌      | 651/1826 [00:07<00:13, 84.88it/s]
 36%|███▌      | 660/1826 [00:07<00:13, 86.28it/s]
 37%|███▋      | 670/1826 [00:07<00:13, 87.60it/s]
 37%|███▋      | 679/1826 [00:07<00:13, 87.43it/s]
 38%|███▊      | 689/1826 [00:08<00:12, 88.58it/s]
 38%|███▊      | 699/1826 [00:08<00:12, 90.30it/s]
 39%|███▉      | 709/1826 [00:08<00:12, 91.17it/s]
 39%|███▉      | 719/1826 [00:08<00:11, 92.57it/s]
 40%|███▉      | 729/1826 [00:08<00:11, 93.00it/s]
 40%|████      | 739/1826 [00:08<00:12, 90.46it/s]
 41%|████      | 749/1826 [00:08<00:11, 91.75it/s]
 42%|████▏     | 759/1826 [00:08<00:11, 91.80it/s]
 42%|████▏     | 769/1826 [00:08<00:11, 93.48it/s]
 43%|████▎     | 779/1826 [00:09<00:11, 90.59it/s]
 43%|████▎     | 789/1826 [00:09<00:11, 89.82it/s]
 44%|████▍     | 799/1826 [00:09<00:11, 88.45it/s]
 44%|████▍     | 809/1826 [00:09<00:11, 88.97it/s]
 45%|████▍     | 819/1826 [00:09<00:11, 89.58it/s]
 45%|████▌     | 829/1826 [00:09<00:10, 90.95it/s]
 46%|████▌     | 839/1826 [00:09<00:11, 87.82it/s]
 46%|████▋     | 849/1826 [00:09<00:10, 90.77it/s]
 47%|████▋     | 859/1826 [00:09<00:10, 93.15it/s]
 48%|████▊     | 869/1826 [00:10<00:10, 90.43it/s]
 48%|████▊     | 879/1826 [00:10<00:10, 89.07it/s]
 49%|████▊     | 888/1826 [00:10<00:10, 88.03it/s]
 49%|████▉     | 898/1826 [00:10<00:10, 89.92it/s]
 50%|████▉     | 908/1826 [00:10<00:10, 91.01it/s]
 50%|█████     | 918/1826 [00:10<00:09, 91.51it/s]
 51%|█████     | 928/1826 [00:10<00:09, 91.40it/s]
 51%|█████▏    | 938/1826 [00:10<00:10, 88.19it/s]
 52%|█████▏    | 947/1826 [00:10<00:10, 84.32it/s]
 52%|█████▏    | 956/1826 [00:11<00:10, 85.23it/s]
 53%|█████▎    | 966/1826 [00:11<00:09, 86.86it/s]
 53%|█████▎    | 975/1826 [00:11<00:09, 85.98it/s]
 54%|█████▍    | 985/1826 [00:11<00:09, 87.66it/s]
 54%|█████▍    | 994/1826 [00:11<00:09, 85.02it/s]
 55%|█████▍    | 1003/1826 [00:11<00:09, 83.12it/s]
 55%|█████▌    | 1012/1826 [00:11<00:09, 84.61it/s]
 56%|█████▌    | 1022/1826 [00:11<00:09, 88.01it/s]
 57%|█████▋    | 1032/1826 [00:11<00:08, 90.28it/s]
 57%|█████▋    | 1042/1826 [00:12<00:09, 86.92it/s]
 58%|█████▊    | 1051/1826 [00:12<00:09, 85.51it/s]
 58%|█████▊    | 1060/1826 [00:12<00:08, 86.69it/s]
 59%|█████▊    | 1070/1826 [00:12<00:08, 89.80it/s]
 59%|█████▉    | 1080/1826 [00:12<00:08, 91.79it/s]
 60%|█████▉    | 1090/1826 [00:12<00:07, 92.06it/s]
 60%|██████    | 1100/1826 [00:12<00:07, 92.64it/s]
 61%|██████    | 1110/1826 [00:12<00:08, 89.48it/s]
 61%|██████▏   | 1119/1826 [00:12<00:08, 87.58it/s]
 62%|██████▏   | 1129/1826 [00:13<00:07, 89.36it/s]
 62%|██████▏   | 1138/1826 [00:13<00:07, 88.55it/s]
 63%|██████▎   | 1147/1826 [00:13<00:07, 87.81it/s]
 63%|██████▎   | 1156/1826 [00:13<00:07, 87.34it/s]
 64%|██████▍   | 1165/1826 [00:13<00:07, 83.98it/s]
 64%|██████▍   | 1174/1826 [00:13<00:07, 84.35it/s]
 65%|██████▍   | 1183/1826 [00:13<00:07, 82.32it/s]
 65%|██████▌   | 1193/1826 [00:13<00:07, 84.81it/s]
 66%|██████▌   | 1202/1826 [00:13<00:07, 85.14it/s]
 66%|██████▋   | 1211/1826 [00:13<00:07, 83.20it/s]
 67%|██████▋   | 1220/1826 [00:14<00:07, 81.78it/s]
 67%|██████▋   | 1230/1826 [00:14<00:06, 85.50it/s]
 68%|██████▊   | 1240/1826 [00:14<00:06, 86.53it/s]
 68%|██████▊   | 1250/1826 [00:14<00:06, 85.10it/s]
 69%|██████▉   | 1260/1826 [00:14<00:06, 86.63it/s]
 69%|██████▉   | 1269/1826 [00:14<00:06, 87.13it/s]
 70%|██████▉   | 1278/1826 [00:14<00:06, 87.73it/s]
 70%|███████   | 1287/1826 [00:14<00:06, 88.07it/s]
 71%|███████   | 1297/1826 [00:14<00:05, 90.99it/s]
 72%|███████▏  | 1307/1826 [00:15<00:05, 91.31it/s]
 72%|███████▏  | 1317/1826 [00:15<00:05, 93.37it/s]
 73%|███████▎  | 1327/1826 [00:15<00:05, 92.71it/s]
 73%|███████▎  | 1337/1826 [00:15<00:05, 88.69it/s]
 74%|███████▎  | 1346/1826 [00:15<00:05, 85.16it/s]
 74%|███████▍  | 1355/1826 [00:15<00:05, 86.29it/s]
 75%|███████▍  | 1364/1826 [00:15<00:05, 87.05it/s]
 75%|███████▌  | 1373/1826 [00:15<00:05, 85.50it/s]
 76%|███████▌  | 1383/1826 [00:15<00:05, 88.26it/s]
 76%|███████▋  | 1393/1826 [00:16<00:04, 89.45it/s]
 77%|███████▋  | 1403/1826 [00:16<00:04, 91.42it/s]
 77%|███████▋  | 1413/1826 [00:16<00:04, 92.32it/s]
 78%|███████▊  | 1423/1826 [00:16<00:04, 89.99it/s]
 78%|███████▊  | 1433/1826 [00:16<00:04, 90.52it/s]
 79%|███████▉  | 1443/1826 [00:16<00:04, 85.95it/s]
 80%|███████▉  | 1453/1826 [00:16<00:04, 88.09it/s]
 80%|████████  | 1462/1826 [00:16<00:04, 88.33it/s]
 81%|████████  | 1471/1826 [00:16<00:04, 84.47it/s]
 81%|████████  | 1480/1826 [00:17<00:04, 84.58it/s]
 82%|████████▏ | 1489/1826 [00:17<00:04, 80.66it/s]
 82%|████████▏ | 1498/1826 [00:17<00:04, 81.96it/s]
 83%|████████▎ | 1507/1826 [00:17<00:03, 82.12it/s]
 83%|████████▎ | 1517/1826 [00:17<00:03, 84.73it/s]
 84%|████████▎ | 1526/1826 [00:17<00:03, 82.53it/s]
 84%|████████▍ | 1535/1826 [00:17<00:03, 81.76it/s]
 85%|████████▍ | 1544/1826 [00:17<00:03, 82.99it/s]
 85%|████████▌ | 1553/1826 [00:17<00:03, 83.81it/s]
 86%|████████▌ | 1562/1826 [00:18<00:03, 82.96it/s]
 86%|████████▌ | 1571/1826 [00:18<00:03, 81.80it/s]
 87%|████████▋ | 1580/1826 [00:18<00:02, 83.72it/s]
 87%|████████▋ | 1589/1826 [00:18<00:02, 82.90it/s]
 88%|████████▊ | 1598/1826 [00:18<00:02, 84.03it/s]
 88%|████████▊ | 1608/1826 [00:18<00:02, 86.25it/s]
 89%|████████▊ | 1617/1826 [00:18<00:02, 85.18it/s]
 89%|████████▉ | 1626/1826 [00:18<00:02, 77.48it/s]
 90%|████████▉ | 1636/1826 [00:18<00:02, 80.59it/s]
 90%|█████████ | 1645/1826 [00:19<00:02, 82.51it/s]
 91%|█████████ | 1655/1826 [00:19<00:02, 85.30it/s]
 91%|█████████ | 1664/1826 [00:19<00:01, 84.48it/s]
 92%|█████████▏| 1673/1826 [00:19<00:01, 85.56it/s]
 92%|█████████▏| 1683/1826 [00:19<00:01, 88.32it/s]
 93%|█████████▎| 1692/1826 [00:19<00:01, 86.33it/s]
 93%|█████████▎| 1701/1826 [00:19<00:01, 85.33it/s]
 94%|█████████▎| 1711/1826 [00:19<00:01, 87.29it/s]
 94%|█████████▍| 1720/1826 [00:19<00:01, 84.39it/s]
 95%|█████████▍| 1730/1826 [00:20<00:01, 87.06it/s]
 95%|█████████▌| 1740/1826 [00:20<00:00, 87.99it/s]
 96%|█████████▌| 1750/1826 [00:20<00:00, 88.28it/s]
 96%|█████████▋| 1759/1826 [00:20<00:00, 87.89it/s]
 97%|█████████▋| 1768/1826 [00:20<00:00, 87.33it/s]
 97%|█████████▋| 1777/1826 [00:20<00:00, 88.09it/s]
 98%|█████████▊| 1787/1826 [00:20<00:00, 88.65it/s]
 98%|█████████▊| 1796/1826 [00:20<00:00, 83.22it/s]
 99%|█████████▉| 1806/1826 [00:20<00:00, 86.07it/s]
 99%|█████████▉| 1815/1826 [00:20<00:00, 86.85it/s]
+[INFO|trainer.py:3478] 2024-08-30 22:16:43,468 >> Saving model checkpoint to /content/dissertation/scripts/ner/output
+[INFO|configuration_utils.py:472] 2024-08-30 22:16:43,469 >> Configuration saved in /content/dissertation/scripts/ner/output/config.json
+[INFO|modeling_utils.py:2690] 2024-08-30 22:16:44,625 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
+[INFO|tokenization_utils_base.py:2574] 2024-08-30 22:16:44,626 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2583] 2024-08-30 22:16:44,627 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+***** predict metrics *****
+  predict_accuracy           =     0.9981
+  predict_f1                 =     0.8966
+  predict_loss               =     0.0107
+  predict_precision          =     0.8769
+  predict_recall             =     0.9171
+  predict_runtime            = 0:00:27.70
+  predict_samples_per_second =    527.095
+  predict_steps_per_second   =       65.9
+

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 10.0,
-    "total_flos": 1.0996932656642544e+16,
-    "train_loss": 0.003382195293697344,
-    "train_runtime": 1039.0596,
-    "train_samples": 27768,
-    "train_samples_per_second": 267.242,
-    "train_steps_per_second": 4.177
+    "total_flos": 1.4262694978690116e+16,
+    "train_loss": 0.0022696754537961062,
+    "train_runtime": 1261.5031,
+    "train_samples": 27198,
+    "train_samples_per_second": 215.6,
+    "train_steps_per_second": 3.369
 }

trainer_state.json

@@ -1,201 +1,201 @@
 {
-  "best_metric": 0.9314045730284647,
-  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-3038",
+  "best_metric": 0.936867469879518,
+  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-4250",
   "epoch": 10.0,
   "eval_steps": 500,
-  "global_step": 4340,
+  "global_step": 4250,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9980813775011861,
-      "eval_f1": 0.8937558247903076,
-      "eval_loss": 0.005673006176948547,
-      "eval_precision": 0.8937558247903076,
-      "eval_recall": 0.8937558247903076,
-      "eval_runtime": 13.5502,
-      "eval_samples_per_second": 512.611,
-      "eval_steps_per_second": 64.132,
-      "step": 434
-    },
-    {
-      "epoch": 1.1520737327188941,
-      "grad_norm": 0.324382483959198,
-      "learning_rate": 4.423963133640553e-05,
-      "loss": 0.0182,
+      "eval_accuracy": 0.9981303557517528,
+      "eval_f1": 0.8939962476547841,
+      "eval_loss": 0.00556989898905158,
+      "eval_precision": 0.8671519563239308,
+      "eval_recall": 0.9225556631171346,
+      "eval_runtime": 14.25,
+      "eval_samples_per_second": 477.051,
+      "eval_steps_per_second": 59.649,
+      "step": 425
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 0.1311497986316681,
+      "learning_rate": 4.411764705882353e-05,
+      "loss": 0.0104,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9985418469009014,
-      "eval_f1": 0.9160165213400643,
-      "eval_loss": 0.004419892560690641,
-      "eval_precision": 0.9023508137432188,
-      "eval_recall": 0.9301025163094129,
-      "eval_runtime": 13.4449,
-      "eval_samples_per_second": 516.626,
-      "eval_steps_per_second": 64.634,
-      "step": 868
-    },
-    {
-      "epoch": 2.3041474654377883,
-      "grad_norm": 0.0693276971578598,
-      "learning_rate": 3.847926267281106e-05,
-      "loss": 0.0039,
+      "eval_accuracy": 0.9985782913528953,
+      "eval_f1": 0.9216722729456991,
+      "eval_loss": 0.0041933078318834305,
+      "eval_precision": 0.9150763358778626,
+      "eval_recall": 0.9283639883833494,
+      "eval_runtime": 14.1751,
+      "eval_samples_per_second": 479.575,
+      "eval_steps_per_second": 59.965,
+      "step": 850
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 0.002636878052726388,
+      "learning_rate": 3.8235294117647055e-05,
+      "loss": 0.0034,
       "step": 1000
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9986534758462869,
-      "eval_f1": 0.9205175600739371,
-      "eval_loss": 0.0044819144532084465,
-      "eval_precision": 0.9129239230064161,
-      "eval_recall": 0.9282385834109972,
-      "eval_runtime": 13.422,
-      "eval_samples_per_second": 517.51,
-      "eval_steps_per_second": 64.745,
-      "step": 1302
-    },
-    {
-      "epoch": 3.456221198156682,
-      "grad_norm": 0.06998981535434723,
-      "learning_rate": 3.271889400921659e-05,
-      "loss": 0.0024,
+      "eval_accuracy": 0.9985133731498312,
+      "eval_f1": 0.9155339805825242,
+      "eval_loss": 0.004266700241714716,
+      "eval_precision": 0.9182083739045764,
+      "eval_recall": 0.9128751210067764,
+      "eval_runtime": 14.285,
+      "eval_samples_per_second": 475.883,
+      "eval_steps_per_second": 59.503,
+      "step": 1275
+    },
+    {
+      "epoch": 3.5294117647058822,
+      "grad_norm": 0.08669757843017578,
+      "learning_rate": 3.235294117647059e-05,
+      "loss": 0.0022,
       "step": 1500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9983255658192169,
-      "eval_f1": 0.9076923076923078,
-      "eval_loss": 0.005129755008965731,
-      "eval_precision": 0.882145998240985,
-      "eval_recall": 0.934762348555452,
-      "eval_runtime": 13.6729,
-      "eval_samples_per_second": 508.013,
-      "eval_steps_per_second": 63.556,
-      "step": 1736
-    },
-    {
-      "epoch": 4.6082949308755765,
-      "grad_norm": 0.0027600331231951714,
-      "learning_rate": 2.6958525345622122e-05,
-      "loss": 0.0017,
+      "eval_accuracy": 0.9985847831732018,
+      "eval_f1": 0.9250367466927977,
+      "eval_loss": 0.0043651387095451355,
+      "eval_precision": 0.9365079365079365,
+      "eval_recall": 0.9138431752178122,
+      "eval_runtime": 14.5173,
+      "eval_samples_per_second": 468.27,
+      "eval_steps_per_second": 58.551,
+      "step": 1700
+    },
+    {
+      "epoch": 4.705882352941177,
+      "grad_norm": 0.27693310379981995,
+      "learning_rate": 2.647058823529412e-05,
+      "loss": 0.0012,
       "step": 2000
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9986464990372004,
-      "eval_f1": 0.9285051067780873,
-      "eval_loss": 0.0056963409297168255,
-      "eval_precision": 0.9250693802035153,
-      "eval_recall": 0.9319664492078286,
-      "eval_runtime": 13.4387,
-      "eval_samples_per_second": 516.864,
-      "eval_steps_per_second": 64.664,
-      "step": 2170
-    },
-    {
-      "epoch": 5.76036866359447,
-      "grad_norm": 0.018086101859807968,
-      "learning_rate": 2.1198156682027652e-05,
-      "loss": 0.0012,
+      "eval_accuracy": 0.9985393404310569,
+      "eval_f1": 0.919463087248322,
+      "eval_loss": 0.006118799094110727,
+      "eval_precision": 0.9107312440645774,
+      "eval_recall": 0.9283639883833494,
+      "eval_runtime": 14.2824,
+      "eval_samples_per_second": 475.97,
+      "eval_steps_per_second": 59.514,
+      "step": 2125
+    },
+    {
+      "epoch": 5.882352941176471,
+      "grad_norm": 0.008007431402802467,
+      "learning_rate": 2.058823529411765e-05,
+      "loss": 0.0009,
       "step": 2500
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9984023107191695,
-      "eval_f1": 0.9116835326586937,
-      "eval_loss": 0.006062328349798918,
-      "eval_precision": 0.9000908265213442,
-      "eval_recall": 0.923578751164958,
-      "eval_runtime": 13.3783,
-      "eval_samples_per_second": 519.199,
-      "eval_steps_per_second": 64.956,
-      "step": 2604
-    },
-    {
-      "epoch": 6.912442396313364,
-      "grad_norm": 0.009607589803636074,
-      "learning_rate": 1.543778801843318e-05,
-      "loss": 0.0009,
-      "step": 3000
+      "eval_accuracy": 0.9986626850168787,
+      "eval_f1": 0.9221213569039655,
+      "eval_loss": 0.005954863503575325,
+      "eval_precision": 0.910377358490566,
+      "eval_recall": 0.9341723136495643,
+      "eval_runtime": 14.2494,
+      "eval_samples_per_second": 477.072,
+      "eval_steps_per_second": 59.652,
+      "step": 2550
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9986953367008066,
-      "eval_f1": 0.9314045730284647,
-      "eval_loss": 0.005624314770102501,
-      "eval_precision": 0.9327102803738317,
-      "eval_recall": 0.9301025163094129,
-      "eval_runtime": 13.399,
-      "eval_samples_per_second": 518.397,
-      "eval_steps_per_second": 64.856,
-      "step": 3038
+      "eval_accuracy": 0.9986691768371851,
+      "eval_f1": 0.9314148681055155,
+      "eval_loss": 0.006543714087456465,
+      "eval_precision": 0.9230038022813688,
+      "eval_recall": 0.9399806389157793,
+      "eval_runtime": 14.386,
+      "eval_samples_per_second": 472.542,
+      "eval_steps_per_second": 59.085,
+      "step": 2975
+    },
+    {
+      "epoch": 7.0588235294117645,
+      "grad_norm": 0.0017305670771747828,
+      "learning_rate": 1.4705882352941177e-05,
+      "loss": 0.0005,
+      "step": 3000
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9985837077554209,
-      "eval_f1": 0.9231477220432582,
-      "eval_loss": 0.006831143982708454,
-      "eval_precision": 0.9118181818181819,
-      "eval_recall": 0.934762348555452,
-      "eval_runtime": 13.3718,
-      "eval_samples_per_second": 519.45,
-      "eval_steps_per_second": 64.987,
-      "step": 3472
-    },
-    {
-      "epoch": 8.064516129032258,
-      "grad_norm": 0.0005392630700953305,
-      "learning_rate": 9.67741935483871e-06,
-      "loss": 0.0006,
+      "eval_accuracy": 0.9986886522981044,
+      "eval_f1": 0.9280540801545147,
+      "eval_loss": 0.005883762612938881,
+      "eval_precision": 0.9258188824662813,
+      "eval_recall": 0.9303000968054211,
+      "eval_runtime": 14.3169,
+      "eval_samples_per_second": 474.822,
+      "eval_steps_per_second": 59.37,
+      "step": 3400
+    },
+    {
+      "epoch": 8.235294117647058,
+      "grad_norm": 0.00020609228522516787,
+      "learning_rate": 8.823529411764707e-06,
+      "loss": 0.0004,
       "step": 3500
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.9986813830826333,
-      "eval_f1": 0.9288702928870293,
-      "eval_loss": 0.0071532572619616985,
-      "eval_precision": 0.9267161410018553,
-      "eval_recall": 0.9310344827586207,
-      "eval_runtime": 13.5255,
-      "eval_samples_per_second": 513.548,
-      "eval_steps_per_second": 64.249,
-      "step": 3906
-    },
-    {
-      "epoch": 9.216589861751151,
-      "grad_norm": 0.0045371875166893005,
-      "learning_rate": 3.9170506912442395e-06,
-      "loss": 0.0004,
+      "eval_accuracy": 0.9987276032199429,
+      "eval_f1": 0.9317307692307693,
+      "eval_loss": 0.00656876852735877,
+      "eval_precision": 0.9255014326647565,
+      "eval_recall": 0.9380445304937076,
+      "eval_runtime": 14.5715,
+      "eval_samples_per_second": 466.526,
+      "eval_steps_per_second": 58.333,
+      "step": 3825
+    },
+    {
+      "epoch": 9.411764705882353,
+      "grad_norm": 0.00026785818045027554,
+      "learning_rate": 2.9411764705882355e-06,
+      "loss": 0.0001,
       "step": 4000
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.998618591800854,
-      "eval_f1": 0.9259944495837187,
-      "eval_loss": 0.007294897455722094,
-      "eval_precision": 0.9191919191919192,
-      "eval_recall": 0.9328984156570364,
-      "eval_runtime": 13.8041,
-      "eval_samples_per_second": 503.184,
-      "eval_steps_per_second": 62.952,
-      "step": 4340
+      "eval_accuracy": 0.9988184887042326,
+      "eval_f1": 0.936867469879518,
+      "eval_loss": 0.006724909413605928,
+      "eval_precision": 0.9328214971209213,
+      "eval_recall": 0.9409486931268151,
+      "eval_runtime": 14.3451,
+      "eval_samples_per_second": 473.891,
+      "eval_steps_per_second": 59.254,
+      "step": 4250
     },
     {
       "epoch": 10.0,
-      "step": 4340,
-      "total_flos": 1.0996932656642544e+16,
-      "train_loss": 0.003382195293697344,
-      "train_runtime": 1039.0596,
-      "train_samples_per_second": 267.242,
-      "train_steps_per_second": 4.177
+      "step": 4250,
+      "total_flos": 1.4262694978690116e+16,
+      "train_loss": 0.0022696754537961062,
+      "train_runtime": 1261.5031,
+      "train_samples_per_second": 215.6,
+      "train_steps_per_second": 3.369
     }
   ],
   "logging_steps": 500,
-  "max_steps": 4340,
+  "max_steps": 4250,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
@@ -211,7 +211,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.0996932656642544e+16,
+  "total_flos": 1.4262694978690116e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null