End of training

README.md

@@ -3,9 +3,10 @@ library_name: transformers
 license: apache-2.0
 base_model: michiyasunaga/BioLinkBERT-base
 tags:
+- token-classification
 - generated_from_trainer
 datasets:
-- drugtemist-en-75-ner
+- Rodrigo1771/drugtemist-en-75-ner
 metrics:
 - precision
 - recall
@@ -18,24 +19,24 @@ model-index:
       name: Token Classification
       type: token-classification
     dataset:
-      name: drugtemist-en-75-ner
-      type: drugtemist-en-75-ner
+      name: Rodrigo1771/drugtemist-en-75-ner
+      type: Rodrigo1771/drugtemist-en-75-ner
       config: DrugTEMIST English NER
       split: validation
       args: DrugTEMIST English NER
     metrics:
     - name: Precision
       type: precision
-      value: 0.921028466483012
+      value: 0.9342105263157895
     - name: Recall
       type: recall
-      value: 0.934762348555452
+      value: 0.9263746505125815
     - name: F1
       type: f1
-      value: 0.9278445883441258
+      value: 0.930276087973795
     - name: Accuracy
       type: accuracy
-      value: 0.9986883598917199
+      value: 0.9987162671280663
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -43,12 +44,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # output
 
-This model is a fine-tuned version of [michiyasunaga/BioLinkBERT-base](https://huggingface.co/michiyasunaga/BioLinkBERT-base) on the drugtemist-en-75-ner dataset.
+This model is a fine-tuned version of [michiyasunaga/BioLinkBERT-base](https://huggingface.co/michiyasunaga/BioLinkBERT-base) on the Rodrigo1771/drugtemist-en-75-ner dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0083
-- Precision: 0.9210
-- Recall: 0.9348
-- F1: 0.9278
+- Loss: 0.0065
+- Precision: 0.9342
+- Recall: 0.9264
+- F1: 0.9303
 - Accuracy: 0.9987
 
 ## Model description

all_results.json

@@ -0,0 +1,26 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.9987162671280663,
+    "eval_f1": 0.930276087973795,
+    "eval_loss": 0.0064844791777431965,
+    "eval_precision": 0.9342105263157895,
+    "eval_recall": 0.9263746505125815,
+    "eval_runtime": 13.1881,
+    "eval_samples": 6946,
+    "eval_samples_per_second": 526.688,
+    "eval_steps_per_second": 65.893,
+    "predict_accuracy": 0.9986488402874071,
+    "predict_f1": 0.9212396929201025,
+    "predict_loss": 0.006953952368348837,
+    "predict_precision": 0.9025069637883009,
+    "predict_recall": 0.9407665505226481,
+    "predict_runtime": 26.178,
+    "predict_samples_per_second": 562.112,
+    "predict_steps_per_second": 70.288,
+    "total_flos": 1.394320679130096e+16,
+    "train_loss": 0.0030765269683407886,
+    "train_runtime": 1249.6681,
+    "train_samples": 32232,
+    "train_samples_per_second": 257.924,
+    "train_steps_per_second": 4.033
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.9987162671280663,
+    "eval_f1": 0.930276087973795,
+    "eval_loss": 0.0064844791777431965,
+    "eval_precision": 0.9342105263157895,
+    "eval_recall": 0.9263746505125815,
+    "eval_runtime": 13.1881,
+    "eval_samples": 6946,
+    "eval_samples_per_second": 526.688,
+    "eval_steps_per_second": 65.893
+}

predict_results.json

@@ -0,0 +1,10 @@
+{
+    "predict_accuracy": 0.9986488402874071,
+    "predict_f1": 0.9212396929201025,
+    "predict_loss": 0.006953952368348837,
+    "predict_precision": 0.9025069637883009,
+    "predict_recall": 0.9407665505226481,
+    "predict_runtime": 26.178,
+    "predict_samples_per_second": 562.112,
+    "predict_steps_per_second": 70.288
+}

tb/events.out.tfevents.1725527837.6cb9bed92fd1.4510.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24dfba6e51d8a0727fb46592f105dea95aa3a1892fc3bb573fbee1166feae180
+size 560

train.log

@@ -1289,3 +1289,51 @@ Training completed. Do not forget to share your model on huggingface.co/models =
 {'eval_loss': 0.008253143168985844, 'eval_precision': 0.921028466483012, 'eval_recall': 0.934762348555452, 'eval_f1': 0.9278445883441258, 'eval_accuracy': 0.9986883598917199, 'eval_runtime': 13.7257, 'eval_samples_per_second': 506.057, 'eval_steps_per_second': 63.312, 'epoch': 10.0}
 {'train_runtime': 1249.6681, 'train_samples_per_second': 257.924, 'train_steps_per_second': 4.033, 'train_loss': 0.0030765269683407886, 'epoch': 10.0}
 
+***** train metrics *****
+  epoch                    =       10.0
+  total_flos               = 12985623GF
+  train_loss               =     0.0031
+  train_runtime            = 0:20:49.66
+  train_samples            =      32232
+  train_samples_per_second =    257.924
+  train_steps_per_second   =      4.033
+09/05/2024 09:17:04 - INFO - __main__ -   *** Evaluate ***
+[INFO|trainer.py:811] 2024-09-05 09:17:04,803 >> The following columns in the evaluation set don't have a corresponding argument in `BertForTokenClassification.forward` and have been ignored: tokens, id, ner_tags. If tokens, id, ner_tags are not expected by `BertForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3819] 2024-09-05 09:17:04,805 >> 
+***** Running Evaluation *****
+[INFO|trainer.py:3821] 2024-09-05 09:17:04,805 >>   Num examples = 6946
+[INFO|trainer.py:3824] 2024-09-05 09:17:04,805 >>   Batch size = 8
+
  0%|          | 0/869 [00:00<?, ?it/s]
  1%|          | 10/869 [00:00<00:08, 97.10it/s]
  2%|▏         | 20/869 [00:00<00:09, 90.01it/s]
  3%|▎         | 30/869 [00:00<00:09, 89.18it/s]
  5%|▍         | 40/869 [00:00<00:09, 90.30it/s]
  6%|▌         | 50/869 [00:00<00:08, 92.35it/s]
  7%|▋         | 60/869 [00:00<00:08, 92.23it/s]
  8%|▊         | 70/869 [00:00<00:08, 91.59it/s]
  9%|▉         | 81/869 [00:00<00:08, 94.55it/s]
 10%|█         | 91/869 [00:00<00:08, 96.01it/s]
 12%|█▏        | 101/869 [00:01<00:07, 97.12it/s]
 13%|█▎        | 111/869 [00:01<00:07, 94.84it/s]
 14%|█▍        | 121/869 [00:01<00:07, 95.04it/s]
 15%|█▌        | 131/869 [00:01<00:07, 93.42it/s]
 16%|█▌        | 141/869 [00:01<00:07, 93.24it/s]
 17%|█▋        | 151/869 [00:01<00:07, 93.28it/s]
 19%|█▊        | 161/869 [00:01<00:07, 92.79it/s]
 20%|█▉        | 171/869 [00:01<00:07, 91.74it/s]
 21%|██        | 181/869 [00:01<00:07, 92.14it/s]
 22%|██▏       | 191/869 [00:02<00:07, 91.28it/s]
 23%|██▎       | 201/869 [00:02<00:07, 91.88it/s]
 24%|██▍       | 211/869 [00:02<00:07, 92.23it/s]
 25%|██▌       | 221/869 [00:02<00:06, 93.41it/s]
 27%|██▋       | 231/869 [00:02<00:06, 92.39it/s]
 28%|██▊       | 241/869 [00:02<00:07, 86.85it/s]
 29%|██▉       | 251/869 [00:02<00:06, 89.50it/s]
 30%|███       | 261/869 [00:02<00:06, 92.32it/s]
 31%|███       | 271/869 [00:02<00:06, 92.37it/s]
 32%|███▏      | 281/869 [00:03<00:06, 91.65it/s]
 33%|███▎      | 291/869 [00:03<00:06, 91.10it/s]
 35%|███▍      | 301/869 [00:03<00:06, 90.53it/s]
 36%|███▌      | 311/869 [00:03<00:06, 90.71it/s]
 37%|███▋      | 321/869 [00:03<00:06, 90.54it/s]
 38%|███▊      | 331/869 [00:03<00:05, 92.74it/s]
 39%|███▉      | 341/869 [00:03<00:05, 92.85it/s]
 40%|████      | 351/869 [00:03<00:05, 89.75it/s]
 42%|████▏     | 361/869 [00:03<00:05, 91.39it/s]
 43%|████▎     | 371/869 [00:04<00:05, 92.76it/s]
 44%|████▍     | 381/869 [00:04<00:05, 90.33it/s]
 45%|████▍     | 391/869 [00:04<00:05, 87.76it/s]
 46%|████▌     | 401/869 [00:04<00:05, 89.51it/s]
 47%|████▋     | 411/869 [00:04<00:04, 91.66it/s]
 48%|████▊     | 421/869 [00:04<00:04, 92.04it/s]
 50%|████▉     | 431/869 [00:04<00:04, 93.80it/s]
 51%|█████     | 441/869 [00:04<00:04, 90.42it/s]
 52%|█████▏    | 451/869 [00:04<00:04, 90.86it/s]
 53%|█████▎    | 461/869 [00:05<00:04, 90.51it/s]
 54%|█████▍    | 471/869 [00:05<00:04, 92.69it/s]
 55%|█████▌    | 482/869 [00:05<00:04, 95.27it/s]
 57%|█████▋    | 492/869 [00:05<00:04, 92.82it/s]
 58%|█████▊    | 502/869 [00:05<00:04, 91.33it/s]
 59%|█████▉    | 512/869 [00:05<00:03, 93.04it/s]
 60%|██████    | 522/869 [00:05<00:03, 90.12it/s]
 61%|██████    | 532/869 [00:05<00:03, 91.17it/s]
 62%|██████▏   | 542/869 [00:05<00:03, 90.57it/s]
 64%|██████▎   | 552/869 [00:06<00:03, 92.71it/s]
 65%|██████▍   | 562/869 [00:06<00:03, 94.03it/s]
 66%|██████▌   | 572/869 [00:06<00:03, 93.29it/s]
 67%|██████▋   | 582/869 [00:06<00:03, 91.83it/s]
 68%|██████▊   | 592/869 [00:06<00:03, 91.81it/s]
 69%|██████▉   | 602/869 [00:06<00:02, 93.95it/s]
 70%|███████   | 612/869 [00:06<00:02, 92.85it/s]
 72%|███████▏  | 622/869 [00:06<00:02, 92.26it/s]
 73%|███████▎  | 632/869 [00:06<00:02, 92.74it/s]
 74%|███████▍  | 642/869 [00:06<00:02, 91.09it/s]
 75%|███████▌  | 652/869 [00:07<00:02, 92.01it/s]
 76%|███████▌  | 662/869 [00:07<00:02, 91.29it/s]
 77%|███████▋  | 672/869 [00:07<00:02, 93.02it/s]
 78%|███████▊  | 682/869 [00:07<00:02, 86.69it/s]
 80%|███████▉  | 692/869 [00:07<00:01, 89.55it/s]
 81%|████████  | 702/869 [00:07<00:01, 89.54it/s]
 82%|████████▏ | 712/869 [00:07<00:01, 91.99it/s]
 83%|████████▎ | 722/869 [00:07<00:01, 91.92it/s]
 84%|████████▍ | 732/869 [00:07<00:01, 93.22it/s]
 85%|████████▌ | 742/869 [00:08<00:01, 93.18it/s]
 87%|████████▋ | 752/869 [00:08<00:01, 92.20it/s]
 88%|████████▊ | 763/869 [00:08<00:01, 95.00it/s]
 89%|████████▉ | 773/869 [00:08<00:01, 92.74it/s]
 90%|█████████ | 783/869 [00:08<00:00, 87.07it/s]
 91%|█████████▏| 793/869 [00:08<00:00, 89.65it/s]
 92%|█████████▏| 803/869 [00:08<00:00, 91.00it/s]
 94%|█████████▎| 813/869 [00:08<00:00, 90.83it/s]
 95%|█████████▍| 823/869 [00:08<00:00, 91.40it/s]
 96%|█████████▌| 833/869 [00:09<00:00, 90.29it/s]
 97%|█████████▋| 843/869 [00:09<00:00, 92.01it/s]
 98%|█████████▊| 853/869 [00:09<00:00, 93.22it/s]
 99%|█████████▉| 863/869 [00:09<00:00, 90.24it/s]
+***** eval metrics *****
+  epoch                   =       10.0
+  eval_accuracy           =     0.9987
+  eval_f1                 =     0.9303
+  eval_loss               =     0.0065
+  eval_precision          =     0.9342
+  eval_recall             =     0.9264
+  eval_runtime            = 0:00:13.18
+  eval_samples            =       6946
+  eval_samples_per_second =    526.688
+  eval_steps_per_second   =     65.893
+09/05/2024 09:17:17 - INFO - __main__ -   *** Predict ***
+[INFO|trainer.py:811] 2024-09-05 09:17:17,999 >> The following columns in the test set don't have a corresponding argument in `BertForTokenClassification.forward` and have been ignored: tokens, id, ner_tags. If tokens, id, ner_tags are not expected by `BertForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3819] 2024-09-05 09:17:18,002 >> 
+***** Running Prediction *****
+[INFO|trainer.py:3821] 2024-09-05 09:17:18,002 >>   Num examples = 14715
+[INFO|trainer.py:3824] 2024-09-05 09:17:18,002 >>   Batch size = 8
+
  0%|          | 0/1840 [00:00<?, ?it/s]
  1%|          | 11/1840 [00:00<00:17, 101.79it/s]
  1%|          | 22/1840 [00:00<00:19, 91.52it/s] 
  2%|▏         | 32/1840 [00:00<00:19, 94.89it/s]
  2%|▏         | 42/1840 [00:00<00:18, 94.92it/s]
  3%|▎         | 52/1840 [00:00<00:18, 94.68it/s]
  3%|▎         | 62/1840 [00:00<00:19, 91.54it/s]
  4%|▍         | 72/1840 [00:00<00:19, 91.04it/s]
  4%|▍         | 82/1840 [00:00<00:19, 92.43it/s]
  5%|▌         | 92/1840 [00:00<00:19, 90.95it/s]
  6%|▌         | 102/1840 [00:01<00:18, 92.60it/s]
  6%|▌         | 112/1840 [00:01<00:18, 93.15it/s]
  7%|▋         | 122/1840 [00:01<00:18, 92.68it/s]
  7%|▋         | 132/1840 [00:01<00:18, 92.00it/s]
  8%|▊         | 142/1840 [00:01<00:18, 91.89it/s]
  8%|▊         | 152/1840 [00:01<00:18, 92.54it/s]
  9%|▉         | 162/1840 [00:01<00:18, 89.05it/s]
  9%|▉         | 172/1840 [00:01<00:18, 90.17it/s]
 10%|▉         | 182/1840 [00:01<00:17, 92.61it/s]
 10%|█         | 192/1840 [00:02<00:17, 93.99it/s]
 11%|█         | 202/1840 [00:02<00:17, 92.45it/s]
 12%|█▏        | 212/1840 [00:02<00:17, 92.36it/s]
 12%|█▏        | 222/1840 [00:02<00:17, 91.82it/s]
 13%|█▎        | 232/1840 [00:02<00:17, 90.21it/s]
 13%|█▎        | 242/1840 [00:02<00:18, 87.53it/s]
 14%|█▎        | 251/1840 [00:02<00:18, 87.80it/s]
 14%|█▍        | 260/1840 [00:02<00:18, 87.51it/s]
 15%|█▍        | 270/1840 [00:02<00:17, 89.69it/s]
 15%|█▌        | 280/1840 [00:03<00:16, 92.41it/s]
 16%|█▌        | 290/1840 [00:03<00:16, 92.07it/s]
 16%|█▋        | 300/1840 [00:03<00:16, 93.43it/s]
 17%|█▋        | 310/1840 [00:03<00:16, 93.16it/s]
 17%|█▋        | 320/1840 [00:03<00:16, 93.68it/s]
 18%|█▊        | 330/1840 [00:03<00:15, 95.47it/s]
 18%|█▊        | 340/1840 [00:03<00:15, 95.33it/s]
 19%|█▉        | 350/1840 [00:03<00:15, 95.92it/s]
 20%|█▉        | 360/1840 [00:03<00:15, 93.05it/s]
 20%|██        | 370/1840 [00:04<00:15, 93.79it/s]
 21%|██        | 380/1840 [00:04<00:15, 94.45it/s]
 21%|██        | 390/1840 [00:04<00:15, 94.48it/s]
 22%|██▏       | 400/1840 [00:04<00:15, 94.37it/s]
 22%|██▏       | 410/1840 [00:04<00:15, 94.57it/s]
 23%|██▎       | 420/1840 [00:04<00:15, 90.65it/s]
 23%|██▎       | 430/1840 [00:04<00:15, 90.83it/s]
 24%|██▍       | 440/1840 [00:04<00:15, 92.12it/s]
 24%|██▍       | 450/1840 [00:04<00:15, 91.31it/s]
 25%|██▌       | 460/1840 [00:04<00:14, 93.20it/s]
 26%|██▌       | 470/1840 [00:05<00:14, 94.92it/s]
 26%|██▌       | 480/1840 [00:05<00:15, 90.56it/s]
 27%|██▋       | 490/1840 [00:05<00:14, 91.66it/s]
 27%|██▋       | 500/1840 [00:05<00:14, 90.02it/s]
 28%|██▊       | 510/1840 [00:05<00:14, 90.04it/s]
 28%|██▊       | 520/1840 [00:05<00:14, 91.31it/s]
 29%|██▉       | 530/1840 [00:05<00:14, 93.38it/s]
 29%|██▉       | 540/1840 [00:05<00:14, 91.96it/s]
 30%|██▉       | 550/1840 [00:05<00:13, 92.37it/s]
 30%|███       | 560/1840 [00:06<00:13, 91.82it/s]
 31%|███       | 570/1840 [00:06<00:13, 93.85it/s]
 32%|███▏      | 580/1840 [00:06<00:13, 95.51it/s]
 32%|███▏      | 590/1840 [00:06<00:13, 94.30it/s]
 33%|███▎      | 600/1840 [00:06<00:13, 92.23it/s]
 33%|███▎      | 610/1840 [00:06<00:13, 92.56it/s]
 34%|███▎      | 620/1840 [00:06<00:13, 90.55it/s]
 34%|███▍      | 630/1840 [00:06<00:13, 90.47it/s]
 35%|███▍      | 640/1840 [00:06<00:13, 92.07it/s]
 35%|███▌      | 650/1840 [00:07<00:12, 92.80it/s]
 36%|███▌      | 660/1840 [00:07<00:13, 90.37it/s]
 36%|███▋      | 670/1840 [00:07<00:12, 91.10it/s]
 37%|███▋      | 680/1840 [00:07<00:12, 90.92it/s]
 38%|███▊      | 690/1840 [00:07<00:12, 91.68it/s]
 38%|███▊      | 700/1840 [00:07<00:12, 91.93it/s]
 39%|███▊      | 710/1840 [00:07<00:12, 90.93it/s]
 39%|███▉      | 720/1840 [00:07<00:12, 93.12it/s]
 40%|███▉      | 731/1840 [00:07<00:11, 95.27it/s]
 40%|████      | 741/1840 [00:08<00:11, 96.12it/s]
 41%|████      | 751/1840 [00:08<00:11, 95.23it/s]
 41%|████▏     | 761/1840 [00:08<00:11, 95.80it/s]
 42%|████▏     | 771/1840 [00:08<00:11, 94.59it/s]
 42%|████▏     | 781/1840 [00:08<00:11, 94.61it/s]
 43%|████▎     | 791/1840 [00:08<00:11, 92.06it/s]
 44%|████▎     | 801/1840 [00:08<00:11, 92.15it/s]
 44%|████▍     | 811/1840 [00:08<00:11, 90.36it/s]
 45%|████▍     | 821/1840 [00:08<00:11, 90.07it/s]
 45%|████▌     | 831/1840 [00:09<00:11, 90.22it/s]
 46%|████▌     | 841/1840 [00:09<00:10, 90.83it/s]
 46%|████▋     | 851/1840 [00:09<00:10, 92.27it/s]
 47%|████▋     | 861/1840 [00:09<00:10, 93.30it/s]
 47%|████▋     | 871/1840 [00:09<00:10, 95.19it/s]
 48%|████▊     | 881/1840 [00:09<00:10, 94.02it/s]
 48%|████▊     | 891/1840 [00:09<00:10, 94.19it/s]
 49%|████▉     | 901/1840 [00:09<00:10, 93.31it/s]
 50%|████▉     | 911/1840 [00:09<00:09, 94.54it/s]
 50%|█████     | 921/1840 [00:09<00:09, 94.97it/s]
 51%|█████     | 931/1840 [00:10<00:09, 94.63it/s]
 51%|█████     | 941/1840 [00:10<00:09, 91.83it/s]
 52%|█████▏    | 951/1840 [00:10<00:09, 91.34it/s]
 52%|█████▏    | 961/1840 [00:10<00:09, 90.38it/s]
 53%|█████▎    | 971/1840 [00:10<00:09, 91.94it/s]
 53%|█████▎    | 981/1840 [00:10<00:09, 91.00it/s]
 54%|█████▍    | 991/1840 [00:10<00:09, 92.18it/s]
 54%|█████▍    | 1001/1840 [00:10<00:09, 92.24it/s]
 55%|█████▍    | 1011/1840 [00:10<00:08, 92.66it/s]
 55%|█████▌    | 1021/1840 [00:11<00:08, 94.00it/s]
 56%|█████▌    | 1031/1840 [00:11<00:08, 95.57it/s]
 57%|█████▋    | 1041/1840 [00:11<00:08, 94.84it/s]
 57%|█████▋    | 1051/1840 [00:11<00:08, 94.30it/s]
 58%|█████▊    | 1061/1840 [00:11<00:08, 93.87it/s]
 58%|█████▊    | 1071/1840 [00:11<00:08, 94.72it/s]
 59%|█████▉    | 1081/1840 [00:11<00:08, 85.95it/s]
 59%|█████▉    | 1091/1840 [00:11<00:08, 89.18it/s]
 60%|█████▉    | 1101/1840 [00:11<00:08, 91.72it/s]
 60%|██████    | 1111/1840 [00:12<00:07, 92.63it/s]
 61%|██████    | 1121/1840 [00:12<00:07, 92.18it/s]
 61%|██████▏   | 1131/1840 [00:12<00:07, 92.87it/s]
 62%|██████▏   | 1141/1840 [00:12<00:07, 92.80it/s]
 63%|██████▎   | 1151/1840 [00:12<00:07, 93.16it/s]
 63%|██████▎   | 1161/1840 [00:12<00:07, 94.60it/s]
 64%|██████▎   | 1171/1840 [00:12<00:07, 92.20it/s]
 64%|██████▍   | 1181/1840 [00:12<00:07, 92.67it/s]
 65%|██████▍   | 1191/1840 [00:12<00:07, 90.14it/s]
 65%|██████▌   | 1201/1840 [00:12<00:06, 91.67it/s]
 66%|██████▌   | 1211/1840 [00:13<00:06, 93.09it/s]
 66%|██████▋   | 1221/1840 [00:13<00:06, 91.59it/s]
 67%|██████▋   | 1231/1840 [00:13<00:06, 91.56it/s]
 67%|██████▋   | 1241/1840 [00:13<00:06, 92.51it/s]
 68%|██████▊   | 1251/1840 [00:13<00:06, 93.68it/s]
 69%|██████▊   | 1261/1840 [00:13<00:06, 91.47it/s]
 69%|██████▉   | 1271/1840 [00:13<00:06, 91.63it/s]
 70%|██████▉   | 1281/1840 [00:13<00:06, 92.14it/s]
 70%|███████   | 1291/1840 [00:13<00:05, 93.88it/s]
 71%|███████   | 1301/1840 [00:14<00:05, 94.62it/s]
 71%|███████▏  | 1312/1840 [00:14<00:05, 96.68it/s]
 72%|███████▏  | 1322/1840 [00:14<00:05, 96.97it/s]
 72%|███████▏  | 1332/1840 [00:14<00:05, 96.37it/s]
 73%|███████▎  | 1342/1840 [00:14<00:05, 96.05it/s]
 73%|███████▎  | 1352/1840 [00:14<00:05, 95.04it/s]
 74%|███████▍  | 1362/1840 [00:14<00:05, 94.28it/s]
 75%|███████▍  | 1372/1840 [00:14<00:04, 94.95it/s]
 75%|███████▌  | 1382/1840 [00:14<00:04, 94.71it/s]
 76%|███████▌  | 1393/1840 [00:15<00:04, 96.04it/s]
 76%|███████▋  | 1403/1840 [00:15<00:04, 96.26it/s]
 77%|███████▋  | 1414/1840 [00:15<00:04, 97.83it/s]
 77%|███████▋  | 1425/1840 [00:15<00:04, 99.04it/s]
 78%|███████▊  | 1435/1840 [00:15<00:04, 97.69it/s]
 79%|███████▊  | 1445/1840 [00:15<00:04, 94.55it/s]
 79%|███████▉  | 1455/1840 [00:15<00:04, 93.85it/s]
 80%|███████▉  | 1466/1840 [00:15<00:03, 95.88it/s]
 80%|████████  | 1476/1840 [00:15<00:03, 96.72it/s]
 81%|████████  | 1486/1840 [00:15<00:03, 92.87it/s]
 81%|████████▏ | 1496/1840 [00:16<00:03, 92.19it/s]
 82%|████████▏ | 1506/1840 [00:16<00:03, 90.11it/s]
 82%|████████▏ | 1516/1840 [00:16<00:03, 89.92it/s]
 83%|████████▎ | 1526/1840 [00:16<00:03, 91.95it/s]
 83%|████████▎ | 1536/1840 [00:16<00:03, 90.92it/s]
 84%|████████▍ | 1546/1840 [00:16<00:03, 90.96it/s]
 85%|████████▍ | 1556/1840 [00:16<00:03, 90.87it/s]
 85%|████████▌ | 1566/1840 [00:16<00:02, 91.37it/s]
 86%|████████▌ | 1576/1840 [00:16<00:02, 90.52it/s]
 86%|████████▌ | 1586/1840 [00:17<00:02, 90.93it/s]
 87%|████████▋ | 1596/1840 [00:17<00:02, 91.78it/s]
 87%|████████▋ | 1606/1840 [00:17<00:02, 92.49it/s]
 88%|████████▊ | 1616/1840 [00:17<00:02, 93.36it/s]
 88%|████████▊ | 1626/1840 [00:17<00:02, 92.76it/s]
 89%|████████▉ | 1636/1840 [00:17<00:02, 86.37it/s]
 89%|████████▉ | 1646/1840 [00:17<00:02, 87.94it/s]
 90%|█████████ | 1656/1840 [00:17<00:02, 89.41it/s]
 91%|█████████ | 1666/1840 [00:17<00:01, 90.62it/s]
 91%|█████████ | 1676/1840 [00:18<00:01, 90.24it/s]
 92%|█████████▏| 1686/1840 [00:18<00:01, 90.23it/s]
 92%|█████████▏| 1696/1840 [00:18<00:01, 92.31it/s]
 93%|█████████▎| 1706/1840 [00:18<00:01, 91.76it/s]
 93%|█████████▎| 1716/1840 [00:18<00:01, 91.75it/s]
 94%|█████████▍| 1726/1840 [00:18<00:01, 92.98it/s]
 94%|█████████▍| 1736/1840 [00:18<00:01, 90.98it/s]
 95%|█████████▍| 1746/1840 [00:18<00:01, 91.89it/s]
 95%|█████████▌| 1756/1840 [00:18<00:00, 93.73it/s]
 96%|█████████▌| 1766/1840 [00:19<00:00, 95.14it/s]
 97%|█████████▋| 1776/1840 [00:19<00:00, 94.88it/s]
 97%|█████████▋| 1786/1840 [00:19<00:00, 95.14it/s]
 98%|█████████▊| 1796/1840 [00:19<00:00, 96.49it/s]
 98%|█████████▊| 1806/1840 [00:19<00:00, 94.16it/s]
 99%|█████████▊| 1816/1840 [00:19<00:00, 93.01it/s]
 99%|█████████▉| 1826/1840 [00:19<00:00, 94.52it/s]
+[INFO|trainer.py:3503] 2024-09-05 09:17:44,719 >> Saving model checkpoint to /content/dissertation/scripts/ner/output
+[INFO|configuration_utils.py:472] 2024-09-05 09:17:44,720 >> Configuration saved in /content/dissertation/scripts/ner/output/config.json
+[INFO|modeling_utils.py:2799] 2024-09-05 09:17:46,010 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
+[INFO|tokenization_utils_base.py:2684] 2024-09-05 09:17:46,011 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2693] 2024-09-05 09:17:46,011 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+***** predict metrics *****
+  predict_accuracy           =     0.9986
+  predict_f1                 =     0.9212
+  predict_loss               =      0.007
+  predict_precision          =     0.9025
+  predict_recall             =     0.9408
+  predict_runtime            = 0:00:26.17
+  predict_samples_per_second =    562.112
+  predict_steps_per_second   =     70.288
+

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.394320679130096e+16,
+    "train_loss": 0.0030765269683407886,
+    "train_runtime": 1249.6681,
+    "train_samples": 32232,
+    "train_samples_per_second": 257.924,
+    "train_steps_per_second": 4.033
+}

trainer_state.json

@@ -0,0 +1,232 @@
+{
+  "best_metric": 0.930276087973795,
+  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-2016",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 5040,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.9920634920634921,
+      "grad_norm": 0.06281786412000656,
+      "learning_rate": 4.503968253968254e-05,
+      "loss": 0.0189,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9983953339100828,
+      "eval_f1": 0.9040358744394619,
+      "eval_loss": 0.00518822530284524,
+      "eval_precision": 0.8712186689714779,
+      "eval_recall": 0.9394221808014911,
+      "eval_runtime": 13.2131,
+      "eval_samples_per_second": 525.689,
+      "eval_steps_per_second": 65.768,
+      "step": 504
+    },
+    {
+      "epoch": 1.9841269841269842,
+      "grad_norm": 0.20689290761947632,
+      "learning_rate": 4.007936507936508e-05,
+      "loss": 0.0047,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9986883598917199,
+      "eval_f1": 0.9244402985074627,
+      "eval_loss": 0.004834321793168783,
+      "eval_precision": 0.9253034547152195,
+      "eval_recall": 0.923578751164958,
+      "eval_runtime": 13.2434,
+      "eval_samples_per_second": 524.487,
+      "eval_steps_per_second": 65.618,
+      "step": 1008
+    },
+    {
+      "epoch": 2.9761904761904763,
+      "grad_norm": 0.02256501279771328,
+      "learning_rate": 3.511904761904762e-05,
+      "loss": 0.0027,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9986255686099406,
+      "eval_f1": 0.9239384041063929,
+      "eval_loss": 0.005881821736693382,
+      "eval_precision": 0.9252336448598131,
+      "eval_recall": 0.9226467847157502,
+      "eval_runtime": 13.1715,
+      "eval_samples_per_second": 527.351,
+      "eval_steps_per_second": 65.976,
+      "step": 1512
+    },
+    {
+      "epoch": 3.9682539682539684,
+      "grad_norm": 0.0326182059943676,
+      "learning_rate": 3.0158730158730158e-05,
+      "loss": 0.0015,
+      "step": 2000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9987162671280663,
+      "eval_f1": 0.930276087973795,
+      "eval_loss": 0.0064844791777431965,
+      "eval_precision": 0.9342105263157895,
+      "eval_recall": 0.9263746505125815,
+      "eval_runtime": 13.5939,
+      "eval_samples_per_second": 510.963,
+      "eval_steps_per_second": 63.926,
+      "step": 2016
+    },
+    {
+      "epoch": 4.9603174603174605,
+      "grad_norm": 0.1600140929222107,
+      "learning_rate": 2.5198412698412697e-05,
+      "loss": 0.0011,
+      "step": 2500
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9986116149917673,
+      "eval_f1": 0.923076923076923,
+      "eval_loss": 0.007346163038164377,
+      "eval_precision": 0.9072907290729073,
+      "eval_recall": 0.9394221808014911,
+      "eval_runtime": 13.3438,
+      "eval_samples_per_second": 520.543,
+      "eval_steps_per_second": 65.124,
+      "step": 2520
+    },
+    {
+      "epoch": 5.9523809523809526,
+      "grad_norm": 0.09641193598508835,
+      "learning_rate": 2.023809523809524e-05,
+      "loss": 0.0005,
+      "step": 3000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.998444171573689,
+      "eval_f1": 0.9204281060958585,
+      "eval_loss": 0.009004838764667511,
+      "eval_precision": 0.9191449814126395,
+      "eval_recall": 0.9217148182665424,
+      "eval_runtime": 13.3195,
+      "eval_samples_per_second": 521.491,
+      "eval_steps_per_second": 65.243,
+      "step": 3024
+    },
+    {
+      "epoch": 6.944444444444445,
+      "grad_norm": 0.029784763231873512,
+      "learning_rate": 1.527777777777778e-05,
+      "loss": 0.0007,
+      "step": 3500
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9985767309463344,
+      "eval_f1": 0.9190432382704691,
+      "eval_loss": 0.008385799825191498,
+      "eval_precision": 0.9073569482288828,
+      "eval_recall": 0.9310344827586207,
+      "eval_runtime": 13.4485,
+      "eval_samples_per_second": 516.487,
+      "eval_steps_per_second": 64.617,
+      "step": 3528
+    },
+    {
+      "epoch": 7.936507936507937,
+      "grad_norm": 0.0010318646673113108,
+      "learning_rate": 1.0317460317460318e-05,
+      "loss": 0.0004,
+      "step": 4000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9985558005190746,
+      "eval_f1": 0.9213793103448276,
+      "eval_loss": 0.008501913398504257,
+      "eval_precision": 0.9092558983666061,
+      "eval_recall": 0.9338303821062441,
+      "eval_runtime": 13.3613,
+      "eval_samples_per_second": 519.86,
+      "eval_steps_per_second": 65.039,
+      "step": 4032
+    },
+    {
+      "epoch": 8.928571428571429,
+      "grad_norm": 0.0005677491426467896,
+      "learning_rate": 5.357142857142857e-06,
+      "loss": 0.0003,
+      "step": 4500
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.9987023135098931,
+      "eval_f1": 0.9270544783010157,
+      "eval_loss": 0.008021777495741844,
+      "eval_precision": 0.918572735590119,
+      "eval_recall": 0.9356943150046598,
+      "eval_runtime": 13.487,
+      "eval_samples_per_second": 515.014,
+      "eval_steps_per_second": 64.432,
+      "step": 4536
+    },
+    {
+      "epoch": 9.920634920634921,
+      "grad_norm": 0.0005077613168396056,
+      "learning_rate": 3.9682539682539683e-07,
+      "loss": 0.0002,
+      "step": 5000
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.9986883598917199,
+      "eval_f1": 0.9278445883441258,
+      "eval_loss": 0.008253143168985844,
+      "eval_precision": 0.921028466483012,
+      "eval_recall": 0.934762348555452,
+      "eval_runtime": 13.7257,
+      "eval_samples_per_second": 506.057,
+      "eval_steps_per_second": 63.312,
+      "step": 5040
+    },
+    {
+      "epoch": 10.0,
+      "step": 5040,
+      "total_flos": 1.394320679130096e+16,
+      "train_loss": 0.0030765269683407886,
+      "train_runtime": 1249.6681,
+      "train_samples_per_second": 257.924,
+      "train_steps_per_second": 4.033
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 5040,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.394320679130096e+16,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}