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-fasttext-75-ner
+- Rodrigo1771/drugtemist-en-fasttext-75-ner
 metrics:
 - precision
 - recall
@@ -18,24 +19,24 @@ model-index:
       name: Token Classification
       type: token-classification
     dataset:
-      name: drugtemist-en-fasttext-75-ner
-      type: drugtemist-en-fasttext-75-ner
+      name: Rodrigo1771/drugtemist-en-fasttext-75-ner
+      type: Rodrigo1771/drugtemist-en-fasttext-75-ner
       config: DrugTEMIST English NER
       split: validation
       args: DrugTEMIST English NER
     metrics:
     - name: Precision
       type: precision
-      value: 0.9194876486733761
+      value: 0.9249771271729186
     - name: Recall
       type: recall
-      value: 0.9366262814538676
+      value: 0.9422180801491147
     - name: F1
       type: f1
-      value: 0.92797783933518
+      value: 0.9335180055401663
     - name: Accuracy
       type: accuracy
-      value: 0.9987511511734993
+      value: 0.998772081600759
 ---
 
 <!-- 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-fasttext-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-fasttext-75-ner dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0077
-- Precision: 0.9195
-- Recall: 0.9366
-- F1: 0.9280
+- Loss: 0.0076
+- Precision: 0.9250
+- Recall: 0.9422
+- F1: 0.9335
 - Accuracy: 0.9988
 
 ## Model description

all_results.json

@@ -1,26 +1,26 @@
 {
-    "epoch": 9.967845659163988,
-    "eval_accuracy": 0.9496615226667522,
-    "eval_f1": 0.6946045049764275,
-    "eval_loss": 0.24502409994602203,
-    "eval_precision": 0.6659969864389754,
-    "eval_recall": 0.7257799671592775,
-    "eval_runtime": 5.9273,
-    "eval_samples": 2519,
-    "eval_samples_per_second": 424.981,
-    "eval_steps_per_second": 53.144,
-    "predict_accuracy": 0.9468015056363083,
-    "predict_f1": 0.6798307475317349,
-    "predict_loss": 0.26582667231559753,
-    "predict_precision": 0.6624923640806353,
-    "predict_recall": 0.6981010621177985,
-    "predict_runtime": 9.8107,
-    "predict_samples_per_second": 412.509,
-    "predict_steps_per_second": 51.576,
-    "total_flos": 4644619911314910.0,
-    "train_loss": 0.050868147861573,
-    "train_runtime": 855.1929,
-    "train_samples": 9929,
-    "train_samples_per_second": 116.102,
-    "train_steps_per_second": 1.812
+    "epoch": 10.0,
+    "eval_accuracy": 0.998772081600759,
+    "eval_f1": 0.9335180055401663,
+    "eval_loss": 0.007628325838595629,
+    "eval_precision": 0.9249771271729186,
+    "eval_recall": 0.9422180801491147,
+    "eval_runtime": 15.1819,
+    "eval_samples": 6946,
+    "eval_samples_per_second": 457.519,
+    "eval_steps_per_second": 57.239,
+    "predict_accuracy": 0.9986685364931299,
+    "predict_f1": 0.9202592279515356,
+    "predict_loss": 0.007816320285201073,
+    "predict_precision": 0.8938149972632731,
+    "predict_recall": 0.9483159117305459,
+    "predict_runtime": 28.7456,
+    "predict_samples_per_second": 511.904,
+    "predict_steps_per_second": 64.01,
+    "total_flos": 1.394810359803495e+16,
+    "train_loss": 0.0028968164414402532,
+    "train_runtime": 2196.5741,
+    "train_samples": 32447,
+    "train_samples_per_second": 147.716,
+    "train_steps_per_second": 2.308
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 9.967845659163988,
-    "eval_accuracy": 0.9496615226667522,
-    "eval_f1": 0.6946045049764275,
-    "eval_loss": 0.24502409994602203,
-    "eval_precision": 0.6659969864389754,
-    "eval_recall": 0.7257799671592775,
-    "eval_runtime": 5.9273,
-    "eval_samples": 2519,
-    "eval_samples_per_second": 424.981,
-    "eval_steps_per_second": 53.144
+    "epoch": 10.0,
+    "eval_accuracy": 0.998772081600759,
+    "eval_f1": 0.9335180055401663,
+    "eval_loss": 0.007628325838595629,
+    "eval_precision": 0.9249771271729186,
+    "eval_recall": 0.9422180801491147,
+    "eval_runtime": 15.1819,
+    "eval_samples": 6946,
+    "eval_samples_per_second": 457.519,
+    "eval_steps_per_second": 57.239
 }

predict_results.json

@@ -1,10 +1,10 @@
 {
-    "predict_accuracy": 0.9468015056363083,
-    "predict_f1": 0.6798307475317349,
-    "predict_loss": 0.26582667231559753,
-    "predict_precision": 0.6624923640806353,
-    "predict_recall": 0.6981010621177985,
-    "predict_runtime": 9.8107,
-    "predict_samples_per_second": 412.509,
-    "predict_steps_per_second": 51.576
+    "predict_accuracy": 0.9986685364931299,
+    "predict_f1": 0.9202592279515356,
+    "predict_loss": 0.007816320285201073,
+    "predict_precision": 0.8938149972632731,
+    "predict_recall": 0.9483159117305459,
+    "predict_runtime": 28.7456,
+    "predict_samples_per_second": 511.904,
+    "predict_steps_per_second": 64.01
 }

tb/events.out.tfevents.1725888457.0a1c9bec2a53.24273.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bb5583c37fdbd9ea659142a09fba0fae95a78d6f0eaac64a0292bc00e64215d
+size 560

train.log

@@ -1429,3 +1429,51 @@ Training completed. Do not forget to share your model on huggingface.co/models =
 {'eval_loss': 0.00769586768001318, 'eval_precision': 0.9194876486733761, 'eval_recall': 0.9366262814538676, 'eval_f1': 0.92797783933518, 'eval_accuracy': 0.9987511511734993, 'eval_runtime': 15.2047, 'eval_samples_per_second': 456.832, 'eval_steps_per_second': 57.153, 'epoch': 10.0}
 {'train_runtime': 2196.5741, 'train_samples_per_second': 147.716, 'train_steps_per_second': 2.308, 'train_loss': 0.0028968164414402532, 'epoch': 10.0}
 
+***** train metrics *****
+  epoch                    =       10.0
+  total_flos               = 12990183GF
+  train_loss               =     0.0029
+  train_runtime            = 0:36:36.57
+  train_samples            =      32447
+  train_samples_per_second =    147.716
+  train_steps_per_second   =      2.308
+09/09/2024 13:27:22 - INFO - __main__ -   *** Evaluate ***
+[INFO|trainer.py:811] 2024-09-09 13:27:22,340 >> 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:3819] 2024-09-09 13:27:22,343 >> 
+***** Running Evaluation *****
+[INFO|trainer.py:3821] 2024-09-09 13:27:22,343 >>   Num examples = 6946
+[INFO|trainer.py:3824] 2024-09-09 13:27:22,343 >>   Batch size = 8
+
  0%|          | 0/869 [00:00<?, ?it/s]
  1%|          | 10/869 [00:00<00:09, 93.14it/s]
  2%|▏         | 20/869 [00:00<00:10, 79.68it/s]
  3%|▎         | 29/869 [00:00<00:10, 78.36it/s]
  4%|▍         | 37/869 [00:00<00:11, 75.35it/s]
  5%|▌         | 46/869 [00:00<00:10, 78.54it/s]
  6%|▋         | 55/869 [00:00<00:10, 81.34it/s]
  7%|▋         | 64/869 [00:00<00:10, 77.39it/s]
  8%|▊         | 72/869 [00:00<00:10, 77.10it/s]
  9%|▉         | 81/869 [00:01<00:09, 80.79it/s]
 10%|█         | 90/869 [00:01<00:09, 83.27it/s]
 12%|█▏        | 100/869 [00:01<00:08, 85.89it/s]
 13%|█▎        | 109/869 [00:01<00:09, 82.19it/s]
 14%|█▎        | 118/869 [00:01<00:09, 80.71it/s]
 15%|█▍        | 127/869 [00:01<00:09, 79.46it/s]
 16%|█▌        | 136/869 [00:01<00:09, 80.95it/s]
 17%|█▋        | 145/869 [00:01<00:09, 76.58it/s]
 18%|█▊        | 154/869 [00:01<00:09, 78.77it/s]
 19%|█▉        | 163/869 [00:02<00:08, 78.83it/s]
 20%|█▉        | 171/869 [00:02<00:09, 76.42it/s]
 21%|██        | 179/869 [00:02<00:09, 75.56it/s]
 22%|██▏       | 188/869 [00:02<00:08, 77.42it/s]
 23%|██▎       | 196/869 [00:02<00:09, 74.78it/s]
 24%|██▎       | 205/869 [00:02<00:08, 77.47it/s]
 25%|██▍       | 214/869 [00:02<00:08, 79.04it/s]
 26%|██▌       | 223/869 [00:02<00:08, 80.38it/s]
 27%|██▋       | 232/869 [00:02<00:08, 76.59it/s]
 28%|██▊       | 240/869 [00:03<00:08, 77.10it/s]
 29%|██▊       | 248/869 [00:03<00:08, 71.77it/s]
 30%|██▉       | 257/869 [00:03<00:08, 74.54it/s]
 30%|███       | 265/869 [00:03<00:08, 73.96it/s]
 31%|███▏      | 273/869 [00:03<00:07, 74.71it/s]
 32%|███▏      | 281/869 [00:03<00:07, 74.68it/s]
 33%|███▎      | 289/869 [00:03<00:07, 73.95it/s]
 34%|███▍      | 297/869 [00:03<00:07, 74.29it/s]
 35%|███▌      | 305/869 [00:03<00:08, 70.32it/s]
 36%|███▌      | 313/869 [00:04<00:07, 71.72it/s]
 37%|███▋      | 321/869 [00:04<00:07, 71.00it/s]
 38%|███▊      | 330/869 [00:04<00:07, 74.74it/s]
 39%|███▉      | 339/869 [00:04<00:06, 76.71it/s]
 40%|███▉      | 347/869 [00:04<00:07, 72.09it/s]
 41%|████      | 355/869 [00:04<00:07, 72.99it/s]
 42%|████▏     | 364/869 [00:04<00:06, 76.76it/s]
 43%|████▎     | 373/869 [00:04<00:06, 74.39it/s]
 44%|████▍     | 381/869 [00:04<00:06, 75.28it/s]
 45%|████▍     | 389/869 [00:05<00:06, 70.47it/s]
 46%|████▌     | 398/869 [00:05<00:06, 73.16it/s]
 47%|████▋     | 407/869 [00:05<00:06, 75.53it/s]
 48%|████▊     | 416/869 [00:05<00:05, 78.44it/s]
 49%|████▉     | 424/869 [00:05<00:05, 78.67it/s]
 50%|████▉     | 433/869 [00:05<00:05, 79.51it/s]
 51%|█████     | 441/869 [00:05<00:05, 77.23it/s]
 52%|█████▏    | 449/869 [00:05<00:05, 75.66it/s]
 53%|█████▎    | 457/869 [00:05<00:05, 76.56it/s]
 54%|█████▎    | 465/869 [00:06<00:05, 74.67it/s]
 55%|█████▍    | 475/869 [00:06<00:04, 79.79it/s]
 56%|█████▌    | 484/869 [00:06<00:04, 82.19it/s]
 57%|█████▋    | 493/869 [00:06<00:04, 76.64it/s]
 58%|█████▊    | 501/869 [00:06<00:04, 76.35it/s]
 59%|█████▊    | 509/869 [00:06<00:04, 73.64it/s]
 60%|█████▉    | 518/869 [00:06<00:04, 76.46it/s]
 61%|██████    | 526/869 [00:06<00:04, 71.73it/s]
 61%|██████▏   | 534/869 [00:06<00:04, 73.69it/s]
 62%|██████▏   | 542/869 [00:07<00:04, 70.92it/s]
 63%|██████▎   | 551/869 [00:07<00:04, 74.82it/s]
 64%|██████▍   | 559/869 [00:07<00:04, 74.53it/s]
 65%|██████▌   | 567/869 [00:07<00:04, 73.79it/s]
 66%|██████▌   | 575/869 [00:07<00:03, 74.82it/s]
 67%|██████▋   | 583/869 [00:07<00:03, 73.69it/s]
 68%|██████▊   | 591/869 [00:07<00:03, 74.15it/s]
 69%|██████▉   | 600/869 [00:07<00:03, 77.09it/s]
 70%|███████   | 609/869 [00:07<00:03, 78.40it/s]
 71%|███████   | 617/869 [00:08<00:03, 77.00it/s]
 72%|███████▏  | 625/869 [00:08<00:03, 75.42it/s]
 73%|███████▎  | 633/869 [00:08<00:03, 75.64it/s]
 74%|███████▍  | 641/869 [00:08<00:03, 75.26it/s]
 75%|███████▍  | 650/869 [00:08<00:02, 78.48it/s]
 76%|███████▌  | 658/869 [00:08<00:02, 77.25it/s]
 77%|███████▋  | 667/869 [00:08<00:02, 78.48it/s]
 78%|███████▊  | 677/869 [00:08<00:02, 82.17it/s]
 79%|███████▉  | 686/869 [00:09<00:02, 70.63it/s]
 80%|███████▉  | 694/869 [00:09<00:02, 72.95it/s]
 81%|████████  | 702/869 [00:09<00:02, 72.96it/s]
 82%|████████▏ | 711/869 [00:09<00:02, 75.37it/s]
 83%|████████▎ | 719/869 [00:09<00:01, 75.52it/s]
 84%|████████▍ | 728/869 [00:09<00:01, 77.16it/s]
 85%|████████▍ | 736/869 [00:09<00:01, 77.53it/s]
 86%|████████▌ | 744/869 [00:09<00:01, 77.74it/s]
 87%|████████▋ | 752/869 [00:09<00:01, 74.45it/s]
 88%|████████▊ | 761/869 [00:09<00:01, 77.76it/s]
 88%|████████▊ | 769/869 [00:10<00:01, 73.83it/s]
 89%|████████▉ | 777/869 [00:10<00:01, 65.48it/s]
 90%|█████████ | 785/869 [00:10<00:01, 67.59it/s]
 91%|█████████▏| 794/869 [00:10<00:01, 71.22it/s]
 92%|█████████▏| 802/869 [00:10<00:00, 73.53it/s]
 93%|█████████▎| 810/869 [00:10<00:00, 74.76it/s]
 94%|█████████▍| 818/869 [00:10<00:00, 75.19it/s]
 95%|█████████▌| 826/869 [00:10<00:00, 75.37it/s]
 96%|█████████▌| 834/869 [00:10<00:00, 74.91it/s]
 97%|█████████▋| 843/869 [00:11<00:00, 77.97it/s]
 98%|█████████▊| 852/869 [00:11<00:00, 79.15it/s]
 99%|█████████▉| 860/869 [00:11<00:00, 74.82it/s]
+***** eval metrics *****
+  epoch                   =       10.0
+  eval_accuracy           =     0.9988
+  eval_f1                 =     0.9335
+  eval_loss               =     0.0076
+  eval_precision          =      0.925
+  eval_recall             =     0.9422
+  eval_runtime            = 0:00:15.18
+  eval_samples            =       6946
+  eval_samples_per_second =    457.519
+  eval_steps_per_second   =     57.239
+09/09/2024 13:27:37 - INFO - __main__ -   *** Predict ***
+[INFO|trainer.py:811] 2024-09-09 13:27:37,532 >> 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:3819] 2024-09-09 13:27:37,534 >> 
+***** Running Prediction *****
+[INFO|trainer.py:3821] 2024-09-09 13:27:37,534 >>   Num examples = 14715
+[INFO|trainer.py:3824] 2024-09-09 13:27:37,534 >>   Batch size = 8
+
  0%|          | 0/1840 [00:00<?, ?it/s]
  1%|          | 11/1840 [00:00<00:18, 98.85it/s]
  1%|          | 21/1840 [00:00<00:22, 79.68it/s]
  2%|▏         | 30/1840 [00:00<00:22, 79.32it/s]
  2%|▏         | 39/1840 [00:00<00:22, 79.99it/s]
  3%|▎         | 48/1840 [00:00<00:22, 81.01it/s]
  3%|▎         | 57/1840 [00:00<00:22, 80.70it/s]
  4%|▎         | 66/1840 [00:00<00:22, 78.15it/s]
  4%|▍         | 74/1840 [00:00<00:22, 78.25it/s]
  5%|▍         | 83/1840 [00:01<00:21, 80.09it/s]
  5%|▌         | 92/1840 [00:01<00:21, 79.74it/s]
  6%|▌         | 102/1840 [00:01<00:20, 83.78it/s]
  6%|▌         | 111/1840 [00:01<00:20, 84.96it/s]
  7%|▋         | 120/1840 [00:01<00:20, 83.45it/s]
  7%|▋         | 129/1840 [00:01<00:21, 79.72it/s]
  8%|▊         | 138/1840 [00:01<00:20, 81.35it/s]
  8%|▊         | 147/1840 [00:01<00:20, 82.23it/s]
  8%|▊         | 156/1840 [00:01<00:21, 78.29it/s]
  9%|▉         | 164/1840 [00:02<00:21, 76.51it/s]
  9%|▉         | 173/1840 [00:02<00:21, 78.43it/s]
 10%|▉         | 183/1840 [00:02<00:19, 82.86it/s]
 10%|█         | 192/1840 [00:02<00:19, 84.17it/s]
 11%|█         | 201/1840 [00:02<00:19, 82.31it/s]
 11%|█▏        | 210/1840 [00:02<00:20, 79.83it/s]
 12%|█▏        | 219/1840 [00:02<00:20, 80.69it/s]
 12%|█▏        | 228/1840 [00:02<00:19, 82.07it/s]
 13%|█▎        | 237/1840 [00:02<00:21, 76.28it/s]
 13%|█▎        | 246/1840 [00:03<00:20, 78.24it/s]
 14%|█▍        | 254/1840 [00:03<00:20, 78.47it/s]
 14%|█▍        | 262/1840 [00:03<00:20, 78.54it/s]
 15%|█▍        | 270/1840 [00:03<00:20, 78.47it/s]
 15%|█▌        | 280/1840 [00:03<00:18, 83.71it/s]
 16%|█▌        | 290/1840 [00:03<00:18, 83.65it/s]
 16%|█▋        | 299/1840 [00:03<00:18, 85.16it/s]
 17%|█▋        | 308/1840 [00:03<00:18, 83.08it/s]
 17%|█▋        | 317/1840 [00:03<00:17, 84.68it/s]
 18%|█▊        | 326/1840 [00:04<00:18, 83.75it/s]
 18%|█▊        | 335/1840 [00:04<00:18, 82.34it/s]
 19%|█▉        | 345/1840 [00:04<00:17, 84.84it/s]
 19%|█▉        | 354/1840 [00:04<00:18, 80.58it/s]
 20%|█▉        | 363/1840 [00:04<00:18, 80.45it/s]
 20%|██        | 373/1840 [00:04<00:17, 82.94it/s]
 21%|██        | 382/1840 [00:04<00:17, 84.75it/s]
 21%|██▏       | 391/1840 [00:04<00:17, 83.12it/s]
 22%|██▏       | 400/1840 [00:04<00:17, 81.50it/s]
 22%|██▏       | 409/1840 [00:05<00:17, 82.12it/s]
 23%|██▎       | 418/1840 [00:05<00:17, 79.95it/s]
 23%|██▎       | 427/1840 [00:05<00:17, 80.56it/s]
 24%|██▎       | 436/1840 [00:05<00:17, 79.79it/s]
 24%|██▍       | 445/1840 [00:05<00:17, 80.57it/s]
 25%|██▍       | 454/1840 [00:05<00:17, 80.30it/s]
 25%|██▌       | 463/1840 [00:05<00:16, 82.64it/s]
 26%|██▌       | 472/1840 [00:05<00:16, 84.64it/s]
 26%|██▌       | 481/1840 [00:05<00:17, 76.17it/s]
 27%|██▋       | 490/1840 [00:06<00:17, 79.25it/s]
 27%|██▋       | 499/1840 [00:06<00:17, 76.61it/s]
 28%|██▊       | 508/1840 [00:06<00:17, 77.81it/s]
 28%|██▊       | 518/1840 [00:06<00:16, 81.49it/s]
 29%|██▊       | 527/1840 [00:06<00:15, 82.26it/s]
 29%|██▉       | 536/1840 [00:06<00:16, 81.29it/s]
 30%|██▉       | 545/1840 [00:06<00:15, 81.42it/s]
 30%|███       | 554/1840 [00:06<00:15, 82.78it/s]
 31%|███       | 563/1840 [00:06<00:15, 80.10it/s]
 31%|███       | 573/1840 [00:07<00:15, 83.39it/s]
 32%|███▏      | 583/1840 [00:07<00:14, 85.46it/s]
 32%|███▏      | 592/1840 [00:07<00:15, 82.90it/s]
 33%|███▎      | 601/1840 [00:07<00:15, 81.01it/s]
 33%|███▎      | 611/1840 [00:07<00:14, 82.55it/s]
 34%|███▎      | 620/1840 [00:07<00:15, 79.88it/s]
 34%|███▍      | 629/1840 [00:07<00:15, 79.21it/s]
 35%|███▍      | 639/1840 [00:07<00:14, 83.45it/s]
 35%|███▌      | 649/1840 [00:07<00:13, 85.57it/s]
 36%|███▌      | 658/1840 [00:08<00:14, 81.64it/s]
 36%|███▋      | 667/1840 [00:08<00:14, 80.85it/s]
 37%|███▋      | 677/1840 [00:08<00:13, 84.92it/s]
 37%|███▋      | 686/1840 [00:08<00:13, 82.87it/s]
 38%|███▊      | 695/1840 [00:08<00:13, 83.04it/s]
 38%|███▊      | 705/1840 [00:08<00:13, 86.04it/s]
 39%|███▉      | 714/1840 [00:08<00:13, 84.27it/s]
 39%|███▉      | 724/1840 [00:08<00:12, 87.27it/s]
 40%|███▉      | 734/1840 [00:08<00:12, 89.45it/s]
 40%|████      | 743/1840 [00:09<00:12, 89.01it/s]
 41%|████      | 752/1840 [00:09<00:12, 86.83it/s]
 41%|████▏     | 762/1840 [00:09<00:12, 89.06it/s]
 42%|████▏     | 771/1840 [00:09<00:12, 87.52it/s]
 42%|████▏     | 780/1840 [00:09<00:12, 87.85it/s]
 43%|████▎     | 789/1840 [00:09<00:12, 84.43it/s]
 43%|████▎     | 798/1840 [00:09<00:12, 84.40it/s]
 44%|████▍     | 807/1840 [00:09<00:12, 82.54it/s]
 44%|████▍     | 816/1840 [00:09<00:12, 83.63it/s]
 45%|████▍     | 825/1840 [00:10<00:12, 84.00it/s]
 45%|████▌     | 835/1840 [00:10<00:11, 87.17it/s]
 46%|████▌     | 844/1840 [00:10<00:12, 82.78it/s]
 46%|████▋     | 853/1840 [00:10<00:11, 84.72it/s]
 47%|████▋     | 863/1840 [00:10<00:11, 86.79it/s]
 47%|████▋     | 873/1840 [00:10<00:11, 86.72it/s]
 48%|████▊     | 882/1840 [00:10<00:11, 82.75it/s]
 48%|████▊     | 891/1840 [00:10<00:11, 82.52it/s]
 49%|████▉     | 900/1840 [00:10<00:11, 81.40it/s]
 49%|████▉     | 910/1840 [00:11<00:10, 85.07it/s]
 50%|████▉     | 919/1840 [00:11<00:10, 86.09it/s]
 50%|█████     | 929/1840 [00:11<00:10, 87.65it/s]
 51%|█████     | 938/1840 [00:11<00:10, 84.08it/s]
 51%|█████▏    | 947/1840 [00:11<00:10, 84.62it/s]
 52%|█████▏    | 956/1840 [00:11<00:10, 81.28it/s]
 52%|█████▏    | 965/1840 [00:11<00:10, 81.73it/s]
 53%|█████▎    | 975/1840 [00:11<00:10, 84.56it/s]
 53%|█████▎    | 984/1840 [00:11<00:10, 81.54it/s]
 54%|█████▍    | 994/1840 [00:12<00:10, 84.11it/s]
 55%|█████▍    | 1003/1840 [00:12<00:10, 82.91it/s]
 55%|█████▌    | 1012/1840 [00:12<00:10, 81.54it/s]
 55%|█████▌    | 1021/1840 [00:12<00:09, 82.96it/s]
 56%|█████▌    | 1031/1840 [00:12<00:09, 86.38it/s]
 57%|█████▋    | 1041/1840 [00:12<00:09, 86.05it/s]
 57%|█████▋    | 1050/1840 [00:12<00:09, 85.53it/s]
 58%|█████▊    | 1059/1840 [00:12<00:09, 84.24it/s]
 58%|█████▊    | 1068/1840 [00:12<00:09, 83.89it/s]
 59%|█████▊    | 1078/1840 [00:13<00:08, 86.49it/s]
 59%|█████▉    | 1088/1840 [00:13<00:08, 88.79it/s]
 60%|█████▉    | 1098/1840 [00:13<00:08, 90.45it/s]
 60%|██████    | 1108/1840 [00:13<00:08, 90.80it/s]
 61%|██████    | 1118/1840 [00:13<00:08, 86.64it/s]
 61%|██████▏   | 1127/1840 [00:13<00:08, 85.05it/s]
 62%|██████▏   | 1137/1840 [00:13<00:08, 86.61it/s]
 62%|██████▏   | 1146/1840 [00:13<00:07, 87.34it/s]
 63%|██████▎   | 1155/1840 [00:13<00:07, 87.26it/s]
 63%|██████▎   | 1165/1840 [00:14<00:07, 88.13it/s]
 64%|██████▍   | 1174/1840 [00:14<00:07, 83.58it/s]
 64%|██████▍   | 1183/1840 [00:14<00:07, 84.91it/s]
 65%|██████▍   | 1192/1840 [00:14<00:07, 81.30it/s]
 65%|██████▌   | 1202/1840 [00:14<00:07, 85.01it/s]
 66%|██████▌   | 1211/1840 [00:14<00:07, 84.77it/s]
 66%|██████▋   | 1220/1840 [00:14<00:07, 80.78it/s]
 67%|██████▋   | 1229/1840 [00:14<00:07, 77.99it/s]
 67%|██████▋   | 1239/1840 [00:14<00:07, 82.43it/s]
 68%|██████▊   | 1249/1840 [00:15<00:07, 84.02it/s]
 68%|██████▊   | 1259/1840 [00:15<00:06, 86.05it/s]
 69%|██████▉   | 1268/1840 [00:15<00:07, 81.33it/s]
 69%|██████▉   | 1277/1840 [00:15<00:06, 81.61it/s]
 70%|██████▉   | 1286/1840 [00:15<00:06, 83.73it/s]
 70%|███████   | 1295/1840 [00:15<00:06, 85.09it/s]
 71%|███████   | 1305/1840 [00:15<00:06, 87.44it/s]
 71%|███████▏  | 1315/1840 [00:15<00:05, 89.24it/s]
 72%|███████▏  | 1325/1840 [00:15<00:05, 90.82it/s]
 73%|███████▎  | 1335/1840 [00:16<00:05, 89.40it/s]
 73%|███████▎  | 1344/1840 [00:16<00:05, 88.30it/s]
 74%|███████▎  | 1353/1840 [00:16<00:05, 86.55it/s]
 74%|███████▍  | 1362/1840 [00:16<00:05, 85.78it/s]
 75%|███████▍  | 1372/1840 [00:16<00:05, 87.53it/s]
 75%|███████▌  | 1381/1840 [00:16<00:05, 83.66it/s]
 76%|███████▌  | 1391/1840 [00:16<00:05, 86.29it/s]
 76%|███████▌  | 1400/1840 [00:16<00:05, 84.99it/s]
 77%|███████▋  | 1410/1840 [00:16<00:04, 87.43it/s]
 77%|███████▋  | 1420/1840 [00:17<00:04, 89.10it/s]
 78%|███████▊  | 1429/1840 [00:17<00:04, 88.03it/s]
 78%|███████▊  | 1438/1840 [00:17<00:04, 83.40it/s]
 79%|███████▊  | 1447/1840 [00:17<00:04, 80.09it/s]
 79%|███████▉  | 1456/1840 [00:17<00:04, 79.92it/s]
 80%|███████▉  | 1466/1840 [00:17<00:04, 82.65it/s]
 80%|████████  | 1475/1840 [00:17<00:04, 83.62it/s]
 81%|████████  | 1484/1840 [00:17<00:04, 78.30it/s]
 81%|████████  | 1493/1840 [00:17<00:04, 80.11it/s]
 82%|████████▏ | 1502/1840 [00:18<00:04, 75.31it/s]
 82%|████████▏ | 1511/1840 [00:18<00:04, 75.48it/s]
 83%|████████▎ | 1520/1840 [00:18<00:04, 77.35it/s]
 83%|████████▎ | 1529/1840 [00:18<00:03, 79.48it/s]
 84%|████████▎ | 1538/1840 [00:18<00:03, 77.28it/s]
 84%|████████▍ | 1547/1840 [00:18<00:03, 78.65it/s]
 85%|████████▍ | 1556/1840 [00:18<00:03, 79.70it/s]
 85%|████████▌ | 1565/1840 [00:18<00:03, 79.11it/s]
 86%|████████▌ | 1574/1840 [00:18<00:03, 79.32it/s]
 86%|████████▌ | 1582/1840 [00:19<00:03, 76.10it/s]
 86%|████████▋ | 1591/1840 [00:19<00:03, 77.80it/s]
 87%|████████▋ | 1599/1840 [00:19<00:03, 76.77it/s]
 87%|████████▋ | 1608/1840 [00:19<00:02, 78.00it/s]
 88%|████████▊ | 1617/1840 [00:19<00:02, 80.43it/s]
 88%|████████▊ | 1626/1840 [00:19<00:02, 81.36it/s]
 89%|████████▉ | 1635/1840 [00:19<00:02, 71.94it/s]
 89%|████████▉ | 1643/1840 [00:19<00:02, 73.05it/s]
 90%|████████▉ | 1652/1840 [00:20<00:02, 75.07it/s]
 90%|█████████ | 1661/1840 [00:20<00:02, 78.13it/s]
 91%|█████████ | 1670/1840 [00:20<00:02, 78.46it/s]
 91%|█████████ | 1678/1840 [00:20<00:02, 78.03it/s]
 92%|█████████▏| 1687/1840 [00:20<00:01, 79.05it/s]
 92%|█████████▏| 1697/1840 [00:20<00:01, 82.97it/s]
 93%|█████████▎| 1706/1840 [00:20<00:01, 80.42it/s]
 93%|██���██████▎| 1715/1840 [00:20<00:01, 78.43it/s]
 94%|█████████▎| 1724/1840 [00:20<00:01, 80.51it/s]
 94%|█████████▍| 1733/1840 [00:21<00:01, 77.32it/s]
 95%|█████████▍| 1742/1840 [00:21<00:01, 80.56it/s]
 95%|█████████▌| 1752/1840 [00:21<00:01, 83.60it/s]
 96%|█████████▌| 1761/1840 [00:21<00:00, 84.69it/s]
 96%|█████████▌| 1770/1840 [00:21<00:00, 82.30it/s]
 97%|█████████▋| 1779/1840 [00:21<00:00, 82.74it/s]
 97%|█████████▋| 1788/1840 [00:21<00:00, 81.59it/s]
 98%|█████████▊| 1797/1840 [00:21<00:00, 83.44it/s]
 98%|█████████▊| 1806/1840 [00:21<00:00, 79.67it/s]
 99%|█████████▊| 1815/1840 [00:22<00:00, 77.02it/s]
 99%|█████████▉| 1824/1840 [00:22<00:00, 79.80it/s]
+[INFO|trainer.py:3503] 2024-09-09 13:28:06,818 >> Saving model checkpoint to /content/dissertation/scripts/ner/output
+[INFO|configuration_utils.py:472] 2024-09-09 13:28:06,820 >> Configuration saved in /content/dissertation/scripts/ner/output/config.json
+[INFO|modeling_utils.py:2799] 2024-09-09 13:28:08,001 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
+[INFO|tokenization_utils_base.py:2684] 2024-09-09 13:28:08,002 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2693] 2024-09-09 13:28:08,002 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+***** predict metrics *****
+  predict_accuracy           =     0.9987
+  predict_f1                 =     0.9203
+  predict_loss               =     0.0078
+  predict_precision          =     0.8938
+  predict_recall             =     0.9483
+  predict_runtime            = 0:00:28.74
+  predict_samples_per_second =    511.904
+  predict_steps_per_second   =      64.01
+

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 9.967845659163988,
-    "total_flos": 4644619911314910.0,
-    "train_loss": 0.050868147861573,
-    "train_runtime": 855.1929,
-    "train_samples": 9929,
-    "train_samples_per_second": 116.102,
-    "train_steps_per_second": 1.812
+    "epoch": 10.0,
+    "total_flos": 1.394810359803495e+16,
+    "train_loss": 0.0028968164414402532,
+    "train_runtime": 2196.5741,
+    "train_samples": 32447,
+    "train_samples_per_second": 147.716,
+    "train_steps_per_second": 2.308
 }

trainer_state.json

@@ -1,166 +1,215 @@
 {
-  "best_metric": 0.6946045049764275,
-  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-1088",
-  "epoch": 9.967845659163988,
+  "best_metric": 0.9335180055401663,
+  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-4056",
+  "epoch": 10.0,
   "eval_steps": 500,
-  "global_step": 1550,
+  "global_step": 5070,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.9967845659163987,
-      "eval_accuracy": 0.9435978055118868,
-      "eval_f1": 0.593056643174106,
-      "eval_loss": 0.14849668741226196,
-      "eval_precision": 0.5668662674650699,
-      "eval_recall": 0.6217843459222769,
-      "eval_runtime": 5.8633,
-      "eval_samples_per_second": 429.618,
-      "eval_steps_per_second": 53.724,
-      "step": 155
+      "epoch": 0.9861932938856016,
+      "grad_norm": 0.16175590455532074,
+      "learning_rate": 4.5069033530571994e-05,
+      "loss": 0.0183,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9985418469009014,
+      "eval_f1": 0.9170464904284412,
+      "eval_loss": 0.005470008589327335,
+      "eval_precision": 0.8974130240856378,
+      "eval_recall": 0.9375582479030755,
+      "eval_runtime": 15.2218,
+      "eval_samples_per_second": 456.319,
+      "eval_steps_per_second": 57.089,
+      "step": 507
+    },
+    {
+      "epoch": 1.972386587771203,
+      "grad_norm": 0.13494741916656494,
+      "learning_rate": 4.0138067061143986e-05,
+      "loss": 0.0043,
+      "step": 1000
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9408546953704001,
-      "eval_f1": 0.6299060919817,
-      "eval_loss": 0.16094118356704712,
-      "eval_precision": 0.5623387790197765,
-      "eval_recall": 0.715927750410509,
-      "eval_runtime": 5.9155,
-      "eval_samples_per_second": 425.831,
-      "eval_steps_per_second": 53.25,
-      "step": 311
-    },
-    {
-      "epoch": 2.996784565916399,
-      "eval_accuracy": 0.9486669447207161,
-      "eval_f1": 0.6676790685902303,
-      "eval_loss": 0.1634686440229416,
-      "eval_precision": 0.6209981167608286,
-      "eval_recall": 0.7219485495347564,
-      "eval_runtime": 5.9009,
-      "eval_samples_per_second": 426.884,
-      "eval_steps_per_second": 53.382,
-      "step": 466
-    },
-    {
-      "epoch": 3.215434083601286,
-      "grad_norm": 0.9288749098777771,
-      "learning_rate": 3.387096774193548e-05,
-      "loss": 0.1246,
-      "step": 500
+      "eval_accuracy": 0.9985976613735942,
+      "eval_f1": 0.9208103130755064,
+      "eval_loss": 0.005941031500697136,
+      "eval_precision": 0.9099181073703366,
+      "eval_recall": 0.9319664492078286,
+      "eval_runtime": 15.129,
+      "eval_samples_per_second": 459.117,
+      "eval_steps_per_second": 57.439,
+      "step": 1014
+    },
+    {
+      "epoch": 2.9585798816568047,
+      "grad_norm": 0.14869354665279388,
+      "learning_rate": 3.520710059171598e-05,
+      "loss": 0.0022,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.998513939664555,
+      "eval_f1": 0.9155963302752294,
+      "eval_loss": 0.005716267507523298,
+      "eval_precision": 0.9015356820234869,
+      "eval_recall": 0.9301025163094129,
+      "eval_runtime": 15.1542,
+      "eval_samples_per_second": 458.356,
+      "eval_steps_per_second": 57.344,
+      "step": 1521
+    },
+    {
+      "epoch": 3.9447731755424065,
+      "grad_norm": 0.15148845314979553,
+      "learning_rate": 3.027613412228797e-05,
+      "loss": 0.0018,
+      "step": 2000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9492604831723828,
-      "eval_f1": 0.6711919630735812,
-      "eval_loss": 0.20466655492782593,
-      "eval_precision": 0.665948275862069,
-      "eval_recall": 0.6765188834154351,
-      "eval_runtime": 5.9155,
-      "eval_samples_per_second": 425.831,
-      "eval_steps_per_second": 53.25,
-      "step": 622
-    },
-    {
-      "epoch": 4.996784565916399,
-      "eval_accuracy": 0.9479771567904007,
-      "eval_f1": 0.6827731092436975,
-      "eval_loss": 0.2133806049823761,
-      "eval_precision": 0.6562342251388188,
-      "eval_recall": 0.7115489874110563,
-      "eval_runtime": 5.8759,
-      "eval_samples_per_second": 428.701,
-      "eval_steps_per_second": 53.609,
-      "step": 777
+      "eval_accuracy": 0.9985767309463344,
+      "eval_f1": 0.9227166276346604,
+      "eval_loss": 0.0072021945379674435,
+      "eval_precision": 0.9274952919020716,
+      "eval_recall": 0.9179869524697111,
+      "eval_runtime": 15.3453,
+      "eval_samples_per_second": 452.646,
+      "eval_steps_per_second": 56.63,
+      "step": 2028
+    },
+    {
+      "epoch": 4.930966469428008,
+      "grad_norm": 0.17791983485221863,
+      "learning_rate": 2.5345167652859964e-05,
+      "loss": 0.0009,
+      "step": 2500
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9986604526553735,
+      "eval_f1": 0.9215236346948141,
+      "eval_loss": 0.0063728527165949345,
+      "eval_precision": 0.9077757685352622,
+      "eval_recall": 0.9356943150046598,
+      "eval_runtime": 15.1945,
+      "eval_samples_per_second": 457.14,
+      "eval_steps_per_second": 57.192,
+      "step": 2535
+    },
+    {
+      "epoch": 5.9171597633136095,
+      "grad_norm": 0.027111981064081192,
+      "learning_rate": 2.0414201183431952e-05,
+      "loss": 0.0007,
+      "step": 3000
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.949388815810581,
-      "eval_f1": 0.6795913020696882,
-      "eval_loss": 0.2258971482515335,
-      "eval_precision": 0.6517587939698493,
-      "eval_recall": 0.7099069512862616,
-      "eval_runtime": 5.9106,
-      "eval_samples_per_second": 426.187,
-      "eval_steps_per_second": 53.294,
-      "step": 933
-    },
-    {
-      "epoch": 6.430868167202572,
-      "grad_norm": 0.18494442105293274,
-      "learning_rate": 1.774193548387097e-05,
-      "loss": 0.0242,
-      "step": 1000
+      "eval_accuracy": 0.9987302207462395,
+      "eval_f1": 0.9274826789838336,
+      "eval_loss": 0.006401057820767164,
+      "eval_precision": 0.9194139194139194,
+      "eval_recall": 0.9356943150046598,
+      "eval_runtime": 15.0971,
+      "eval_samples_per_second": 460.088,
+      "eval_steps_per_second": 57.561,
+      "step": 3042
     },
     {
-      "epoch": 6.996784565916399,
-      "eval_accuracy": 0.9496615226667522,
-      "eval_f1": 0.6946045049764275,
-      "eval_loss": 0.24502409994602203,
-      "eval_precision": 0.6659969864389754,
-      "eval_recall": 0.7257799671592775,
-      "eval_runtime": 5.8918,
-      "eval_samples_per_second": 427.542,
-      "eval_steps_per_second": 53.464,
-      "step": 1088
+      "epoch": 6.903353057199211,
+      "grad_norm": 0.0044996230863034725,
+      "learning_rate": 1.5483234714003947e-05,
+      "loss": 0.0004,
+      "step": 3500
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9987651047916725,
+      "eval_f1": 0.9332096474953618,
+      "eval_loss": 0.00720419455319643,
+      "eval_precision": 0.9289012003693444,
+      "eval_recall": 0.9375582479030755,
+      "eval_runtime": 15.1686,
+      "eval_samples_per_second": 457.919,
+      "eval_steps_per_second": 57.289,
+      "step": 3549
+    },
+    {
+      "epoch": 7.889546351084813,
+      "grad_norm": 0.0005134351667948067,
+      "learning_rate": 1.0552268244575937e-05,
+      "loss": 0.0004,
+      "step": 4000
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9490679842150855,
-      "eval_f1": 0.6841015018125324,
-      "eval_loss": 0.26501980423927307,
-      "eval_precision": 0.6491400491400492,
-      "eval_recall": 0.7230432402846196,
-      "eval_runtime": 5.9184,
-      "eval_samples_per_second": 425.624,
-      "eval_steps_per_second": 53.224,
-      "step": 1244
-    },
-    {
-      "epoch": 8.996784565916398,
-      "eval_accuracy": 0.9497577721454009,
-      "eval_f1": 0.687797147385103,
-      "eval_loss": 0.27453720569610596,
-      "eval_precision": 0.664624808575804,
-      "eval_recall": 0.7126436781609196,
-      "eval_runtime": 5.9283,
-      "eval_samples_per_second": 424.913,
-      "eval_steps_per_second": 53.135,
-      "step": 1399
-    },
-    {
-      "epoch": 9.646302250803858,
-      "grad_norm": 0.35425594449043274,
-      "learning_rate": 1.6129032258064516e-06,
-      "loss": 0.0083,
-      "step": 1500
+      "eval_accuracy": 0.998772081600759,
+      "eval_f1": 0.9335180055401663,
+      "eval_loss": 0.007628325838595629,
+      "eval_precision": 0.9249771271729186,
+      "eval_recall": 0.9422180801491147,
+      "eval_runtime": 15.1289,
+      "eval_samples_per_second": 459.122,
+      "eval_steps_per_second": 57.44,
+      "step": 4056
+    },
+    {
+      "epoch": 8.875739644970414,
+      "grad_norm": 0.0011633747490122914,
+      "learning_rate": 5.621301775147929e-06,
+      "loss": 0.0003,
+      "step": 4500
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.9987162671280663,
+      "eval_f1": 0.9262672811059908,
+      "eval_loss": 0.00773056922480464,
+      "eval_precision": 0.9161349134001823,
+      "eval_recall": 0.9366262814538676,
+      "eval_runtime": 15.3347,
+      "eval_samples_per_second": 452.96,
+      "eval_steps_per_second": 56.669,
+      "step": 4563
+    },
+    {
+      "epoch": 9.861932938856016,
+      "grad_norm": 0.001042825635522604,
+      "learning_rate": 6.903353057199211e-07,
+      "loss": 0.0002,
+      "step": 5000
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.9987511511734993,
+      "eval_f1": 0.92797783933518,
+      "eval_loss": 0.00769586768001318,
+      "eval_precision": 0.9194876486733761,
+      "eval_recall": 0.9366262814538676,
+      "eval_runtime": 15.2047,
+      "eval_samples_per_second": 456.832,
+      "eval_steps_per_second": 57.153,
+      "step": 5070
     },
     {
-      "epoch": 9.967845659163988,
-      "eval_accuracy": 0.950255061118419,
-      "eval_f1": 0.6896008403361344,
-      "eval_loss": 0.27744925022125244,
-      "eval_precision": 0.662796567390207,
-      "eval_recall": 0.7186644772851669,
-      "eval_runtime": 6.1493,
-      "eval_samples_per_second": 409.643,
-      "eval_steps_per_second": 51.226,
-      "step": 1550
-    },
-    {
-      "epoch": 9.967845659163988,
-      "step": 1550,
-      "total_flos": 4644619911314910.0,
-      "train_loss": 0.050868147861573,
-      "train_runtime": 855.1929,
-      "train_samples_per_second": 116.102,
-      "train_steps_per_second": 1.812
+      "epoch": 10.0,
+      "step": 5070,
+      "total_flos": 1.394810359803495e+16,
+      "train_loss": 0.0028968164414402532,
+      "train_runtime": 2196.5741,
+      "train_samples_per_second": 147.716,
+      "train_steps_per_second": 2.308
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1550,
+  "max_steps": 5070,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
@@ -176,7 +225,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4644619911314910.0,
+  "total_flos": 1.394810359803495e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null