End of training

README.md

@@ -2,9 +2,10 @@
 license: apache-2.0
 base_model: PlanTL-GOB-ES/bsc-bio-ehr-es
 tags:
+- token-classification
 - generated_from_trainer
 datasets:
-- distemist-ner
+- Rodrigo1771/distemist-ner
 metrics:
 - precision
 - recall
@@ -17,24 +18,24 @@ model-index:
       name: Token Classification
       type: token-classification
     dataset:
-      name: distemist-ner
-      type: distemist-ner
+      name: Rodrigo1771/distemist-ner
+      type: Rodrigo1771/distemist-ner
       config: DisTEMIST NER
       split: validation
       args: DisTEMIST NER
     metrics:
     - name: Precision
       type: precision
-      value: 0.7882031427920747
+      value: 0.7938948817994033
     - name: Recall
       type: recall
-      value: 0.8097800655124006
+      value: 0.8093121197941039
     - name: F1
       type: f1
-      value: 0.7988459319099828
+      value: 0.8015293708724366
     - name: Accuracy
       type: accuracy
-      value: 0.9766776058330014
+      value: 0.9767668584453568
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -42,13 +43,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 # output
 
-This model is a fine-tuned version of [PlanTL-GOB-ES/bsc-bio-ehr-es](https://huggingface.co/PlanTL-GOB-ES/bsc-bio-ehr-es) on the distemist-ner dataset.
+This model is a fine-tuned version of [PlanTL-GOB-ES/bsc-bio-ehr-es](https://huggingface.co/PlanTL-GOB-ES/bsc-bio-ehr-es) on the Rodrigo1771/distemist-ner dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1367
-- Precision: 0.7882
-- Recall: 0.8098
-- F1: 0.7988
-- Accuracy: 0.9767
+- Loss: 0.1294
+- Precision: 0.7939
+- Recall: 0.8093
+- F1: 0.8015
+- Accuracy: 0.9768
 
 ## Model description
 

all_results.json

@@ -1,26 +1,26 @@
 {
     "epoch": 9.988249118683902,
-    "eval_accuracy": 0.7932840841995413,
-    "eval_f1": 0.17333222008850296,
-    "eval_loss": 1.0502684116363525,
-    "eval_precision": 0.09532555790247038,
-    "eval_recall": 0.9540441176470589,
-    "eval_runtime": 14.6851,
+    "eval_accuracy": 0.9767668584453568,
+    "eval_f1": 0.8015293708724366,
+    "eval_loss": 0.1294233798980713,
+    "eval_precision": 0.7938948817994033,
+    "eval_recall": 0.8093121197941039,
+    "eval_runtime": 14.4843,
     "eval_samples": 6810,
-    "eval_samples_per_second": 463.735,
-    "eval_steps_per_second": 58.018,
-    "predict_accuracy": 0.8808544760053879,
-    "predict_f1": 0.24078298623886968,
-    "predict_loss": 0.62894606590271,
-    "predict_precision": 0.1378032345013477,
-    "predict_recall": 0.9528246942341293,
-    "predict_runtime": 29.8715,
-    "predict_samples_per_second": 489.229,
-    "predict_steps_per_second": 61.162,
-    "total_flos": 1.2649810588547778e+16,
-    "train_loss": 0.10639642311544979,
-    "train_runtime": 1208.2019,
+    "eval_samples_per_second": 470.164,
+    "eval_steps_per_second": 58.822,
+    "predict_accuracy": 0.9765722208397027,
+    "predict_f1": 0.8011714125490684,
+    "predict_loss": 0.12331698834896088,
+    "predict_precision": 0.7871923594955308,
+    "predict_recall": 0.8156559248921593,
+    "predict_runtime": 29.3641,
+    "predict_samples_per_second": 497.683,
+    "predict_steps_per_second": 62.219,
+    "total_flos": 1.2649124434987926e+16,
+    "train_loss": 0.02590363489880281,
+    "train_runtime": 1203.9865,
     "train_samples": 27229,
-    "train_samples_per_second": 225.368,
-    "train_steps_per_second": 3.518
+    "train_samples_per_second": 226.157,
+    "train_steps_per_second": 3.53
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 9.988249118683902,
-    "eval_accuracy": 0.7932840841995413,
-    "eval_f1": 0.17333222008850296,
-    "eval_loss": 1.0502684116363525,
-    "eval_precision": 0.09532555790247038,
-    "eval_recall": 0.9540441176470589,
-    "eval_runtime": 14.6851,
+    "eval_accuracy": 0.9767668584453568,
+    "eval_f1": 0.8015293708724366,
+    "eval_loss": 0.1294233798980713,
+    "eval_precision": 0.7938948817994033,
+    "eval_recall": 0.8093121197941039,
+    "eval_runtime": 14.4843,
     "eval_samples": 6810,
-    "eval_samples_per_second": 463.735,
-    "eval_steps_per_second": 58.018
+    "eval_samples_per_second": 470.164,
+    "eval_steps_per_second": 58.822
 }

predict_results.json

@@ -1,10 +1,10 @@
 {
-    "predict_accuracy": 0.8808544760053879,
-    "predict_f1": 0.24078298623886968,
-    "predict_loss": 0.62894606590271,
-    "predict_precision": 0.1378032345013477,
-    "predict_recall": 0.9528246942341293,
-    "predict_runtime": 29.8715,
-    "predict_samples_per_second": 489.229,
-    "predict_steps_per_second": 61.162
+    "predict_accuracy": 0.9765722208397027,
+    "predict_f1": 0.8011714125490684,
+    "predict_loss": 0.12331698834896088,
+    "predict_precision": 0.7871923594955308,
+    "predict_recall": 0.8156559248921593,
+    "predict_runtime": 29.3641,
+    "predict_samples_per_second": 497.683,
+    "predict_steps_per_second": 62.219
 }

tb/events.out.tfevents.1725050776.6b97e535edda.20735.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:695bf5e0a27bceb5f4eb0bc971d0b2f97725e31e30a7b30d5ffedc4ae0991cc9
+size 560

train.log

@@ -1509,3 +1509,51 @@ Training completed. Do not forget to share your model on huggingface.co/models =
 {'eval_loss': 0.13674204051494598, 'eval_precision': 0.7882031427920747, 'eval_recall': 0.8097800655124006, 'eval_f1': 0.7988459319099828, 'eval_accuracy': 0.9766776058330014, 'eval_runtime': 14.8828, 'eval_samples_per_second': 457.575, 'eval_steps_per_second': 57.247, 'epoch': 9.99}
 {'train_runtime': 1203.9865, 'train_samples_per_second': 226.157, 'train_steps_per_second': 3.53, 'train_loss': 0.02590363489880281, 'epoch': 9.99}
 
+***** train metrics *****
+  epoch                    =     9.9882
+  total_flos               = 11780415GF
+  train_loss               =     0.0259
+  train_runtime            = 0:20:03.98
+  train_samples            =      27229
+  train_samples_per_second =    226.157
+  train_steps_per_second   =       3.53
+08/30/2024 20:46:02 - INFO - __main__ -   *** Evaluate ***
+[INFO|trainer.py:805] 2024-08-30 20:46:02,283 >> The following columns in the evaluation set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: tokens, ner_tags, id. If tokens, ner_tags, id are not expected by `RobertaForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3788] 2024-08-30 20:46:02,286 >> 
+***** Running Evaluation *****
+[INFO|trainer.py:3790] 2024-08-30 20:46:02,286 >>   Num examples = 6810
+[INFO|trainer.py:3793] 2024-08-30 20:46:02,286 >>   Batch size = 8
+
  0%|          | 0/852 [00:00<?, ?it/s]
  1%|          | 10/852 [00:00<00:09, 91.43it/s]
  2%|▏         | 20/852 [00:00<00:10, 81.20it/s]
  3%|▎         | 29/852 [00:00<00:10, 75.61it/s]
  4%|▍         | 37/852 [00:00<00:10, 76.54it/s]
  5%|▌         | 46/852 [00:00<00:10, 79.06it/s]
  6%|▋         | 55/852 [00:00<00:09, 80.42it/s]
  8%|▊         | 64/852 [00:00<00:09, 80.69it/s]
  9%|▊         | 73/852 [00:00<00:09, 78.84it/s]
 10%|▉         | 81/852 [00:01<00:09, 78.62it/s]
 11%|█         | 90/852 [00:01<00:09, 79.21it/s]
 12%|█▏        | 98/852 [00:01<00:09, 79.05it/s]
 13%|█▎        | 107/852 [00:01<00:09, 78.52it/s]
 13%|█▎        | 115/852 [00:01<00:09, 78.70it/s]
 15%|█▍        | 124/852 [00:01<00:09, 80.68it/s]
 16%|█▌        | 133/852 [00:01<00:09, 76.87it/s]
 17%|█▋        | 141/852 [00:01<00:09, 77.17it/s]
 17%|█▋        | 149/852 [00:01<00:09, 76.66it/s]
 19%|█▊        | 158/852 [00:02<00:08, 78.34it/s]
 20%|█▉        | 167/852 [00:02<00:08, 79.21it/s]
 21%|██        | 175/852 [00:02<00:08, 79.28it/s]
 22%|██▏       | 184/852 [00:02<00:08, 79.85it/s]
 23%|██▎       | 192/852 [00:02<00:08, 77.56it/s]
 24%|██▎       | 201/852 [00:02<00:08, 79.28it/s]
 25%|██▍       | 209/852 [00:02<00:08, 78.02it/s]
 25%|██▌       | 217/852 [00:02<00:08, 78.36it/s]
 26%|██▋       | 225/852 [00:02<00:07, 78.57it/s]
 27%|██▋       | 234/852 [00:02<00:07, 79.89it/s]
 28%|██▊       | 242/852 [00:03<00:07, 76.69it/s]
 29%|██▉       | 250/852 [00:03<00:07, 77.50it/s]
 30%|███       | 259/852 [00:03<00:07, 78.25it/s]
 31%|███▏      | 267/852 [00:03<00:07, 77.79it/s]
 32%|███▏      | 276/852 [00:03<00:07, 79.12it/s]
 33%|███▎      | 285/852 [00:03<00:07, 79.97it/s]
 34%|███▍      | 293/852 [00:03<00:07, 79.57it/s]
 35%|███▌      | 302/852 [00:03<00:06, 80.12it/s]
 37%|███▋      | 311/852 [00:03<00:06, 81.11it/s]
 38%|███▊      | 320/852 [00:04<00:06, 77.46it/s]
 38%|███▊      | 328/852 [00:04<00:06, 76.62it/s]
 39%|███▉      | 336/852 [00:04<00:06, 76.07it/s]
 40%|████      | 344/852 [00:04<00:06, 77.03it/s]
 41%|████▏     | 352/852 [00:04<00:06, 77.09it/s]
 42%|████▏     | 360/852 [00:04<00:06, 75.43it/s]
 43%|████▎     | 368/852 [00:04<00:06, 76.71it/s]
 44%|████▍     | 376/852 [00:04<00:06, 77.23it/s]
 45%|████▌     | 384/852 [00:04<00:06, 76.84it/s]
 46%|████▌     | 392/852 [00:05<00:05, 77.70it/s]
 47%|████▋     | 400/852 [00:05<00:05, 77.92it/s]
 48%|████▊     | 408/852 [00:05<00:05, 76.03it/s]
 49%|████▉     | 417/852 [00:05<00:05, 77.02it/s]
 50%|████▉     | 425/852 [00:05<00:05, 75.06it/s]
 51%|█████     | 434/852 [00:05<00:05, 77.48it/s]
 52%|█████▏    | 443/852 [00:05<00:05, 77.92it/s]
 53%|█████▎    | 452/852 [00:05<00:05, 78.84it/s]
 54%|█████▍    | 460/852 [00:05<00:04, 78.43it/s]
 55%|█████▍    | 468/852 [00:05<00:04, 76.81it/s]
 56%|█████▌    | 476/852 [00:06<00:05, 74.29it/s]
 57%|█████▋    | 484/852 [00:06<00:04, 74.53it/s]
 58%|█████▊    | 493/852 [00:06<00:04, 76.53it/s]
 59%|█████▉    | 502/852 [00:06<00:04, 77.44it/s]
 60%|█████▉    | 510/852 [00:06<00:04, 77.79it/s]
 61%|██████    | 519/852 [00:06<00:04, 79.51it/s]
 62%|██████▏   | 527/852 [00:06<00:04, 77.59it/s]
 63%|██████▎   | 536/852 [00:06<00:03, 79.53it/s]
 64%|██████▍   | 545/852 [00:06<00:03, 80.00it/s]
 65%|██████▌   | 554/852 [00:07<00:03, 77.58it/s]
 66%|██████▌   | 562/852 [00:07<00:03, 76.79it/s]
 67%|██████▋   | 571/852 [00:07<00:03, 78.19it/s]
 68%|██████▊   | 579/852 [00:07<00:03, 76.97it/s]
 69%|██████▉   | 587/852 [00:07<00:03, 76.55it/s]
 70%|██████▉   | 596/852 [00:07<00:03, 78.37it/s]
 71%|███████   | 604/852 [00:07<00:03, 78.38it/s]
 72%|███████▏  | 612/852 [00:07<00:03, 77.21it/s]
 73%|███████▎  | 620/852 [00:07<00:02, 77.48it/s]
 74%|███████▎  | 628/852 [00:08<00:02, 76.66it/s]
 75%|███████▍  | 636/852 [00:08<00:02, 76.28it/s]
 76%|███████▌  | 644/852 [00:08<00:02, 70.79it/s]
 77%|███████▋  | 652/852 [00:08<00:02, 71.23it/s]
 78%|███████▊  | 661/852 [00:08<00:02, 74.26it/s]
 79%|███████▊  | 669/852 [00:08<00:02, 74.79it/s]
 79%|███████▉  | 677/852 [00:08<00:02, 75.78it/s]
 81%|████████  | 686/852 [00:08<00:02, 77.22it/s]
 82%|████████▏ | 695/852 [00:08<00:01, 78.85it/s]
 83%|████████▎ | 704/852 [00:09<00:01, 80.60it/s]
 84%|████████▎ | 713/852 [00:09<00:01, 81.65it/s]
 85%|████████▍ | 722/852 [00:09<00:01, 80.17it/s]
 86%|████████▌ | 731/852 [00:09<00:01, 79.99it/s]
 87%|████████▋ | 740/852 [00:09<00:01, 80.97it/s]
 88%|████████▊ | 749/852 [00:09<00:01, 81.10it/s]
 89%|████████▉ | 758/852 [00:09<00:01, 81.97it/s]
 90%|█████████ | 767/852 [00:09<00:01, 80.55it/s]
 91%|█████████ | 776/852 [00:09<00:00, 80.68it/s]
 92%|█████████▏| 785/852 [00:10<00:00, 79.64it/s]
 93%|█████████▎| 793/852 [00:10<00:00, 79.71it/s]
 94%|█████████▍| 801/852 [00:10<00:00, 79.62it/s]
 95%|█████████▍| 809/852 [00:10<00:00, 77.20it/s]
 96%|█████████▌| 817/852 [00:10<00:00, 76.64it/s]
 97%|█████████▋| 826/852 [00:10<00:00, 78.45it/s]
 98%|█████████▊| 835/852 [00:10<00:00, 79.43it/s]
 99%|█████████▉| 843/852 [00:10<00:00, 78.34it/s]
+***** eval metrics *****
+  epoch                   =     9.9882
+  eval_accuracy           =     0.9768
+  eval_f1                 =     0.8015
+  eval_loss               =     0.1294
+  eval_precision          =     0.7939
+  eval_recall             =     0.8093
+  eval_runtime            = 0:00:14.48
+  eval_samples            =       6810
+  eval_samples_per_second =    470.164
+  eval_steps_per_second   =     58.822
+08/30/2024 20:46:16 - INFO - __main__ -   *** Predict ***
+[INFO|trainer.py:805] 2024-08-30 20:46:16,773 >> The following columns in the test set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: tokens, ner_tags, id. If tokens, ner_tags, id are not expected by `RobertaForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3788] 2024-08-30 20:46:16,775 >> 
+***** Running Prediction *****
+[INFO|trainer.py:3790] 2024-08-30 20:46:16,775 >>   Num examples = 14614
+[INFO|trainer.py:3793] 2024-08-30 20:46:16,775 >>   Batch size = 8
+
  0%|          | 0/1827 [00:00<?, ?it/s]
  1%|          | 10/1827 [00:00<00:20, 88.35it/s]
  1%|          | 19/1827 [00:00<00:23, 78.49it/s]
  1%|▏         | 27/1827 [00:00<00:23, 77.09it/s]
  2%|▏         | 35/1827 [00:00<00:23, 76.98it/s]
  2%|▏         | 44/1827 [00:00<00:22, 78.90it/s]
  3%|▎         | 52/1827 [00:00<00:22, 77.56it/s]
  3%|▎         | 60/1827 [00:00<00:23, 75.36it/s]
  4%|▎         | 68/1827 [00:00<00:23, 75.95it/s]
  4%|▍         | 76/1827 [00:00<00:22, 76.90it/s]
  5%|▍         | 84/1827 [00:01<00:22, 77.57it/s]
  5%|▌         | 92/1827 [00:01<00:22, 78.03it/s]
  5%|▌         | 100/1827 [00:01<00:22, 77.44it/s]
  6%|▌         | 109/1827 [00:01<00:21, 79.23it/s]
  6%|▋         | 117/1827 [00:01<00:21, 77.97it/s]
  7%|▋         | 125/1827 [00:01<00:21, 77.49it/s]
  7%|▋         | 134/1827 [00:01<00:21, 78.51it/s]
  8%|▊         | 143/1827 [00:01<00:21, 80.08it/s]
  8%|▊         | 152/1827 [00:01<00:20, 79.98it/s]
  9%|▉         | 160/1827 [00:02<00:21, 76.81it/s]
  9%|▉         | 169/1827 [00:02<00:21, 78.19it/s]
 10%|▉         | 178/1827 [00:02<00:20, 80.05it/s]
 10%|█         | 187/1827 [00:02<00:20, 81.05it/s]
 11%|█         | 196/1827 [00:02<00:20, 80.06it/s]
 11%|█         | 205/1827 [00:02<00:20, 79.21it/s]
 12%|█▏        | 214/1827 [00:02<00:20, 78.84it/s]
 12%|█▏        | 222/1827 [00:02<00:20, 78.64it/s]
 13%|█▎        | 230/1827 [00:02<00:20, 77.74it/s]
 13%|█▎        | 238/1827 [00:03<00:20, 76.28it/s]
 14%|█▎        | 247/1827 [00:03<00:20, 78.47it/s]
 14%|█▍        | 256/1827 [00:03<00:19, 79.66it/s]
 14%|█▍        | 264/1827 [00:03<00:19, 79.04it/s]
 15%|█▍        | 273/1827 [00:03<00:19, 80.83it/s]
 15%|█▌        | 282/1827 [00:03<00:18, 81.79it/s]
 16%|█▌        | 291/1827 [00:03<00:19, 80.01it/s]
 16%|█▋        | 300/1827 [00:03<00:18, 80.85it/s]
 17%|█▋        | 309/1827 [00:03<00:19, 79.16it/s]
 17%|█▋        | 317/1827 [00:04<00:19, 79.20it/s]
 18%|█▊        | 326/1827 [00:04<00:18, 80.90it/s]
 18%|█▊        | 335/1827 [00:04<00:18, 80.27it/s]
 19%|█▉        | 344/1827 [00:04<00:18, 81.58it/s]
 19%|█▉        | 353/1827 [00:04<00:19, 76.68it/s]
 20%|█▉        | 361/1827 [00:04<00:18, 77.41it/s]
 20%|██        | 370/1827 [00:04<00:18, 78.63it/s]
 21%|██        | 379/1827 [00:04<00:18, 79.94it/s]
 21%|██        | 388/1827 [00:04<00:18, 79.31it/s]
 22%|██▏       | 396/1827 [00:05<00:18, 78.99it/s]
 22%|██▏       | 405/1827 [00:05<00:17, 80.86it/s]
 23%|██▎       | 414/1827 [00:05<00:17, 79.00it/s]
 23%|██▎       | 423/1827 [00:05<00:17, 80.11it/s]
 24%|██▎       | 432/1827 [00:05<00:17, 79.18it/s]
 24%|██▍       | 440/1827 [00:05<00:17, 79.33it/s]
 25%|██▍       | 448/1827 [00:05<00:17, 79.41it/s]
 25%|██▌       | 457/1827 [00:05<00:17, 80.52it/s]
 26%|██▌       | 466/1827 [00:05<00:16, 80.24it/s]
 26%|██▌       | 475/1827 [00:06<00:17, 78.66it/s]
 26%|██▋       | 483/1827 [00:06<00:17, 78.64it/s]
 27%|██▋       | 491/1827 [00:06<00:17, 77.35it/s]
 27%|██▋       | 500/1827 [00:06<00:16, 78.45it/s]
 28%|██▊       | 508/1827 [00:06<00:16, 78.32it/s]
 28%|██▊       | 516/1827 [00:06<00:16, 78.75it/s]
 29%|██▊       | 525/1827 [00:06<00:16, 78.79it/s]
 29%|██▉       | 533/1827 [00:06<00:16, 78.81it/s]
 30%|██▉       | 541/1827 [00:06<00:16, 79.14it/s]
 30%|███       | 550/1827 [00:06<00:16, 79.35it/s]
 31%|███       | 559/1827 [00:07<00:16, 79.12it/s]
 31%|███       | 568/1827 [00:07<00:15, 81.14it/s]
 32%|███▏      | 577/1827 [00:07<00:15, 82.20it/s]
 32%|███▏      | 586/1827 [00:07<00:15, 81.32it/s]
 33%|███▎      | 595/1827 [00:07<00:15, 81.45it/s]
 33%|███▎      | 604/1827 [00:07<00:14, 81.95it/s]
 34%|███▎      | 613/1827 [00:07<00:15, 80.89it/s]
 34%|███▍      | 622/1827 [00:07<00:14, 80.81it/s]
 35%|███▍      | 631/1827 [00:07<00:14, 82.11it/s]
 35%|███▌      | 640/1827 [00:08<00:14, 82.74it/s]
 36%|███▌      | 649/1827 [00:08<00:14, 80.60it/s]
 36%|███▌      | 658/1827 [00:08<00:14, 78.56it/s]
 37%|███▋      | 667/1827 [00:08<00:14, 80.49it/s]
 37%|███▋      | 676/1827 [00:08<00:14, 80.58it/s]
 37%|███▋      | 685/1827 [00:08<00:13, 81.73it/s]
 38%|███▊      | 694/1827 [00:08<00:13, 82.81it/s]
 38%|███▊      | 703/1827 [00:08<00:13, 81.28it/s]
 39%|███▉      | 712/1827 [00:08<00:13, 82.12it/s]
 39%|███▉      | 721/1827 [00:09<00:13, 83.16it/s]
 40%|███▉      | 730/1827 [00:09<00:13, 83.47it/s]
 40%|████      | 739/1827 [00:09<00:12, 83.79it/s]
 41%|████      | 748/1827 [00:09<00:12, 84.99it/s]
 41%|████▏     | 757/1827 [00:09<00:12, 85.15it/s]
 42%|████▏     | 766/1827 [00:09<00:12, 85.06it/s]
 42%|████▏     | 775/1827 [00:09<00:12, 85.29it/s]
 43%|████▎     | 784/1827 [00:09<00:12, 81.63it/s]
 43%|████▎     | 793/1827 [00:09<00:12, 82.24it/s]
 44%|████▍     | 802/1827 [00:10<00:12, 81.57it/s]
 44%|████▍     | 811/1827 [00:10<00:12, 82.68it/s]
 45%|████▍     | 820/1827 [00:10<00:12, 82.99it/s]
 45%|████▌     | 829/1827 [00:10<00:11, 83.82it/s]
 46%|████▌     | 838/1827 [00:10<00:12, 82.25it/s]
 46%|████▋     | 847/1827 [00:10<00:11, 83.20it/s]
 47%|████▋     | 856/1827 [00:10<00:11, 84.21it/s]
 47%|████▋     | 865/1827 [00:10<00:11, 84.27it/s]
 48%|████▊     | 874/1827 [00:10<00:11, 80.96it/s]
 48%|████▊     | 883/1827 [00:11<00:11, 81.30it/s]
 49%|████▉     | 892/1827 [00:11<00:11, 80.78it/s]
 49%|████▉     | 901/1827 [00:11<00:11, 82.19it/s]
 50%|████▉     | 910/1827 [00:11<00:11, 79.40it/s]
 50%|█████     | 919/1827 [00:11<00:11, 79.95it/s]
 51%|█████     | 928/1827 [00:11<00:11, 80.89it/s]
 51%|█████▏    | 937/1827 [00:11<00:11, 79.25it/s]
 52%|█████▏    | 945/1827 [00:11<00:11, 79.00it/s]
 52%|█████▏    | 953/1827 [00:11<00:11, 77.83it/s]
 53%|█████▎    | 962/1827 [00:12<00:10, 78.70it/s]
 53%|█████▎    | 971/1827 [00:12<00:10, 80.17it/s]
 54%|█████▎    | 980/1827 [00:12<00:10, 79.85it/s]
 54%|█████▍    | 989/1827 [00:12<00:10, 81.39it/s]
 55%|█████▍    | 998/1827 [00:12<00:10, 82.20it/s]
 55%|█████▌    | 1007/1827 [00:12<00:10, 81.27it/s]
 56%|█████▌    | 1016/1827 [00:12<00:10, 80.65it/s]
 56%|█████▌    | 1025/1827 [00:12<00:09, 81.53it/s]
 57%|█████▋    | 1034/1827 [00:12<00:09, 79.74it/s]
 57%|█████▋    | 1043/1827 [00:13<00:09, 80.67it/s]
 58%|█████▊    | 1052/1827 [00:13<00:09, 81.68it/s]
 58%|█████▊    | 1061/1827 [00:13<00:09, 82.32it/s]
 59%|█████▊    | 1070/1827 [00:13<00:09, 83.48it/s]
 59%|█████▉    | 1079/1827 [00:13<00:08, 83.94it/s]
 60%|█████▉    | 1088/1827 [00:13<00:08, 83.86it/s]
 60%|██████    | 1097/1827 [00:13<00:08, 83.97it/s]
 61%|██████    | 1106/1827 [00:13<00:08, 83.02it/s]
 61%|██████    | 1115/1827 [00:13<00:08, 81.44it/s]
 62%|██████▏   | 1124/1827 [00:13<00:08, 81.63it/s]
 62%|██████▏   | 1133/1827 [00:14<00:08, 81.21it/s]
 63%|██████▎   | 1142/1827 [00:14<00:08, 82.28it/s]
 63%|██████▎   | 1151/1827 [00:14<00:08, 83.08it/s]
 63%|██████▎   | 1160/1827 [00:14<00:08, 81.13it/s]
 64%|██████▍   | 1169/1827 [00:14<00:08, 81.26it/s]
 64%|██████▍   | 1178/1827 [00:14<00:07, 82.06it/s]
 65%|██████▍   | 1187/1827 [00:14<00:08, 77.70it/s]
 65%|██████▌   | 1196/1827 [00:14<00:08, 78.65it/s]
 66%|██████▌   | 1205/1827 [00:14<00:07, 80.19it/s]
 66%|██████▋   | 1214/1827 [00:15<00:07, 78.78it/s]
 67%|██████▋   | 1223/1827 [00:15<00:07, 79.88it/s]
 67%|██████▋   | 1232/1827 [00:15<00:07, 81.80it/s]
 68%|██████▊   | 1241/1827 [00:15<00:07, 82.09it/s]
 68%|██████▊   | 1250/1827 [00:15<00:06, 82.44it/s]
 69%|██████▉   | 1259/1827 [00:15<00:07, 80.44it/s]
 69%|██████▉   | 1268/1827 [00:15<00:06, 80.68it/s]
 70%|██████▉   | 1277/1827 [00:15<00:06, 80.44it/s]
 70%|███████   | 1286/1827 [00:15<00:06, 81.54it/s]
 71%|███████   | 1295/1827 [00:16<00:06, 82.61it/s]
 71%|███████▏  | 1304/1827 [00:16<00:06, 83.55it/s]
 72%|███████▏  | 1313/1827 [00:16<00:06, 83.36it/s]
 72%|███████▏  | 1322/1827 [00:16<00:06, 83.66it/s]
 73%|███████▎  | 1331/1827 [00:16<00:05, 84.22it/s]
 73%|███████▎  | 1340/1827 [00:16<00:05, 84.53it/s]
 74%|███████▍  | 1349/1827 [00:16<00:05, 81.75it/s]
 74%|███████▍  | 1358/1827 [00:16<00:05, 80.15it/s]
 75%|███████▍  | 1367/1827 [00:16<00:05, 80.04it/s]
 75%|███████▌  | 1376/1827 [00:17<00:05, 80.07it/s]
 76%|███████▌  | 1385/1827 [00:17<00:05, 81.10it/s]
 76%|███████▋  | 1394/1827 [00:17<00:05, 80.87it/s]
 77%|███████▋  | 1403/1827 [00:17<00:05, 82.04it/s]
 77%|███████▋  | 1412/1827 [00:17<00:05, 82.83it/s]
 78%|███████▊  | 1421/1827 [00:17<00:04, 82.88it/s]
 78%|███████▊  | 1430/1827 [00:17<00:04, 81.23it/s]
 79%|███████▉  | 1439/1827 [00:17<00:04, 79.21it/s]
 79%|███████▉  | 1448/1827 [00:17<00:04, 80.64it/s]
 80%|███████▉  | 1457/1827 [00:18<00:04, 82.29it/s]
 80%|████████  | 1466/1827 [00:18<00:04, 83.72it/s]
 81%|████████  | 1475/1827 [00:18<00:04, 80.74it/s]
 81%|████████  | 1484/1827 [00:18<00:04, 81.51it/s]
 82%|████████▏ | 1493/1827 [00:18<00:04, 79.63it/s]
 82%|████████▏ | 1501/1827 [00:18<00:04, 79.30it/s]
 83%|████████▎ | 1510/1827 [00:18<00:03, 79.91it/s]
 83%|████████▎ | 1519/1827 [00:18<00:03, 80.55it/s]
 84%|████████▎ | 1528/1827 [00:18<00:03, 79.14it/s]
 84%|████████▍ | 1537/1827 [00:19<00:03, 79.70it/s]
 85%|████████▍ | 1546/1827 [00:19<00:03, 80.07it/s]
 85%|████████▌ | 1555/1827 [00:19<00:03, 80.71it/s]
 86%|████████▌ | 1564/1827 [00:19<00:03, 80.13it/s]
 86%|████████▌ | 1573/1827 [00:19<00:03, 80.19it/s]
 87%|████████▋ | 1582/1827 [00:19<00:03, 81.00it/s]
 87%|████████▋ | 1591/1827 [00:19<00:02, 80.84it/s]
 88%|████████▊ | 1600/1827 [00:19<00:02, 80.45it/s]
 88%|████████▊ | 1609/1827 [00:19<00:02, 80.95it/s]
 89%|████████▊ | 1618/1827 [00:20<00:02, 77.39it/s]
 89%|████████▉ | 1626/1827 [00:20<00:02, 75.05it/s]
 89%|████████▉ | 1634/1827 [00:20<00:02, 76.14it/s]
 90%|████████▉ | 1642/1827 [00:20<00:02, 77.16it/s]
 90%|█████████ | 1651/1827 [00:20<00:02, 79.24it/s]
 91%|█████████ | 1660/1827 [00:20<00:02, 79.60it/s]
 91%|��████████▏| 1668/1827 [00:20<00:02, 78.92it/s]
 92%|█████████▏| 1676/1827 [00:20<00:01, 79.10it/s]
 92%|█████████▏| 1684/1827 [00:20<00:01, 78.48it/s]
 93%|█████████▎| 1692/1827 [00:21<00:01, 74.95it/s]
 93%|█████████▎| 1701/1827 [00:21<00:01, 76.99it/s]
 94%|█████████▎| 1710/1827 [00:21<00:01, 78.97it/s]
 94%|█████████▍| 1719/1827 [00:21<00:01, 77.08it/s]
 95%|█████████▍| 1727/1827 [00:21<00:01, 76.46it/s]
 95%|█████████▌| 1736/1827 [00:21<00:01, 77.82it/s]
 96%|█████████▌| 1745/1827 [00:21<00:01, 79.51it/s]
 96%|█████████▌| 1754/1827 [00:21<00:00, 80.76it/s]
 96%|█████████▋| 1763/1827 [00:21<00:00, 80.67it/s]
 97%|█████████▋| 1772/1827 [00:22<00:00, 81.04it/s]
 97%|█████████▋| 1781/1827 [00:22<00:00, 82.22it/s]
 98%|█████████▊| 1790/1827 [00:22<00:00, 82.53it/s]
 98%|█████████▊| 1799/1827 [00:22<00:00, 78.84it/s]
 99%|█████████▉| 1808/1827 [00:22<00:00, 80.29it/s]
 99%|█████████▉| 1817/1827 [00:22<00:00, 81.17it/s]
+[INFO|trainer.py:3478] 2024-08-30 20:46:46,759 >> Saving model checkpoint to /content/dissertation/scripts/ner/output
+[INFO|configuration_utils.py:472] 2024-08-30 20:46:46,762 >> Configuration saved in /content/dissertation/scripts/ner/output/config.json
+[INFO|modeling_utils.py:2690] 2024-08-30 20:46:48,100 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
+[INFO|tokenization_utils_base.py:2574] 2024-08-30 20:46:48,101 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2583] 2024-08-30 20:46:48,101 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+***** predict metrics *****
+  predict_accuracy           =     0.9766
+  predict_f1                 =     0.8012
+  predict_loss               =     0.1233
+  predict_precision          =     0.7872
+  predict_recall             =     0.8157
+  predict_runtime            = 0:00:29.36
+  predict_samples_per_second =    497.683
+  predict_steps_per_second   =     62.219
+

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 9.988249118683902,
-    "total_flos": 1.2649810588547778e+16,
-    "train_loss": 0.10639642311544979,
-    "train_runtime": 1208.2019,
+    "total_flos": 1.2649124434987926e+16,
+    "train_loss": 0.02590363489880281,
+    "train_runtime": 1203.9865,
     "train_samples": 27229,
-    "train_samples_per_second": 225.368,
-    "train_steps_per_second": 3.518
+    "train_samples_per_second": 226.157,
+    "train_steps_per_second": 3.53
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.17333222008850296,
-  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-2127",
+  "best_metric": 0.8015293708724366,
+  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-3404",
   "epoch": 9.988249118683902,
   "eval_steps": 500,
   "global_step": 4250,
@@ -10,188 +10,188 @@
   "log_history": [
     {
       "epoch": 0.9988249118683902,
-      "eval_accuracy": 0.7849561289082346,
-      "eval_f1": 0.1613083366573594,
-      "eval_loss": 0.6610585451126099,
-      "eval_precision": 0.08832008386476806,
-      "eval_recall": 0.9292279411764706,
-      "eval_runtime": 14.7703,
-      "eval_samples_per_second": 461.06,
-      "eval_steps_per_second": 57.683,
+      "eval_accuracy": 0.9733340656624604,
+      "eval_f1": 0.7536426810132258,
+      "eval_loss": 0.07383445650339127,
+      "eval_precision": 0.7233218588640276,
+      "eval_recall": 0.786616752456715,
+      "eval_runtime": 14.6851,
+      "eval_samples_per_second": 463.736,
+      "eval_steps_per_second": 58.018,
       "step": 425
     },
     {
       "epoch": 1.1750881316098707,
-      "grad_norm": 1.4406206607818604,
+      "grad_norm": 1.1428656578063965,
       "learning_rate": 4.411764705882353e-05,
-      "loss": 0.3349,
+      "loss": 0.0996,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7551182940392986,
-      "eval_f1": 0.14506880733944952,
-      "eval_loss": 0.9204075932502747,
-      "eval_precision": 0.07866915422885572,
-      "eval_recall": 0.9301470588235294,
-      "eval_runtime": 14.7453,
-      "eval_samples_per_second": 461.844,
-      "eval_steps_per_second": 57.781,
+      "eval_accuracy": 0.9743364411550661,
+      "eval_f1": 0.7698492462311558,
+      "eval_loss": 0.07865303754806519,
+      "eval_precision": 0.7363811151463363,
+      "eval_recall": 0.8065044454843239,
+      "eval_runtime": 14.4143,
+      "eval_samples_per_second": 472.448,
+      "eval_steps_per_second": 59.108,
       "step": 851
     },
     {
       "epoch": 2.3501762632197414,
-      "grad_norm": 2.988006591796875,
+      "grad_norm": 1.476723074913025,
       "learning_rate": 3.8235294117647055e-05,
-      "loss": 0.1788,
+      "loss": 0.0458,
       "step": 1000
     },
     {
       "epoch": 2.99882491186839,
-      "eval_accuracy": 0.7645035495077375,
-      "eval_f1": 0.15487907225146869,
-      "eval_loss": 0.9544711709022522,
-      "eval_precision": 0.0844496214327315,
-      "eval_recall": 0.9329044117647058,
-      "eval_runtime": 14.8107,
-      "eval_samples_per_second": 459.803,
-      "eval_steps_per_second": 57.526,
+      "eval_accuracy": 0.9759155258351985,
+      "eval_f1": 0.7928563303378454,
+      "eval_loss": 0.07876282930374146,
+      "eval_precision": 0.7715297764002657,
+      "eval_recall": 0.8153954141319607,
+      "eval_runtime": 14.6794,
+      "eval_samples_per_second": 463.917,
+      "eval_steps_per_second": 58.041,
       "step": 1276
     },
     {
       "epoch": 3.525264394829612,
-      "grad_norm": 1.1645787954330444,
+      "grad_norm": 0.4094911515712738,
       "learning_rate": 3.235294117647059e-05,
-      "loss": 0.1227,
+      "loss": 0.0279,
       "step": 1500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7692064756203056,
-      "eval_f1": 0.1618332674832082,
-      "eval_loss": 1.0923608541488647,
-      "eval_precision": 0.08852770813521224,
-      "eval_recall": 0.9411764705882353,
-      "eval_runtime": 14.8211,
-      "eval_samples_per_second": 459.48,
-      "eval_steps_per_second": 57.486,
+      "eval_accuracy": 0.9757026926826589,
+      "eval_f1": 0.7929102344196683,
+      "eval_loss": 0.09221930056810379,
+      "eval_precision": 0.775441735629613,
+      "eval_recall": 0.8111839026672906,
+      "eval_runtime": 14.825,
+      "eval_samples_per_second": 459.361,
+      "eval_steps_per_second": 57.471,
       "step": 1702
     },
     {
       "epoch": 4.700352526439483,
-      "grad_norm": 1.1214195489883423,
+      "grad_norm": 0.49960169196128845,
       "learning_rate": 2.647058823529412e-05,
-      "loss": 0.0856,
+      "loss": 0.0169,
       "step": 2000
     },
     {
       "epoch": 4.9988249118683905,
-      "eval_accuracy": 0.7932840841995413,
-      "eval_f1": 0.17333222008850296,
-      "eval_loss": 1.0502684116363525,
-      "eval_precision": 0.09532555790247038,
-      "eval_recall": 0.9540441176470589,
-      "eval_runtime": 14.797,
-      "eval_samples_per_second": 460.229,
-      "eval_steps_per_second": 57.579,
+      "eval_accuracy": 0.974439424938553,
+      "eval_f1": 0.7863421230561191,
+      "eval_loss": 0.09940312057733536,
+      "eval_precision": 0.7584782608695653,
+      "eval_recall": 0.816331305568554,
+      "eval_runtime": 14.3671,
+      "eval_samples_per_second": 473.999,
+      "eval_steps_per_second": 59.302,
       "step": 2127
     },
     {
       "epoch": 5.875440658049354,
-      "grad_norm": 1.1390776634216309,
+      "grad_norm": 0.6507154703140259,
       "learning_rate": 2.058823529411765e-05,
-      "loss": 0.0597,
+      "loss": 0.0114,
       "step": 2500
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.7787771018990209,
-      "eval_f1": 0.16632958498503356,
-      "eval_loss": 1.2641881704330444,
-      "eval_precision": 0.0911913421449481,
-      "eval_recall": 0.9448529411764706,
-      "eval_runtime": 14.5753,
-      "eval_samples_per_second": 467.228,
-      "eval_steps_per_second": 58.455,
+      "eval_accuracy": 0.9764579070948961,
+      "eval_f1": 0.7909059593523942,
+      "eval_loss": 0.10803968459367752,
+      "eval_precision": 0.7765501691093574,
+      "eval_recall": 0.8058025269068788,
+      "eval_runtime": 14.5335,
+      "eval_samples_per_second": 468.572,
+      "eval_steps_per_second": 58.623,
       "step": 2553
     },
     {
       "epoch": 6.9988249118683905,
-      "eval_accuracy": 0.7828758564817994,
-      "eval_f1": 0.16898640903880793,
-      "eval_loss": 1.3261910676956177,
-      "eval_precision": 0.09275570735214812,
-      "eval_recall": 0.9485294117647058,
-      "eval_runtime": 14.5617,
-      "eval_samples_per_second": 467.665,
-      "eval_steps_per_second": 58.51,
+      "eval_accuracy": 0.975997912861988,
+      "eval_f1": 0.7943099690260411,
+      "eval_loss": 0.11656877398490906,
+      "eval_precision": 0.7792032410533424,
+      "eval_recall": 0.810014038371549,
+      "eval_runtime": 14.6842,
+      "eval_samples_per_second": 463.764,
+      "eval_steps_per_second": 58.022,
       "step": 2978
     },
     {
       "epoch": 7.050528789659224,
-      "grad_norm": 0.5195357799530029,
+      "grad_norm": 0.15811629593372345,
       "learning_rate": 1.4705882352941177e-05,
-      "loss": 0.0458,
+      "loss": 0.0079,
       "step": 3000
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.7848943386381424,
-      "eval_f1": 0.1687846203975236,
-      "eval_loss": 1.3697636127471924,
-      "eval_precision": 0.09259921344297461,
-      "eval_recall": 0.9522058823529411,
-      "eval_runtime": 14.6608,
-      "eval_samples_per_second": 464.504,
-      "eval_steps_per_second": 58.114,
+      "eval_accuracy": 0.9767668584453568,
+      "eval_f1": 0.8015293708724366,
+      "eval_loss": 0.1294233798980713,
+      "eval_precision": 0.7938948817994033,
+      "eval_recall": 0.8093121197941039,
+      "eval_runtime": 14.774,
+      "eval_samples_per_second": 460.946,
+      "eval_steps_per_second": 57.669,
       "step": 3404
     },
     {
       "epoch": 8.225616921269095,
-      "grad_norm": 0.8723571300506592,
+      "grad_norm": 0.11404519528150558,
       "learning_rate": 8.823529411764707e-06,
-      "loss": 0.0343,
+      "loss": 0.0053,
       "step": 3500
     },
     {
       "epoch": 8.99882491186839,
-      "eval_accuracy": 0.782223625853049,
-      "eval_f1": 0.165499476776946,
-      "eval_loss": 1.4433350563049316,
-      "eval_precision": 0.09069254521393913,
-      "eval_recall": 0.9448529411764706,
-      "eval_runtime": 15.0504,
-      "eval_samples_per_second": 452.478,
-      "eval_steps_per_second": 56.61,
+      "eval_accuracy": 0.9766158155629093,
+      "eval_f1": 0.7988929889298894,
+      "eval_loss": 0.13398884236812592,
+      "eval_precision": 0.787630741246021,
+      "eval_recall": 0.8104819840898456,
+      "eval_runtime": 14.4437,
+      "eval_samples_per_second": 471.487,
+      "eval_steps_per_second": 58.988,
       "step": 3829
     },
     {
       "epoch": 9.400705052878966,
-      "grad_norm": 0.6075822710990906,
+      "grad_norm": 0.5636719465255737,
       "learning_rate": 2.9411764705882355e-06,
-      "loss": 0.0292,
+      "loss": 0.0038,
       "step": 4000
     },
     {
       "epoch": 9.988249118683902,
-      "eval_accuracy": 0.7820794485561674,
-      "eval_f1": 0.16674769081186194,
-      "eval_loss": 1.4861844778060913,
-      "eval_precision": 0.0914341567442687,
-      "eval_recall": 0.9457720588235294,
-      "eval_runtime": 15.3988,
-      "eval_samples_per_second": 442.243,
-      "eval_steps_per_second": 55.329,
+      "eval_accuracy": 0.9766776058330014,
+      "eval_f1": 0.7988459319099828,
+      "eval_loss": 0.13674204051494598,
+      "eval_precision": 0.7882031427920747,
+      "eval_recall": 0.8097800655124006,
+      "eval_runtime": 14.8828,
+      "eval_samples_per_second": 457.575,
+      "eval_steps_per_second": 57.247,
       "step": 4250
     },
     {
       "epoch": 9.988249118683902,
       "step": 4250,
-      "total_flos": 1.2649810588547778e+16,
-      "train_loss": 0.10639642311544979,
-      "train_runtime": 1208.2019,
-      "train_samples_per_second": 225.368,
-      "train_steps_per_second": 3.518
+      "total_flos": 1.2649124434987926e+16,
+      "train_loss": 0.02590363489880281,
+      "train_runtime": 1203.9865,
+      "train_samples_per_second": 226.157,
+      "train_steps_per_second": 3.53
     }
   ],
   "logging_steps": 500,
@@ -211,7 +211,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.2649810588547778e+16,
+  "total_flos": 1.2649124434987926e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null