End of training

README.md

@@ -3,9 +3,10 @@ library_name: transformers
 license: apache-2.0
 base_model: PlanTL-GOB-ES/bsc-bio-ehr-es
 tags:
+- token-classification
 - generated_from_trainer
 datasets:
-- cantemist-85-ner
+- Rodrigo1771/cantemist-85-ner
 metrics:
 - precision
 - recall
@@ -18,8 +19,8 @@ model-index:
       name: Token Classification
       type: token-classification
     dataset:
-      name: cantemist-85-ner
-      type: cantemist-85-ner
+      name: Rodrigo1771/cantemist-85-ner
+      type: Rodrigo1771/cantemist-85-ner
       config: CantemistNer
       split: validation
       args: CantemistNer
@@ -35,7 +36,7 @@ model-index:
       value: 0.8506998444790046
     - name: Accuracy
       type: accuracy
-      value: 0.9916486929514279
+      value: 0.9916544445403043
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -43,13 +44,13 @@ 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 cantemist-85-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/cantemist-85-ner dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0503
+- Loss: 0.0496
 - Precision: 0.8399
 - Recall: 0.8618
 - F1: 0.8507
-- Accuracy: 0.9916
+- Accuracy: 0.9917
 
 ## Model description
 

all_results.json

@@ -0,0 +1,26 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.9916544445403043,
+    "eval_f1": 0.8506998444790046,
+    "eval_loss": 0.04956069961190224,
+    "eval_precision": 0.8399232245681382,
+    "eval_recall": 0.8617565970854667,
+    "eval_runtime": 15.5432,
+    "eval_samples": 7354,
+    "eval_samples_per_second": 473.133,
+    "eval_steps_per_second": 59.19,
+    "predict_accuracy": 0.9919921133225054,
+    "predict_f1": 0.8561699809316262,
+    "predict_loss": 0.04697508364915848,
+    "predict_precision": 0.8359042553191489,
+    "predict_recall": 0.8774427694025684,
+    "predict_runtime": 23.0034,
+    "predict_samples_per_second": 471.149,
+    "predict_steps_per_second": 58.904,
+    "total_flos": 1.5369860684670966e+16,
+    "train_loss": 0.009731636565258824,
+    "train_runtime": 1413.3317,
+    "train_samples": 32675,
+    "train_samples_per_second": 231.191,
+    "train_steps_per_second": 3.616
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.9916544445403043,
+    "eval_f1": 0.8506998444790046,
+    "eval_loss": 0.04956069961190224,
+    "eval_precision": 0.8399232245681382,
+    "eval_recall": 0.8617565970854667,
+    "eval_runtime": 15.5432,
+    "eval_samples": 7354,
+    "eval_samples_per_second": 473.133,
+    "eval_steps_per_second": 59.19
+}

predict_results.json

@@ -0,0 +1,10 @@
+{
+    "predict_accuracy": 0.9919921133225054,
+    "predict_f1": 0.8561699809316262,
+    "predict_loss": 0.04697508364915848,
+    "predict_precision": 0.8359042553191489,
+    "predict_recall": 0.8774427694025684,
+    "predict_runtime": 23.0034,
+    "predict_samples_per_second": 471.149,
+    "predict_steps_per_second": 58.904
+}

tb/events.out.tfevents.1725573899.df0b2d2cc7fe.2457.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6423da8ec0617924aeaa8bb9f423f80167211d3925beb4a31c9c0f6412f1122c
+size 560

train.log

@@ -1562,3 +1562,51 @@ Training completed. Do not forget to share your model on huggingface.co/models =
 {'eval_loss': 0.05025585740804672, 'eval_precision': 0.8399232245681382, 'eval_recall': 0.8617565970854667, 'eval_f1': 0.8506998444790046, 'eval_accuracy': 0.9916486929514279, 'eval_runtime': 16.4037, 'eval_samples_per_second': 448.314, 'eval_steps_per_second': 56.085, 'epoch': 10.0}
 {'train_runtime': 1413.3317, 'train_samples_per_second': 231.191, 'train_steps_per_second': 3.616, 'train_loss': 0.009731636565258824, 'epoch': 10.0}
 
+***** train metrics *****
+  epoch                    =       10.0
+  total_flos               = 14314298GF
+  train_loss               =     0.0097
+  train_runtime            = 0:23:33.33
+  train_samples            =      32675
+  train_samples_per_second =    231.191
+  train_steps_per_second   =      3.616
+09/05/2024 22:04:44 - INFO - __main__ -   *** Evaluate ***
+[INFO|trainer.py:811] 2024-09-05 22:04:44,308 >> The following columns in the evaluation set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: tokens, id, ner_tags. If tokens, id, ner_tags are not expected by `RobertaForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3819] 2024-09-05 22:04:44,311 >> 
+***** Running Evaluation *****
+[INFO|trainer.py:3821] 2024-09-05 22:04:44,311 >>   Num examples = 7354
+[INFO|trainer.py:3824] 2024-09-05 22:04:44,311 >>   Batch size = 8
+
  0%|          | 0/920 [00:00<?, ?it/s]
  1%|          | 9/920 [00:00<00:10, 84.54it/s]
  2%|▏         | 18/920 [00:00<00:12, 73.05it/s]
  3%|▎         | 26/920 [00:00<00:11, 74.96it/s]
  4%|▍         | 35/920 [00:00<00:11, 77.46it/s]
  5%|▍         | 43/920 [00:00<00:11, 77.65it/s]
  6%|▌         | 52/920 [00:00<00:10, 79.04it/s]
  7%|▋         | 60/920 [00:00<00:11, 75.68it/s]
  7%|▋         | 68/920 [00:00<00:11, 76.64it/s]
  8%|▊         | 77/920 [00:00<00:10, 78.76it/s]
  9%|▉         | 85/920 [00:01<00:10, 79.00it/s]
 10%|█         | 94/920 [00:01<00:10, 79.41it/s]
 11%|█         | 103/920 [00:01<00:10, 79.93it/s]
 12%|█▏        | 111/920 [00:01<00:10, 79.52it/s]
 13%|█▎        | 120/920 [00:01<00:09, 80.07it/s]
 14%|█▍        | 129/920 [00:01<00:09, 81.24it/s]
 15%|█▌        | 138/920 [00:01<00:09, 81.96it/s]
 16%|█▌        | 147/920 [00:01<00:09, 82.29it/s]
 17%|█▋        | 156/920 [00:01<00:09, 80.75it/s]
 18%|█▊        | 165/920 [00:02<00:09, 81.61it/s]
 19%|█▉        | 174/920 [00:02<00:09, 82.72it/s]
 20%|█▉        | 183/920 [00:02<00:08, 82.84it/s]
 21%|██        | 192/920 [00:02<00:08, 81.41it/s]
 22%|██▏       | 201/920 [00:02<00:08, 80.35it/s]
 23%|██▎       | 210/920 [00:02<00:09, 77.43it/s]
 24%|██▍       | 219/920 [00:02<00:08, 79.21it/s]
 25%|██▍       | 228/920 [00:02<00:08, 80.10it/s]
 26%|██▌       | 237/920 [00:02<00:08, 81.55it/s]
 27%|██▋       | 246/920 [00:03<00:08, 82.11it/s]
 28%|██▊       | 255/920 [00:03<00:08, 81.17it/s]
 29%|██▊       | 264/920 [00:03<00:08, 78.82it/s]
 30%|██▉       | 272/920 [00:03<00:08, 78.57it/s]
 30%|███       | 280/920 [00:03<00:08, 78.09it/s]
 31%|███▏      | 288/920 [00:03<00:08, 78.11it/s]
 32%|███▏      | 297/920 [00:03<00:07, 78.87it/s]
 33%|███▎      | 306/920 [00:03<00:07, 79.62it/s]
 34%|███▍      | 315/920 [00:03<00:07, 80.07it/s]
 35%|███▌      | 324/920 [00:04<00:07, 79.49it/s]
 36%|███▌      | 332/920 [00:04<00:07, 78.63it/s]
 37%|███▋      | 340/920 [00:04<00:07, 77.32it/s]
 38%|███▊      | 349/920 [00:04<00:07, 78.42it/s]
 39%|███▉      | 357/920 [00:04<00:07, 78.09it/s]
 40%|███▉      | 365/920 [00:04<00:07, 78.40it/s]
 41%|████      | 374/920 [00:04<00:06, 79.90it/s]
 42%|████▏     | 383/920 [00:04<00:06, 81.28it/s]
 43%|████▎     | 392/920 [00:04<00:06, 78.97it/s]
 43%|████▎     | 400/920 [00:05<00:06, 77.62it/s]
 44%|████▍     | 408/920 [00:05<00:06, 76.16it/s]
 45%|████▌     | 417/920 [00:05<00:06, 77.84it/s]
 46%|████▋     | 426/920 [00:05<00:06, 78.34it/s]
 47%|████▋     | 435/920 [00:05<00:06, 79.22it/s]
 48%|████▊     | 444/920 [00:05<00:05, 80.13it/s]
 49%|████▉     | 453/920 [00:05<00:05, 81.19it/s]
 50%|█████     | 462/920 [00:05<00:05, 81.32it/s]
 51%|█████     | 471/920 [00:05<00:05, 80.10it/s]
 52%|█████▏    | 480/920 [00:06<00:05, 81.21it/s]
 53%|█████▎    | 489/920 [00:06<00:05, 81.06it/s]
 54%|█████▍    | 498/920 [00:06<00:05, 81.04it/s]
 55%|█████▌    | 507/920 [00:06<00:05, 81.72it/s]
 56%|█████▌    | 516/920 [00:06<00:04, 81.77it/s]
 57%|█████▋    | 525/920 [00:06<00:04, 82.08it/s]
 58%|█████▊    | 534/920 [00:06<00:04, 81.28it/s]
 59%|█████▉    | 543/920 [00:06<00:04, 81.25it/s]
 60%|██████    | 552/920 [00:06<00:04, 78.20it/s]
 61%|██████    | 560/920 [00:07<00:04, 75.62it/s]
 62%|██████▏   | 568/920 [00:07<00:04, 76.50it/s]
 63%|██████▎   | 577/920 [00:07<00:04, 78.56it/s]
 64%|██████▎   | 585/920 [00:07<00:04, 77.87it/s]
 65%|██████▍   | 594/920 [00:07<00:04, 79.73it/s]
 66%|██████▌   | 603/920 [00:07<00:03, 81.43it/s]
 67%|██████▋   | 612/920 [00:07<00:03, 81.46it/s]
 68%|██████▊   | 621/920 [00:07<00:03, 80.50it/s]
 68%|██████▊   | 630/920 [00:07<00:03, 79.92it/s]
 69%|██████▉   | 639/920 [00:08<00:03, 80.53it/s]
 70%|███████   | 648/920 [00:08<00:03, 81.62it/s]
 71%|███████▏  | 657/920 [00:08<00:03, 81.38it/s]
 72%|███████▏  | 666/920 [00:08<00:03, 81.64it/s]
 73%|███████▎  | 675/920 [00:08<00:02, 82.37it/s]
 74%|███████▍  | 684/920 [00:08<00:02, 83.02it/s]
 75%|███████▌  | 693/920 [00:08<00:02, 81.40it/s]
 76%|███████▋  | 702/920 [00:08<00:02, 81.01it/s]
 77%|███████▋  | 711/920 [00:08<00:02, 80.04it/s]
 78%|███████▊  | 720/920 [00:09<00:02, 81.29it/s]
 79%|███████▉  | 729/920 [00:09<00:02, 82.40it/s]
 80%|████████  | 738/920 [00:09<00:02, 81.32it/s]
 81%|████████  | 747/920 [00:09<00:02, 78.84it/s]
 82%|████████▏ | 756/920 [00:09<00:02, 79.45it/s]
 83%|████████▎ | 765/920 [00:09<00:01, 80.42it/s]
 84%|████████▍ | 774/920 [00:09<00:01, 81.22it/s]
 85%|████████▌ | 783/920 [00:09<00:01, 80.34it/s]
 86%|████████▌ | 792/920 [00:09<00:01, 79.82it/s]
 87%|████████▋ | 800/920 [00:10<00:01, 79.49it/s]
 88%|████████▊ | 808/920 [00:10<00:01, 78.30it/s]
 89%|████████▉ | 817/920 [00:10<00:01, 79.52it/s]
 90%|████████▉ | 826/920 [00:10<00:01, 80.37it/s]
 91%|█████████ | 835/920 [00:10<00:01, 81.87it/s]
 92%|█████████▏| 844/920 [00:10<00:00, 80.79it/s]
 93%|█████████▎| 853/920 [00:10<00:00, 76.52it/s]
 94%|█████████▎| 861/920 [00:10<00:00, 75.56it/s]
 95%|█████████▍| 870/920 [00:10<00:00, 77.00it/s]
 96%|█████████▌| 879/920 [00:11<00:00, 78.67it/s]
 97%|█████████▋| 888/920 [00:11<00:00, 79.25it/s]
 97%|█████████▋| 896/920 [00:11<00:00, 76.28it/s]
 98%|█████████▊| 905/920 [00:11<00:00, 77.41it/s]
 99%|█████████▉| 914/920 [00:11<00:00, 79.26it/s]
+***** eval metrics *****
+  epoch                   =       10.0
+  eval_accuracy           =     0.9917
+  eval_f1                 =     0.8507
+  eval_loss               =     0.0496
+  eval_precision          =     0.8399
+  eval_recall             =     0.8618
+  eval_runtime            = 0:00:15.54
+  eval_samples            =       7354
+  eval_samples_per_second =    473.133
+  eval_steps_per_second   =      59.19
+09/05/2024 22:04:59 - INFO - __main__ -   *** Predict ***
+[INFO|trainer.py:811] 2024-09-05 22:04:59,857 >> The following columns in the test set don't have a corresponding argument in `RobertaForTokenClassification.forward` and have been ignored: tokens, id, ner_tags. If tokens, id, ner_tags are not expected by `RobertaForTokenClassification.forward`,  you can safely ignore this message.
+[INFO|trainer.py:3819] 2024-09-05 22:04:59,859 >> 
+***** Running Prediction *****
+[INFO|trainer.py:3821] 2024-09-05 22:04:59,859 >>   Num examples = 10838
+[INFO|trainer.py:3824] 2024-09-05 22:04:59,859 >>   Batch size = 8
+
  0%|          | 0/1355 [00:00<?, ?it/s]
  1%|          | 9/1355 [00:00<00:15, 89.65it/s]
  1%|▏         | 18/1355 [00:00<00:16, 78.98it/s]
  2%|▏         | 27/1355 [00:00<00:16, 80.91it/s]
  3%|▎         | 36/1355 [00:00<00:16, 80.26it/s]
  3%|▎         | 45/1355 [00:00<00:16, 79.74it/s]
  4%|▍         | 53/1355 [00:00<00:17, 76.43it/s]
  5%|▍         | 61/1355 [00:00<00:16, 76.79it/s]
  5%|▌         | 69/1355 [00:00<00:16, 77.27it/s]
  6%|▌         | 78/1355 [00:00<00:16, 78.60it/s]
  6%|▋         | 87/1355 [00:01<00:15, 79.45it/s]
  7%|▋         | 95/1355 [00:01<00:15, 79.17it/s]
  8%|▊         | 103/1355 [00:01<00:15, 79.17it/s]
  8%|▊         | 111/1355 [00:01<00:16, 77.62it/s]
  9%|▉         | 119/1355 [00:01<00:16, 76.95it/s]
  9%|▉         | 128/1355 [00:01<00:15, 77.99it/s]
 10%|█         | 137/1355 [00:01<00:15, 78.88it/s]
 11%|█         | 145/1355 [00:01<00:15, 76.54it/s]
 11%|█▏        | 153/1355 [00:01<00:16, 74.41it/s]
 12%|█▏        | 161/1355 [00:02<00:17, 70.18it/s]
 12%|█▏        | 169/1355 [00:02<00:16, 71.76it/s]
 13%|█▎        | 178/1355 [00:02<00:15, 74.18it/s]
 14%|█▍        | 187/1355 [00:02<00:15, 76.84it/s]
 14%|█▍        | 196/1355 [00:02<00:14, 78.67it/s]
 15%|█▌        | 205/1355 [00:02<00:14, 79.95it/s]
 16%|█▌        | 214/1355 [00:02<00:14, 81.03it/s]
 16%|█▋        | 223/1355 [00:02<00:14, 79.30it/s]
 17%|█▋        | 232/1355 [00:02<00:14, 79.76it/s]
 18%|█▊        | 240/1355 [00:03<00:14, 78.82it/s]
 18%|█▊        | 249/1355 [00:03<00:13, 79.55it/s]
 19%|█▉        | 258/1355 [00:03<00:13, 80.77it/s]
 20%|█▉        | 267/1355 [00:03<00:13, 82.26it/s]
 20%|██        | 276/1355 [00:03<00:12, 83.16it/s]
 21%|██        | 285/1355 [00:03<00:13, 81.80it/s]
 22%|██▏       | 294/1355 [00:03<00:13, 81.16it/s]
 22%|██▏       | 303/1355 [00:03<00:12, 82.02it/s]
 23%|██▎       | 312/1355 [00:03<00:12, 83.08it/s]
 24%|██▎       | 321/1355 [00:04<00:12, 83.67it/s]
 24%|██▍       | 330/1355 [00:04<00:12, 84.72it/s]
 25%|██▌       | 339/1355 [00:04<00:11, 85.11it/s]
 26%|██▌       | 348/1355 [00:04<00:11, 85.47it/s]
 26%|██▋       | 357/1355 [00:04<00:11, 85.91it/s]
 27%|██▋       | 366/1355 [00:04<00:11, 86.27it/s]
 28%|██▊       | 375/1355 [00:04<00:11, 85.79it/s]
 28%|██▊       | 384/1355 [00:04<00:11, 85.61it/s]
 29%|██▉       | 393/1355 [00:04<00:11, 85.43it/s]
 30%|██▉       | 402/1355 [00:04<00:11, 85.11it/s]
 30%|███       | 411/1355 [00:05<00:11, 85.66it/s]
 31%|███       | 420/1355 [00:05<00:11, 84.67it/s]
 32%|███▏      | 429/1355 [00:05<00:10, 85.02it/s]
 32%|███▏      | 438/1355 [00:05<00:10, 84.20it/s]
 33%|███▎      | 447/1355 [00:05<00:10, 84.53it/s]
 34%|███▎      | 456/1355 [00:05<00:11, 81.47it/s]
 34%|███▍      | 465/1355 [00:05<00:10, 81.24it/s]
 35%|███▍      | 474/1355 [00:05<00:10, 81.95it/s]
 36%|███▌      | 483/1355 [00:05<00:10, 82.64it/s]
 36%|███▋      | 492/1355 [00:06<00:10, 83.13it/s]
 37%|███▋      | 501/1355 [00:06<00:10, 82.28it/s]
 38%|███▊      | 510/1355 [00:06<00:10, 79.36it/s]
 38%|███▊      | 518/1355 [00:06<00:10, 78.05it/s]
 39%|███▉      | 527/1355 [00:06<00:10, 79.16it/s]
 40%|███▉      | 536/1355 [00:06<00:10, 79.94it/s]
 40%|████      | 545/1355 [00:06<00:10, 78.66it/s]
 41%|████      | 554/1355 [00:06<00:10, 80.00it/s]
 42%|████▏     | 563/1355 [00:06<00:09, 81.81it/s]
 42%|████▏     | 572/1355 [00:07<00:09, 82.19it/s]
 43%|████▎     | 581/1355 [00:07<00:09, 80.37it/s]
 44%|████▎     | 590/1355 [00:07<00:09, 81.18it/s]
 44%|████▍     | 599/1355 [00:07<00:09, 80.25it/s]
 45%|████▍     | 608/1355 [00:07<00:09, 77.59it/s]
 46%|████▌     | 617/1355 [00:07<00:09, 79.00it/s]
 46%|████▌     | 625/1355 [00:07<00:09, 78.08it/s]
 47%|████▋     | 633/1355 [00:07<00:09, 78.24it/s]
 47%|████▋     | 642/1355 [00:07<00:08, 79.32it/s]
 48%|████▊     | 651/1355 [00:08<00:08, 80.44it/s]
 49%|████▊     | 660/1355 [00:08<00:08, 82.22it/s]
 49%|████▉     | 669/1355 [00:08<00:09, 76.21it/s]
 50%|█████     | 678/1355 [00:08<00:08, 77.55it/s]
 51%|█████     | 687/1355 [00:08<00:08, 78.58it/s]
 51%|█████▏    | 696/1355 [00:08<00:08, 79.87it/s]
 52%|█████▏    | 705/1355 [00:08<00:08, 78.11it/s]
 53%|█████▎    | 714/1355 [00:08<00:07, 80.50it/s]
 53%|█████▎    | 723/1355 [00:08<00:07, 80.12it/s]
 54%|█████▍    | 732/1355 [00:09<00:07, 80.88it/s]
 55%|█████▍    | 741/1355 [00:09<00:07, 78.39it/s]
 55%|█████▌    | 750/1355 [00:09<00:07, 79.64it/s]
 56%|█████▌    | 758/1355 [00:09<00:07, 78.30it/s]
 57%|█████▋    | 766/1355 [00:09<00:07, 78.43it/s]
 57%|█████▋    | 775/1355 [00:09<00:07, 81.16it/s]
 58%|█████▊    | 784/1355 [00:09<00:07, 76.55it/s]
 59%|█████▊    | 793/1355 [00:09<00:07, 78.18it/s]
 59%|█████▉    | 801/1355 [00:09<00:07, 78.59it/s]
 60%|█████▉    | 810/1355 [00:10<00:06, 81.02it/s]
 60%|██████    | 819/1355 [00:10<00:06, 81.88it/s]
 61%|██████    | 828/1355 [00:10<00:06, 81.33it/s]
 62%|██████▏   | 837/1355 [00:10<00:06, 80.79it/s]
 62%|██████▏   | 846/1355 [00:10<00:06, 80.11it/s]
 63%|██████▎   | 855/1355 [00:10<00:06, 80.41it/s]
 64%|██████▍   | 864/1355 [00:10<00:06, 81.38it/s]
 64%|██████▍   | 873/1355 [00:10<00:05, 82.00it/s]
 65%|██████▌   | 882/1355 [00:10<00:05, 82.54it/s]
 66%|██████▌   | 891/1355 [00:11<00:05, 82.98it/s]
 66%|██████▋   | 900/1355 [00:11<00:05, 82.84it/s]
 67%|██████▋   | 909/1355 [00:11<00:05, 83.28it/s]
 68%|██████▊   | 918/1355 [00:11<00:05, 83.30it/s]
 68%|██████▊   | 927/1355 [00:11<00:05, 80.41it/s]
 69%|██████▉   | 936/1355 [00:11<00:05, 79.47it/s]
 70%|██████▉   | 945/1355 [00:11<00:05, 80.81it/s]
 70%|███████   | 954/1355 [00:11<00:04, 81.93it/s]
 71%|███████   | 963/1355 [00:11<00:04, 80.07it/s]
 72%|███████▏  | 972/1355 [00:12<00:05, 76.41it/s]
 72%|███████▏  | 980/1355 [00:12<00:04, 77.20it/s]
 73%|███████▎  | 988/1355 [00:12<00:04, 77.77it/s]
 74%|███████▎  | 996/1355 [00:12<00:04, 78.35it/s]
 74%|███████▍  | 1005/1355 [00:12<00:04, 80.44it/s]
 75%|███████▍  | 1014/1355 [00:12<00:04, 79.12it/s]
 75%|███████▌  | 1022/1355 [00:12<00:04, 78.30it/s]
 76%|███████▌  | 1030/1355 [00:12<00:04, 78.23it/s]
 77%|███████▋  | 1039/1355 [00:12<00:03, 79.57it/s]
 77%|███████▋  | 1048/1355 [00:13<00:03, 81.05it/s]
 78%|███████▊  | 1057/1355 [00:13<00:03, 79.79it/s]
 79%|███████▊  | 1066/1355 [00:13<00:03, 81.05it/s]
 79%|███████▉  | 1075/1355 [00:13<00:03, 81.44it/s]
 80%|████████  | 1084/1355 [00:13<00:03, 80.30it/s]
 81%|████████  | 1093/1355 [00:13<00:03, 80.00it/s]
 81%|████████▏ | 1102/1355 [00:13<00:03, 80.32it/s]
 82%|████████▏ | 1111/1355 [00:13<00:03, 76.79it/s]
 83%|████████▎ | 1119/1355 [00:13<00:03, 76.96it/s]
 83%|████████▎ | 1127/1355 [00:14<00:02, 76.68it/s]
 84%|████████▍ | 1136/1355 [00:14<00:02, 76.22it/s]
 85%|████████▍ | 1145/1355 [00:14<00:02, 78.00it/s]
 85%|████████▌ | 1153/1355 [00:14<00:02, 77.95it/s]
 86%|████████▌ | 1161/1355 [00:14<00:02, 76.16it/s]
 86%|████████▋ | 1169/1355 [00:14<00:02, 76.13it/s]
 87%|████████▋ | 1177/1355 [00:14<00:02, 77.20it/s]
 88%|████████▊ | 1186/1355 [00:14<00:02, 77.96it/s]
 88%|████████▊ | 1195/1355 [00:14<00:02, 78.39it/s]
 89%|████████▉ | 1203/1355 [00:15<00:01, 78.18it/s]
 89%|████████▉ | 1212/1355 [00:15<00:01, 79.03it/s]
 90%|█████████ | 1220/1355 [00:15<00:01, 76.62it/s]
 91%|█████████ | 1229/1355 [00:15<00:01, 77.65it/s]
 91%|█████████▏| 1237/1355 [00:15<00:01, 77.55it/s]
 92%|█████████▏| 1245/1355 [00:15<00:01, 77.78it/s]
 92%|█████████▏| 1253/1355 [00:15<00:01, 78.26it/s]
 93%|█████████▎| 1262/1355 [00:15<00:01, 79.17it/s]
 94%|█████████▎| 1270/1355 [00:15<00:01, 79.22it/s]
 94%|█████████▍| 1279/1355 [00:16<00:00, 80.05it/s]
 95%|█████████▌| 1288/1355 [00:16<00:00, 81.28it/s]
 96%|█████████▌| 1297/1355 [00:16<00:00, 80.80it/s]
 96%|█████████▋| 1306/1355 [00:16<00:00, 79.04it/s]
 97%|█████████▋| 1314/1355 [00:16<00:00, 79.26it/s]
 98%|█████████▊| 1322/1355 [00:16<00:00, 79.15it/s]
 98%|█████████▊| 1330/1355 [00:16<00:00, 79.26it/s]
 99%|█████████▊| 1338/1355 [00:16<00:00, 76.91it/s]
 99%|█████████▉| 1346/1355 [00:16<00:00, 77.65it/s]
+[INFO|trainer.py:3503] 2024-09-05 22:05:23,351 >> Saving model checkpoint to /content/dissertation/scripts/ner/output
+[INFO|configuration_utils.py:472] 2024-09-05 22:05:23,353 >> Configuration saved in /content/dissertation/scripts/ner/output/config.json
+[INFO|modeling_utils.py:2799] 2024-09-05 22:05:24,590 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
+[INFO|tokenization_utils_base.py:2684] 2024-09-05 22:05:24,591 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2693] 2024-09-05 22:05:24,591 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+***** predict metrics *****
+  predict_accuracy           =      0.992
+  predict_f1                 =     0.8562
+  predict_loss               =      0.047
+  predict_precision          =     0.8359
+  predict_recall             =     0.8774
+  predict_runtime            = 0:00:23.00
+  predict_samples_per_second =    471.149
+  predict_steps_per_second   =     58.904
+

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.5369860684670966e+16,
+    "train_loss": 0.009731636565258824,
+    "train_runtime": 1413.3317,
+    "train_samples": 32675,
+    "train_samples_per_second": 231.191,
+    "train_steps_per_second": 3.616
+}

trainer_state.json

@@ -0,0 +1,232 @@
+{
+  "best_metric": 0.8506998444790046,
+  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-4599",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 5110,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.9784735812133072,
+      "grad_norm": 0.2284504771232605,
+      "learning_rate": 4.510763209393347e-05,
+      "loss": 0.0542,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9894745923561384,
+      "eval_f1": 0.7720770551211139,
+      "eval_loss": 0.027118448168039322,
+      "eval_precision": 0.7485207100591716,
+      "eval_recall": 0.7971642378889326,
+      "eval_runtime": 15.6605,
+      "eval_samples_per_second": 469.589,
+      "eval_steps_per_second": 58.746,
+      "step": 511
+    },
+    {
+      "epoch": 1.9569471624266144,
+      "grad_norm": 0.6956239342689514,
+      "learning_rate": 4.021526418786693e-05,
+      "loss": 0.0184,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9906364133091766,
+      "eval_f1": 0.8196286472148541,
+      "eval_loss": 0.027706459164619446,
+      "eval_precision": 0.7897042716319824,
+      "eval_recall": 0.8519102008664828,
+      "eval_runtime": 15.4779,
+      "eval_samples_per_second": 475.128,
+      "eval_steps_per_second": 59.439,
+      "step": 1022
+    },
+    {
+      "epoch": 2.935420743639922,
+      "grad_norm": 0.465753972530365,
+      "learning_rate": 3.53228962818004e-05,
+      "loss": 0.0103,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9913726166853594,
+      "eval_f1": 0.8360814742967991,
+      "eval_loss": 0.03051806427538395,
+      "eval_precision": 0.8237767584097859,
+      "eval_recall": 0.848759354076408,
+      "eval_runtime": 15.7367,
+      "eval_samples_per_second": 467.314,
+      "eval_steps_per_second": 58.462,
+      "step": 1533
+    },
+    {
+      "epoch": 3.9138943248532287,
+      "grad_norm": 0.46147385239601135,
+      "learning_rate": 3.0430528375733857e-05,
+      "loss": 0.0058,
+      "step": 2000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9913208523854715,
+      "eval_f1": 0.8364197530864197,
+      "eval_loss": 0.032038912177085876,
+      "eval_precision": 0.8196597353497165,
+      "eval_recall": 0.8538794801102796,
+      "eval_runtime": 15.567,
+      "eval_samples_per_second": 472.41,
+      "eval_steps_per_second": 59.099,
+      "step": 2044
+    },
+    {
+      "epoch": 4.892367906066536,
+      "grad_norm": 0.1297295242547989,
+      "learning_rate": 2.553816046966732e-05,
+      "loss": 0.0041,
+      "step": 2500
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9916544445403043,
+      "eval_f1": 0.8406766325727772,
+      "eval_loss": 0.037402600049972534,
+      "eval_precision": 0.8396856581532417,
+      "eval_recall": 0.8416699487987397,
+      "eval_runtime": 15.6627,
+      "eval_samples_per_second": 469.523,
+      "eval_steps_per_second": 58.738,
+      "step": 2555
+    },
+    {
+      "epoch": 5.870841487279844,
+      "grad_norm": 0.04726821184158325,
+      "learning_rate": 2.064579256360078e-05,
+      "loss": 0.0026,
+      "step": 3000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9916601961291807,
+      "eval_f1": 0.8435241258058215,
+      "eval_loss": 0.04270775243639946,
+      "eval_precision": 0.8368217054263566,
+      "eval_recall": 0.8503347774714455,
+      "eval_runtime": 15.5727,
+      "eval_samples_per_second": 472.237,
+      "eval_steps_per_second": 59.078,
+      "step": 3066
+    },
+    {
+      "epoch": 6.8493150684931505,
+      "grad_norm": 0.25813549757003784,
+      "learning_rate": 1.5753424657534248e-05,
+      "loss": 0.0015,
+      "step": 3500
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9911828142524373,
+      "eval_f1": 0.8397768801692633,
+      "eval_loss": 0.045068006962537766,
+      "eval_precision": 0.8206766917293233,
+      "eval_recall": 0.85978731784167,
+      "eval_runtime": 15.5993,
+      "eval_samples_per_second": 471.433,
+      "eval_steps_per_second": 58.977,
+      "step": 3577
+    },
+    {
+      "epoch": 7.8277886497064575,
+      "grad_norm": 0.22853463888168335,
+      "learning_rate": 1.086105675146771e-05,
+      "loss": 0.0013,
+      "step": 4000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.991637189773675,
+      "eval_f1": 0.8470906630581867,
+      "eval_loss": 0.04481955245137215,
+      "eval_precision": 0.8318147304479878,
+      "eval_recall": 0.8629381646317448,
+      "eval_runtime": 15.9638,
+      "eval_samples_per_second": 460.669,
+      "eval_steps_per_second": 57.631,
+      "step": 4088
+    },
+    {
+      "epoch": 8.806262230919765,
+      "grad_norm": 0.009220785461366177,
+      "learning_rate": 5.9686888454011745e-06,
+      "loss": 0.0007,
+      "step": 4500
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.9916544445403043,
+      "eval_f1": 0.8506998444790046,
+      "eval_loss": 0.04956069961190224,
+      "eval_precision": 0.8399232245681382,
+      "eval_recall": 0.8617565970854667,
+      "eval_runtime": 15.6877,
+      "eval_samples_per_second": 468.774,
+      "eval_steps_per_second": 58.645,
+      "step": 4599
+    },
+    {
+      "epoch": 9.784735812133073,
+      "grad_norm": 0.5473453402519226,
+      "learning_rate": 1.076320939334638e-06,
+      "loss": 0.0006,
+      "step": 5000
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.9916486929514279,
+      "eval_f1": 0.8506998444790046,
+      "eval_loss": 0.05025585740804672,
+      "eval_precision": 0.8399232245681382,
+      "eval_recall": 0.8617565970854667,
+      "eval_runtime": 16.4037,
+      "eval_samples_per_second": 448.314,
+      "eval_steps_per_second": 56.085,
+      "step": 5110
+    },
+    {
+      "epoch": 10.0,
+      "step": 5110,
+      "total_flos": 1.5369860684670966e+16,
+      "train_loss": 0.009731636565258824,
+      "train_runtime": 1413.3317,
+      "train_samples_per_second": 231.191,
+      "train_steps_per_second": 3.616
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 5110,
+  "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": 1.5369860684670966e+16,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}