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:
-- multi-train-distemist-dev-ner
+- Rodrigo1771/multi-train-distemist-dev-ner
 metrics:
 - precision
 - recall
@@ -17,24 +18,24 @@ model-index:
       name: Token Classification
       type: token-classification
     dataset:
-      name: multi-train-distemist-dev-ner
-      type: multi-train-distemist-dev-ner
+      name: Rodrigo1771/multi-train-distemist-dev-ner
+      type: Rodrigo1771/multi-train-distemist-dev-ner
       config: MultiTrainDisTEMISTDevNER
       split: validation
       args: MultiTrainDisTEMISTDevNER
     metrics:
     - name: Precision
       type: precision
-      value: 0.3145413870246085
+      value: 0.32143181611701643
     - name: Recall
       type: recall
-      value: 0.8224145999064109
+      value: 0.8277959756668226
     - name: F1
       type: f1
-      value: 0.45504563402161957
+      value: 0.46305870034683594
     - name: Accuracy
       type: accuracy
-      value: 0.8526409018818958
+      value: 0.8559776451929613
 ---
 
 <!-- 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 multi-train-distemist-dev-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/multi-train-distemist-dev-ner dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1188
-- Precision: 0.3145
-- Recall: 0.8224
-- F1: 0.4550
-- Accuracy: 0.8526
+- Loss: 0.9499
+- Precision: 0.3214
+- Recall: 0.8278
+- F1: 0.4631
+- Accuracy: 0.8560
 
 ## Model description
 

all_results.json

@@ -1,19 +1,26 @@
 {
-    "eval_accuracy": 0.018743434648577764,
-    "eval_f1": 0.008761828065230522,
-    "eval_loss": 2.3735408782958984,
-    "eval_precision": 0.004537076421380973,
-    "eval_recall": 0.1272812353766963,
-    "eval_runtime": 16.9354,
+    "epoch": 9.997061416397296,
+    "eval_accuracy": 0.8559776451929613,
+    "eval_f1": 0.46305870034683594,
+    "eval_loss": 0.9498798847198486,
+    "eval_precision": 0.32143181611701643,
+    "eval_recall": 0.8277959756668226,
+    "eval_runtime": 16.1603,
     "eval_samples": 6807,
-    "eval_samples_per_second": 401.939,
-    "eval_steps_per_second": 50.25,
-    "predict_accuracy": 0.018743434648577764,
-    "predict_f1": 0.008761828065230522,
-    "predict_loss": 2.3735408782958984,
-    "predict_precision": 0.004537076421380973,
-    "predict_recall": 0.1272812353766963,
-    "predict_runtime": 16.1794,
-    "predict_samples_per_second": 420.719,
-    "predict_steps_per_second": 52.598
+    "eval_samples_per_second": 421.218,
+    "eval_steps_per_second": 52.66,
+    "predict_accuracy": 0.8559776451929613,
+    "predict_f1": 0.46305870034683594,
+    "predict_loss": 0.9498798847198486,
+    "predict_precision": 0.32143181611701643,
+    "predict_recall": 0.8277959756668226,
+    "predict_runtime": 16.4367,
+    "predict_samples_per_second": 414.134,
+    "predict_steps_per_second": 51.774,
+    "total_flos": 6700722040732752.0,
+    "train_loss": 0.08913132946046923,
+    "train_runtime": 3311.208,
+    "train_samples": 27224,
+    "train_samples_per_second": 82.218,
+    "train_steps_per_second": 5.137
 }

eval_results.json

@@ -1,11 +1,12 @@
 {
-    "eval_accuracy": 0.018743434648577764,
-    "eval_f1": 0.008761828065230522,
-    "eval_loss": 2.3735408782958984,
-    "eval_precision": 0.004537076421380973,
-    "eval_recall": 0.1272812353766963,
-    "eval_runtime": 16.9354,
+    "epoch": 9.997061416397296,
+    "eval_accuracy": 0.8559776451929613,
+    "eval_f1": 0.46305870034683594,
+    "eval_loss": 0.9498798847198486,
+    "eval_precision": 0.32143181611701643,
+    "eval_recall": 0.8277959756668226,
+    "eval_runtime": 16.1603,
     "eval_samples": 6807,
-    "eval_samples_per_second": 401.939,
-    "eval_steps_per_second": 50.25
+    "eval_samples_per_second": 421.218,
+    "eval_steps_per_second": 52.66
 }

predict_results.json

@@ -1,10 +1,10 @@
 {
-    "predict_accuracy": 0.018743434648577764,
-    "predict_f1": 0.008761828065230522,
-    "predict_loss": 2.3735408782958984,
-    "predict_precision": 0.004537076421380973,
-    "predict_recall": 0.1272812353766963,
-    "predict_runtime": 16.1794,
-    "predict_samples_per_second": 420.719,
-    "predict_steps_per_second": 52.598
+    "predict_accuracy": 0.8559776451929613,
+    "predict_f1": 0.46305870034683594,
+    "predict_loss": 0.9498798847198486,
+    "predict_precision": 0.32143181611701643,
+    "predict_recall": 0.8277959756668226,
+    "predict_runtime": 16.4367,
+    "predict_samples_per_second": 414.134,
+    "predict_steps_per_second": 51.774
 }

tb/events.out.tfevents.1715603564.dff07dfba241.5879.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d165a43ac9f9c5f183a1937575bd2f2d93ca6034ab66002728aec974fcf320a2
+size 569

train.log

@@ -1640,3 +1640,53 @@ Training completed. Do not forget to share your model on huggingface.co/models =
 [INFO|modeling_utils.py:2590] 2024-05-13 12:32:23,238 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
 [INFO|tokenization_utils_base.py:2488] 2024-05-13 12:32:23,239 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
 [INFO|tokenization_utils_base.py:2497] 2024-05-13 12:32:23,239 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+{'eval_loss': 1.1188451051712036, 'eval_precision': 0.3145413870246085, 'eval_recall': 0.8224145999064109, 'eval_f1': 0.45504563402161957, 'eval_accuracy': 0.8526409018818958, 'eval_runtime': 16.3165, 'eval_samples_per_second': 417.185, 'eval_steps_per_second': 52.156, 'epoch': 10.0}
+{'train_runtime': 3311.208, 'train_samples_per_second': 82.218, 'train_steps_per_second': 5.137, 'train_loss': 0.08913132946046923, 'epoch': 10.0}
+***** train metrics *****
+  epoch                    =     9.9971
+  total_flos               =  6240533GF
+  train_loss               =     0.0891
+  train_runtime            = 0:55:11.20
+  train_samples            =      27224
+  train_samples_per_second =     82.218
+  train_steps_per_second   =      5.137
+05/13/2024 12:32:28 - INFO - __main__ -   *** Evaluate ***
+[INFO|trainer.py:786] 2024-05-13 12:32:28,316 >> 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:3614] 2024-05-13 12:32:28,318 >> ***** Running Evaluation *****
+[INFO|trainer.py:3616] 2024-05-13 12:32:28,318 >>   Num examples = 6807
+[INFO|trainer.py:3619] 2024-05-13 12:32:28,318 >>   Batch size = 8
+
  0%|          | 0/851 [00:00<?, ?it/s]
  1%|          | 10/851 [00:00<00:09, 92.31it/s]
  2%|▏         | 20/851 [00:00<00:10, 77.97it/s]
  3%|▎         | 28/851 [00:00<00:11, 73.38it/s]
  4%|▍         | 36/851 [00:00<00:11, 72.05it/s]
  5%|▌         | 44/851 [00:00<00:11, 73.19it/s]
  6%|▌         | 52/851 [00:00<00:10, 73.90it/s]
  7%|▋         | 60/851 [00:00<00:10, 74.36it/s]
  8%|▊         | 68/851 [00:00<00:11, 68.74it/s]
  9%|▉         | 75/851 [00:01<00:11, 68.45it/s]
 10%|▉         | 83/851 [00:01<00:11, 69.37it/s]
 11%|█         | 91/851 [00:01<00:10, 70.51it/s]
 12%|█▏        | 99/851 [00:01<00:10, 69.26it/s]
 13%|█▎        | 107/851 [00:01<00:10, 67.76it/s]
 13%|█▎        | 114/851 [00:01<00:10, 67.63it/s]
 14%|█▍        | 122/851 [00:01<00:10, 70.32it/s]
 15%|█▌        | 130/851 [00:01<00:11, 64.72it/s]
 16%|█▌        | 137/851 [00:01<00:11, 64.04it/s]
 17%|█▋        | 145/851 [00:02<00:10, 66.06it/s]
 18%|█▊        | 152/851 [00:02<00:10, 64.75it/s]
 19%|█▉        | 160/851 [00:02<00:10, 68.49it/s]
 20%|█▉        | 167/851 [00:02<00:09, 68.88it/s]
 21%|██        | 175/851 [00:02<00:09, 70.11it/s]
 22%|██▏       | 183/851 [00:02<00:09, 70.82it/s]
 22%|██▏       | 191/851 [00:02<00:09, 71.05it/s]
 23%|██▎       | 199/851 [00:02<00:09, 72.33it/s]
 24%|██▍       | 207/851 [00:02<00:09, 71.19it/s]
 25%|██▌       | 215/851 [00:03<00:09, 67.78it/s]
 26%|██▌       | 223/851 [00:03<00:09, 68.86it/s]
 27%|██▋       | 231/851 [00:03<00:08, 69.60it/s]
 28%|██▊       | 238/851 [00:03<00:09, 66.60it/s]
 29%|██▉       | 245/851 [00:03<00:09, 64.21it/s]
 30%|██▉       | 253/851 [00:03<00:08, 67.82it/s]
 31%|███       | 261/851 [00:03<00:08, 70.71it/s]
 32%|███▏      | 269/851 [00:03<00:08, 69.23it/s]
 33%|███▎      | 277/851 [00:03<00:07, 71.99it/s]
 33%|███▎      | 285/851 [00:04<00:07, 73.42it/s]
 34%|███▍      | 293/851 [00:04<00:07, 71.22it/s]
 35%|███▌      | 301/851 [00:04<00:07, 72.04it/s]
 36%|███▋      | 310/851 [00:04<00:07, 74.21it/s]
 37%|███▋      | 318/851 [00:04<00:07, 70.47it/s]
 38%|███▊      | 326/851 [00:04<00:07, 69.69it/s]
 39%|███▉      | 334/851 [00:04<00:07, 69.49it/s]
 40%|████      | 342/851 [00:04<00:07, 70.77it/s]
 41%|████      | 350/851 [00:05<00:07, 70.56it/s]
 42%|████▏     | 358/851 [00:05<00:07, 65.68it/s]
 43%|████▎     | 366/851 [00:05<00:07, 66.28it/s]
 44%|████▍     | 373/851 [00:05<00:07, 66.55it/s]
 45%|████▍     | 380/851 [00:05<00:07, 64.58it/s]
 46%|████▌     | 388/851 [00:05<00:06, 67.42it/s]
 46%|████▋     | 395/851 [00:05<00:06, 67.26it/s]
 47%|████▋     | 402/851 [00:05<00:06, 67.54it/s]
 48%|████▊     | 409/851 [00:05<00:06, 64.14it/s]
 49%|████▉     | 416/851 [00:06<00:06, 65.62it/s]
 50%|████▉     | 424/851 [00:06<00:06, 67.91it/s]
 51%|█████     | 431/851 [00:06<00:06, 66.25it/s]
 52%|█████▏    | 439/851 [00:06<00:06, 68.51it/s]
 52%|██��██▏    | 446/851 [00:06<00:05, 67.97it/s]
 53%|█████▎    | 454/851 [00:06<00:05, 68.93it/s]
 54%|█████▍    | 461/851 [00:06<00:05, 68.17it/s]
 55%|█████▍    | 468/851 [00:06<00:05, 65.16it/s]
 56%|█████▌    | 475/851 [00:06<00:06, 60.96it/s]
 57%|█████▋    | 482/851 [00:07<00:05, 61.68it/s]
 57%|█████▋    | 489/851 [00:07<00:05, 61.89it/s]
 58%|█████▊    | 497/851 [00:07<00:05, 66.34it/s]
 59%|█████▉    | 505/851 [00:07<00:04, 69.56it/s]
 60%|██████    | 513/851 [00:07<00:04, 68.23it/s]
 61%|██████    | 521/851 [00:07<00:04, 66.68it/s]
 62%|██████▏   | 528/851 [00:07<00:05, 63.84it/s]
 63%|██████▎   | 536/851 [00:07<00:04, 66.25it/s]
 64%|██████▍   | 544/851 [00:07<00:04, 68.40it/s]
 65%|██████▍   | 551/851 [00:08<00:04, 65.47it/s]
 66%|██████▌   | 558/851 [00:08<00:04, 66.51it/s]
 67%|██████▋   | 566/851 [00:08<00:04, 69.07it/s]
 67%|██████▋   | 573/851 [00:08<00:04, 69.29it/s]
 68%|██████▊   | 580/851 [00:08<00:04, 67.14it/s]
 69%|██████▉   | 587/851 [00:08<00:04, 63.62it/s]
 70%|██████▉   | 594/851 [00:08<00:03, 65.03it/s]
 71%|███████   | 601/851 [00:08<00:03, 64.50it/s]
 71%|███████▏  | 608/851 [00:08<00:03, 65.17it/s]
 72%|███████▏  | 615/851 [00:09<00:03, 60.27it/s]
 73%|███████▎  | 623/851 [00:09<00:03, 64.78it/s]
 74%|███████▍  | 630/851 [00:09<00:03, 61.49it/s]
 75%|███████▍  | 638/851 [00:09<00:03, 64.58it/s]
 76%|███████▌  | 645/851 [00:09<00:03, 59.82it/s]
 77%|███████▋  | 653/851 [00:09<00:03, 62.97it/s]
 78%|███████▊  | 660/851 [00:09<00:02, 64.81it/s]
 78%|███████▊  | 667/851 [00:09<00:02, 65.46it/s]
 79%|███████▉  | 674/851 [00:09<00:02, 64.18it/s]
 80%|████████  | 681/851 [00:10<00:02, 65.23it/s]
 81%|████████  | 689/851 [00:10<00:02, 69.23it/s]
 82%|████████▏ | 697/851 [00:10<00:02, 70.21it/s]
 83%|████████▎ | 705/851 [00:10<00:02, 71.66it/s]
 84%|████████▍ | 713/851 [00:10<00:01, 72.76it/s]
 85%|████████▍ | 721/851 [00:10<00:01, 69.62it/s]
 86%|████████▌ | 729/851 [00:10<00:01, 70.75it/s]
 87%|████████▋ | 737/851 [00:10<00:01, 71.16it/s]
 88%|████████▊ | 745/851 [00:10<00:01, 71.47it/s]
 88%|████████▊ | 753/851 [00:11<00:01, 70.42it/s]
 89%|████████▉ | 761/851 [00:11<00:01, 72.99it/s]
 90%|█████████ | 769/851 [00:11<00:01, 67.29it/s]
 91%|█████████ | 776/851 [00:11<00:01, 67.64it/s]
 92%|█████████▏| 783/851 [00:11<00:01, 65.98it/s]
 93%|█████████▎| 790/851 [00:11<00:00, 66.89it/s]
 94%|█████████▎| 797/851 [00:11<00:00, 67.62it/s]
 95%|█████████▍| 805/851 [00:11<00:00, 70.55it/s]
 96%|█████████▌| 813/851 [00:11<00:00, 67.17it/s]
 96%|█████████▋| 820/851 [00:12<00:00, 67.58it/s]
 97%|█████████▋| 828/851 [00:12<00:00, 68.89it/s]
 98%|█████████▊| 835/851 [00:12<00:00, 68.68it/s]
 99%|█████████▉| 842/851 [00:12<00:00, 65.37it/s]
+  _warn_prf(average, modifier, msg_start, len(result))
+
+***** eval metrics *****
+  epoch                   =     9.9971
+  eval_accuracy           =      0.856
+  eval_f1                 =     0.4631
+  eval_loss               =     0.9499
+  eval_precision          =     0.3214
+  eval_recall             =     0.8278
+  eval_runtime            = 0:00:16.16
+  eval_samples            =       6807
+  eval_samples_per_second =    421.218
+  eval_steps_per_second   =      52.66
+05/13/2024 12:32:44 - INFO - __main__ -   *** Predict ***
+[INFO|trainer.py:786] 2024-05-13 12:32:44,481 >> 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:3614] 2024-05-13 12:32:44,483 >> ***** Running Prediction *****
+[INFO|trainer.py:3616] 2024-05-13 12:32:44,484 >>   Num examples = 6807
+[INFO|trainer.py:3619] 2024-05-13 12:32:44,484 >>   Batch size = 8
+
  0%|          | 0/851 [00:00<?, ?it/s]
  1%|          | 10/851 [00:00<00:09, 91.52it/s]
  2%|▏         | 20/851 [00:00<00:10, 77.24it/s]
  3%|▎         | 28/851 [00:00<00:11, 73.46it/s]
  4%|▍         | 36/851 [00:00<00:11, 71.95it/s]
  5%|▌         | 44/851 [00:00<00:11, 72.24it/s]
  6%|▌         | 52/851 [00:00<00:10, 72.93it/s]
  7%|▋         | 60/851 [00:00<00:10, 73.18it/s]
  8%|▊         | 68/851 [00:00<00:11, 67.50it/s]
  9%|▉         | 75/851 [00:01<00:11, 67.33it/s]
 10%|▉         | 83/851 [00:01<00:11, 68.35it/s]
 11%|█         | 91/851 [00:01<00:10, 69.50it/s]
 12%|█▏        | 98/851 [00:01<00:11, 68.01it/s]
 12%|█▏        | 105/851 [00:01<00:10, 68.42it/s]
 13%|█▎        | 112/851 [00:01<00:11, 66.54it/s]
 14%|█▍        | 119/851 [00:01<00:10, 67.31it/s]
 15%|█▍        | 126/851 [00:01<00:11, 63.93it/s]
 16%|█▌        | 133/851 [00:01<00:11, 64.45it/s]
 16%|█▋        | 140/851 [00:02<00:11, 64.55it/s]
 17%|█▋        | 147/851 [00:02<00:10, 65.85it/s]
 18%|█▊        | 154/851 [00:02<00:10, 64.70it/s]
 19%|█▉        | 162/851 [00:02<00:10, 66.76it/s]
 20%|█▉        | 170/851 [00:02<00:10, 67.85it/s]
 21%|██        | 178/851 [00:02<00:09, 69.49it/s]
 22%|██▏       | 186/851 [00:02<00:09, 69.84it/s]
 23%|██▎       | 194/851 [00:02<00:09, 70.57it/s]
 24%|██▍       | 203/851 [00:02<00:08, 72.61it/s]
 25%|██▍       | 211/851 [00:03<00:09, 69.42it/s]
 26%|██▌       | 218/851 [00:03<00:09, 66.85it/s]
 27%|██▋       | 226/851 [00:03<00:09, 68.61it/s]
 27%|██▋       | 234/851 [00:03<00:08, 69.75it/s]
 28%|██▊       | 241/851 [00:03<00:09, 64.31it/s]
 29%|██▉       | 249/851 [00:03<00:09, 66.20it/s]
 30%|███       | 257/851 [00:03<00:08, 69.02it/s]
 31%|███       | 265/851 [00:03<00:08, 71.69it/s]
 32%|███▏      | 273/851 [00:03<00:08, 69.07it/s]
 33%|███▎      | 281/851 [00:04<00:07, 71.84it/s]
 34%|███▍      | 289/851 [00:04<00:08, 69.07it/s]
 35%|███▍      | 297/851 [00:04<00:07, 70.91it/s]
 36%|███▌      | 305/851 [00:04<00:07, 73.25it/s]
 37%|███▋      | 313/851 [00:04<00:07, 68.13it/s]
 38%|███▊      | 322/851 [00:04<00:07, 71.73it/s]
 39%|███▉      | 330/851 [00:04<00:07, 69.46it/s]
 40%|███▉      | 338/851 [00:04<00:07, 69.29it/s]
 41%|████      | 346/851 [00:05<00:07, 69.47it/s]
 41%|████▏     | 353/851 [00:05<00:07, 67.00it/s]
 42%|████▏     | 360/851 [00:05<00:07, 65.48it/s]
 43%|████▎     | 368/851 [00:05<00:07, 66.78it/s]
 44%|████▍     | 375/851 [00:05<00:07, 66.05it/s]
 45%|████▍     | 382/851 [00:05<00:07, 64.23it/s]
 46%|████▌     | 390/851 [00:05<00:06, 67.18it/s]
 47%|████▋     | 398/851 [00:05<00:06, 67.83it/s]
 48%|████▊     | 405/851 [00:05<00:07, 62.69it/s]
 48%|████▊     | 412/851 [00:06<00:06, 63.96it/s]
 49%|████▉     | 420/851 [00:06<00:06, 67.12it/s]
 50%|█████     | 427/851 [00:06<00:06, 64.71it/s]
 51%|█████     | 435/851 [00:06<00:06, 67.20it/s]
 52%|█████▏    | 443/851 [00:06<00:05, 68.92it/s]
 53%|█████▎    | 451/851 [00:06<00:05, 69.97it/s]
 54%|█████▍    | 459/851 [00:06<00:05, 70.79it/s]
 55%|█████▍    | 467/851 [00:06<00:05, 64.99it/s]
 56%|█████▌    | 474/851 [00:06<00:06, 60.17it/s]
 57%|█████▋    | 481/851 [00:07<00:06, 61.07it/s]
 57%|█████▋    | 488/851 [00:07<00:05, 62.68it/s]
 58%|█████▊    | 496/851 [00:07<00:05, 65.90it/s]
 59%|█████▉    | 504/851 [00:07<00:05, 69.35it/s]
 60%|██████    | 512/851 [00:07<00:05, 67.73it/s]
 61%|██████    | 520/851 [00:07<00:04, 68.81it/s]
 62%|██████▏   | 527/851 [00:07<00:05, 62.95it/s]
 63%|██████▎   | 535/851 [00:07<00:04, 65.45it/s]
 64%|██████▍   | 543/851 [00:07<00:04, 67.58it/s]
 65%|██████▍   | 550/851 [00:08<00:04, 65.74it/s]
 65%|██████▌   | 557/851 [00:08<00:04, 66.86it/s]
 66%|██████▋   | 565/851 [00:08<00:04, 70.26it/s]
 67%|██████▋   | 573/851 [00:08<00:03, 70.03it/s]
 68%|██████▊   | 581/851 [00:08<00:03, 68.47it/s]
 69%|██████▉   | 588/851 [00:08<00:04, 64.15it/s]
 70%|███████   | 596/851 [00:08<00:03, 65.33it/s]
 71%|███████   | 604/851 [00:08<00:03, 66.81it/s]
 72%|███████▏  | 611/851 [00:09<00:03, 66.06it/s]
 73%|███████▎  | 618/851 [00:09<00:03, 61.98it/s]
 73%|███████▎  | 625/851 [00:09<00:03, 63.84it/s]
 74%|███████▍  | 632/851 [00:09<00:03, 62.09it/s]
 75%|███████▌  | 639/851 [00:09<00:03, 62.91it/s]
 76%|███████▌  | 646/851 [00:09<00:03, 60.42it/s]
 77%|███████▋  | 654/851 [00:09<00:03, 63.78it/s]
 78%|███████▊  | 661/851 [00:09<00:02, 64.94it/s]
 78%|███████▊  | 668/851 [00:09<00:02, 66.04it/s]
 79%|███████▉  | 675/851 [00:10<00:02, 64.72it/s]
 80%|████████  | 682/851 [00:10<00:02, 65.98it/s]
 81%|████████  | 690/851 [00:10<00:02, 69.12it/s]
 82%|████████▏ | 698/851 [00:10<00:02, 70.40it/s]
 83%|████████▎ | 706/851 [00:10<00:02, 71.97it/s]
 84%|████████▍ | 714/851 [00:10<00:01, 71.53it/s]
 85%|████████▍ | 722/851 [00:10<00:01, 69.93it/s]
 86%|████████▌ | 730/851 [00:10<00:01, 71.53it/s]
 87%|████████▋ | 738/851 [00:10<00:01, 72.49it/s]
 88%|████████▊ | 746/851 [00:11<00:01, 71.54it/s]
 89%|████████▊ | 754/851 [00:11<00:01, 71.10it/s]
 90%|████████▉ | 762/851 [00:11<00:01, 72.35it/s]
 90%|█████████ | 770/851 [00:11<00:01, 67.47it/s]
 91%|█████████▏| 777/851 [00:11<00:01, 68.02it/s]
 92%|█████████▏| 784/851 [00:11<00:01, 66.04it/s]
 93%|█████████▎| 791/851 [00:11<00:00, 66.57it/s]
 94%|█████████▍| 798/851 [00:11<00:00, 67.52it/s]
 95%|█████████▍| 806/851 [00:11<00:00, 69.97it/s]
 96%|█████████▌| 814/851 [00:12<00:00, 67.07it/s]
 97%|█████████▋| 822/851 [00:12<00:00, 68.50it/s]
 97%|█████████▋| 829/851 [00:12<00:00, 68.73it/s]
 98%|█████████▊| 836/851 [00:12<00:00, 68.89it/s]
 99%|█████████▉| 843/851 [00:12<00:00, 64.90it/s]
+[INFO|trainer.py:3305] 2024-05-13 12:33:01,240 >> Saving model checkpoint to /content/dissertation/scripts/ner/output
+[INFO|configuration_utils.py:471] 2024-05-13 12:33:01,241 >> Configuration saved in /content/dissertation/scripts/ner/output/config.json
+[INFO|modeling_utils.py:2590] 2024-05-13 12:33:02,436 >> Model weights saved in /content/dissertation/scripts/ner/output/model.safetensors
+[INFO|tokenization_utils_base.py:2488] 2024-05-13 12:33:02,437 >> tokenizer config file saved in /content/dissertation/scripts/ner/output/tokenizer_config.json
+[INFO|tokenization_utils_base.py:2497] 2024-05-13 12:33:02,437 >> Special tokens file saved in /content/dissertation/scripts/ner/output/special_tokens_map.json
+***** predict metrics *****
+  predict_accuracy           =      0.856
+  predict_f1                 =     0.4631
+  predict_loss               =     0.9499
+  predict_precision          =     0.3214
+  predict_recall             =     0.8278
+  predict_runtime            = 0:00:16.43
+  predict_samples_per_second =    414.134
+  predict_steps_per_second   =     51.774
+

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.997061416397296,
+    "total_flos": 6700722040732752.0,
+    "train_loss": 0.08913132946046923,
+    "train_runtime": 3311.208,
+    "train_samples": 27224,
+    "train_samples_per_second": 82.218,
+    "train_steps_per_second": 5.137
+}

trainer_state.json

@@ -0,0 +1,388 @@
+{
+  "best_metric": 0.46305870034683594,
+  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-11910",
+  "epoch": 9.997061416397296,
+  "eval_steps": 500,
+  "global_step": 17010,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.29385836027034967,
+      "grad_norm": 1.9668159484863281,
+      "learning_rate": 4.853027630805409e-05,
+      "loss": 0.4174,
+      "step": 500
+    },
+    {
+      "epoch": 0.5877167205406993,
+      "grad_norm": 3.246731758117676,
+      "learning_rate": 4.7060552616108174e-05,
+      "loss": 0.2765,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8815750808110491,
+      "grad_norm": 2.936720609664917,
+      "learning_rate": 4.559082892416226e-05,
+      "loss": 0.2596,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9997061416397296,
+      "eval_accuracy": 0.8358953937837708,
+      "eval_f1": 0.3926881971617123,
+      "eval_loss": 0.431875616312027,
+      "eval_precision": 0.26165460347108727,
+      "eval_recall": 0.786616752456715,
+      "eval_runtime": 16.2639,
+      "eval_samples_per_second": 418.535,
+      "eval_steps_per_second": 52.325,
+      "step": 1701
+    },
+    {
+      "epoch": 1.1754334410813987,
+      "grad_norm": 2.302980661392212,
+      "learning_rate": 4.4121105232216346e-05,
+      "loss": 0.2135,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4692918013517484,
+      "grad_norm": 3.7150967121124268,
+      "learning_rate": 4.265138154027043e-05,
+      "loss": 0.1839,
+      "step": 2500
+    },
+    {
+      "epoch": 1.7631501616220981,
+      "grad_norm": 1.2019191980361938,
+      "learning_rate": 4.118165784832452e-05,
+      "loss": 0.1853,
+      "step": 3000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8644911466450625,
+      "eval_f1": 0.44846535316981634,
+      "eval_loss": 0.3841294050216675,
+      "eval_precision": 0.3142374154770849,
+      "eval_recall": 0.7828731867103416,
+      "eval_runtime": 16.2484,
+      "eval_samples_per_second": 418.933,
+      "eval_steps_per_second": 52.374,
+      "step": 3403
+    },
+    {
+      "epoch": 2.0570085218924477,
+      "grad_norm": 2.090308904647827,
+      "learning_rate": 3.971193415637861e-05,
+      "loss": 0.1684,
+      "step": 3500
+    },
+    {
+      "epoch": 2.3508668821627974,
+      "grad_norm": 0.9995286464691162,
+      "learning_rate": 3.824221046443269e-05,
+      "loss": 0.1222,
+      "step": 4000
+    },
+    {
+      "epoch": 2.644725242433147,
+      "grad_norm": 1.3526027202606201,
+      "learning_rate": 3.677248677248677e-05,
+      "loss": 0.1277,
+      "step": 4500
+    },
+    {
+      "epoch": 2.938583602703497,
+      "grad_norm": 2.2816500663757324,
+      "learning_rate": 3.530276308054086e-05,
+      "loss": 0.1254,
+      "step": 5000
+    },
+    {
+      "epoch": 2.9997061416397295,
+      "eval_accuracy": 0.843626202360437,
+      "eval_f1": 0.4435178651613316,
+      "eval_loss": 0.6409665942192078,
+      "eval_precision": 0.30552364118426867,
+      "eval_recall": 0.8088441740758072,
+      "eval_runtime": 16.2306,
+      "eval_samples_per_second": 419.394,
+      "eval_steps_per_second": 52.432,
+      "step": 5104
+    },
+    {
+      "epoch": 3.2324419629738466,
+      "grad_norm": 0.6440290808677673,
+      "learning_rate": 3.3833039388594945e-05,
+      "loss": 0.0929,
+      "step": 5500
+    },
+    {
+      "epoch": 3.5263003232441963,
+      "grad_norm": 3.7906153202056885,
+      "learning_rate": 3.2363315696649034e-05,
+      "loss": 0.0896,
+      "step": 6000
+    },
+    {
+      "epoch": 3.820158683514546,
+      "grad_norm": 1.0953032970428467,
+      "learning_rate": 3.0893592004703116e-05,
+      "loss": 0.0823,
+      "step": 6500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.843598739452527,
+      "eval_f1": 0.4336244267135767,
+      "eval_loss": 0.7241775393486023,
+      "eval_precision": 0.2964012711500472,
+      "eval_recall": 0.8074403369209172,
+      "eval_runtime": 16.5937,
+      "eval_samples_per_second": 410.216,
+      "eval_steps_per_second": 51.285,
+      "step": 6806
+    },
+    {
+      "epoch": 4.114017043784895,
+      "grad_norm": 1.7823286056518555,
+      "learning_rate": 2.9423868312757202e-05,
+      "loss": 0.0761,
+      "step": 7000
+    },
+    {
+      "epoch": 4.407875404055245,
+      "grad_norm": 1.165720820426941,
+      "learning_rate": 2.795414462081129e-05,
+      "loss": 0.0603,
+      "step": 7500
+    },
+    {
+      "epoch": 4.701733764325595,
+      "grad_norm": 4.1130452156066895,
+      "learning_rate": 2.648442092886537e-05,
+      "loss": 0.0589,
+      "step": 8000
+    },
+    {
+      "epoch": 4.9955921245959445,
+      "grad_norm": 0.3957385718822479,
+      "learning_rate": 2.501469723691946e-05,
+      "loss": 0.0597,
+      "step": 8500
+    },
+    {
+      "epoch": 4.99970614163973,
+      "eval_accuracy": 0.8501829716239504,
+      "eval_f1": 0.4494872643069798,
+      "eval_loss": 0.7755796909332275,
+      "eval_precision": 0.3133474771700028,
+      "eval_recall": 0.7948058025269069,
+      "eval_runtime": 16.1833,
+      "eval_samples_per_second": 420.619,
+      "eval_steps_per_second": 52.585,
+      "step": 8507
+    },
+    {
+      "epoch": 5.289450484866294,
+      "grad_norm": 1.0087623596191406,
+      "learning_rate": 2.3544973544973546e-05,
+      "loss": 0.0423,
+      "step": 9000
+    },
+    {
+      "epoch": 5.583308845136644,
+      "grad_norm": 2.162200450897217,
+      "learning_rate": 2.2075249853027632e-05,
+      "loss": 0.043,
+      "step": 9500
+    },
+    {
+      "epoch": 5.877167205406994,
+      "grad_norm": 1.1820895671844482,
+      "learning_rate": 2.0605526161081718e-05,
+      "loss": 0.0446,
+      "step": 10000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8483223596130476,
+      "eval_f1": 0.4512908099586152,
+      "eval_loss": 0.8560824394226074,
+      "eval_precision": 0.31372728103023106,
+      "eval_recall": 0.8036967711745437,
+      "eval_runtime": 16.2581,
+      "eval_samples_per_second": 418.684,
+      "eval_steps_per_second": 52.343,
+      "step": 10209
+    },
+    {
+      "epoch": 6.171025565677343,
+      "grad_norm": 2.190476894378662,
+      "learning_rate": 1.91358024691358e-05,
+      "loss": 0.0333,
+      "step": 10500
+    },
+    {
+      "epoch": 6.464883925947693,
+      "grad_norm": 1.5180469751358032,
+      "learning_rate": 1.766607877718989e-05,
+      "loss": 0.0333,
+      "step": 11000
+    },
+    {
+      "epoch": 6.758742286218043,
+      "grad_norm": 1.4279770851135254,
+      "learning_rate": 1.6196355085243976e-05,
+      "loss": 0.0325,
+      "step": 11500
+    },
+    {
+      "epoch": 6.99970614163973,
+      "eval_accuracy": 0.8559776451929613,
+      "eval_f1": 0.46305870034683594,
+      "eval_loss": 0.9498798847198486,
+      "eval_precision": 0.32143181611701643,
+      "eval_recall": 0.8277959756668226,
+      "eval_runtime": 16.1891,
+      "eval_samples_per_second": 420.467,
+      "eval_steps_per_second": 52.566,
+      "step": 11910
+    },
+    {
+      "epoch": 7.052600646488393,
+      "grad_norm": 0.6558970212936401,
+      "learning_rate": 1.472663139329806e-05,
+      "loss": 0.0289,
+      "step": 12000
+    },
+    {
+      "epoch": 7.346459006758742,
+      "grad_norm": 0.26189878582954407,
+      "learning_rate": 1.3256907701352148e-05,
+      "loss": 0.0224,
+      "step": 12500
+    },
+    {
+      "epoch": 7.640317367029092,
+      "grad_norm": 1.370686650276184,
+      "learning_rate": 1.1787184009406232e-05,
+      "loss": 0.0231,
+      "step": 13000
+    },
+    {
+      "epoch": 7.934175727299442,
+      "grad_norm": 0.36619672179222107,
+      "learning_rate": 1.0317460317460318e-05,
+      "loss": 0.022,
+      "step": 13500
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.851041187496138,
+      "eval_f1": 0.45333161323614785,
+      "eval_loss": 1.045249104499817,
+      "eval_precision": 0.3129397096802921,
+      "eval_recall": 0.8221806270472625,
+      "eval_runtime": 16.479,
+      "eval_samples_per_second": 413.071,
+      "eval_steps_per_second": 51.641,
+      "step": 13612
+    },
+    {
+      "epoch": 8.22803408756979,
+      "grad_norm": 1.0557399988174438,
+      "learning_rate": 8.847736625514404e-06,
+      "loss": 0.0164,
+      "step": 14000
+    },
+    {
+      "epoch": 8.521892447840141,
+      "grad_norm": 4.32920503616333,
+      "learning_rate": 7.37801293356849e-06,
+      "loss": 0.0164,
+      "step": 14500
+    },
+    {
+      "epoch": 8.81575080811049,
+      "grad_norm": 0.1674884408712387,
+      "learning_rate": 5.908289241622575e-06,
+      "loss": 0.017,
+      "step": 15000
+    },
+    {
+      "epoch": 8.999706141639729,
+      "eval_accuracy": 0.8523937357107058,
+      "eval_f1": 0.45305514157973176,
+      "eval_loss": 1.102479100227356,
+      "eval_precision": 0.31328972130119187,
+      "eval_recall": 0.8179691155825924,
+      "eval_runtime": 16.1914,
+      "eval_samples_per_second": 420.409,
+      "eval_steps_per_second": 52.559,
+      "step": 15313
+    },
+    {
+      "epoch": 9.10960916838084,
+      "grad_norm": 3.3734261989593506,
+      "learning_rate": 4.438565549676661e-06,
+      "loss": 0.0171,
+      "step": 15500
+    },
+    {
+      "epoch": 9.40346752865119,
+      "grad_norm": 0.19828377664089203,
+      "learning_rate": 2.9688418577307467e-06,
+      "loss": 0.013,
+      "step": 16000
+    },
+    {
+      "epoch": 9.69732588892154,
+      "grad_norm": 1.491190791130066,
+      "learning_rate": 1.4991181657848325e-06,
+      "loss": 0.0133,
+      "step": 16500
+    },
+    {
+      "epoch": 9.991184249191889,
+      "grad_norm": 2.5641109943389893,
+      "learning_rate": 2.9394473838918286e-08,
+      "loss": 0.0135,
+      "step": 17000
+    },
+    {
+      "epoch": 9.997061416397296,
+      "eval_accuracy": 0.8526409018818958,
+      "eval_f1": 0.45504563402161957,
+      "eval_loss": 1.1188451051712036,
+      "eval_precision": 0.3145413870246085,
+      "eval_recall": 0.8224145999064109,
+      "eval_runtime": 16.3165,
+      "eval_samples_per_second": 417.185,
+      "eval_steps_per_second": 52.156,
+      "step": 17010
+    },
+    {
+      "epoch": 9.997061416397296,
+      "step": 17010,
+      "total_flos": 6700722040732752.0,
+      "train_loss": 0.08913132946046923,
+      "train_runtime": 3311.208,
+      "train_samples_per_second": 82.218,
+      "train_steps_per_second": 5.137
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 17010,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 6700722040732752.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}