End of training

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+eval_nbest_predictions.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -4,6 +4,8 @@ license: mit
 base_model: emilyalsentzer/Bio_ClinicalBERT
 tags:
 - generated_from_trainer
+datasets:
+- squad_v2
 model-index:
 - name: Bio_ClinicalBERT_QA
   results: []
@@ -14,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Bio_ClinicalBERT_QA
 
-This model is a fine-tuned version of [emilyalsentzer/Bio_ClinicalBERT](https://huggingface.co/emilyalsentzer/Bio_ClinicalBERT) on an unknown dataset.
+This model is a fine-tuned version of [emilyalsentzer/Bio_ClinicalBERT](https://huggingface.co/emilyalsentzer/Bio_ClinicalBERT) on the squad_v2 dataset.
 
 ## Model description
 

all_results.json

@@ -1,26 +1,26 @@
 {
     "epoch": 2.0,
-    "eval_HasAns_exact": 2.2222222222222223,
-    "eval_HasAns_f1": 8.371200331984646,
-    "eval_HasAns_total": 45,
-    "eval_NoAns_exact": 0.0,
-    "eval_NoAns_f1": 0.0,
-    "eval_NoAns_total": 55,
-    "eval_best_exact": 55.0,
+    "eval_HasAns_exact": 69.87179487179488,
+    "eval_HasAns_f1": 76.91826514384819,
+    "eval_HasAns_total": 5928,
+    "eval_NoAns_exact": 76.16484440706476,
+    "eval_NoAns_f1": 76.16484440706476,
+    "eval_NoAns_total": 5945,
+    "eval_best_exact": 73.02282489682473,
     "eval_best_exact_thresh": 0.0,
-    "eval_best_f1": 55.517647058823535,
+    "eval_best_f1": 76.54101539398066,
     "eval_best_f1_thresh": 0.0,
-    "eval_exact": 1.0,
-    "eval_f1": 3.7670401493930905,
-    "eval_runtime": 2.3757,
-    "eval_samples": 100,
-    "eval_samples_per_second": 42.093,
-    "eval_steps_per_second": 5.472,
-    "eval_total": 100,
-    "total_flos": 51214164320256.0,
-    "train_loss": 4.990250057644314,
-    "train_runtime": 14.83,
-    "train_samples": 100,
-    "train_samples_per_second": 13.486,
-    "train_steps_per_second": 1.214
+    "eval_exact": 73.02282489682473,
+    "eval_f1": 76.54101539398063,
+    "eval_runtime": 310.3268,
+    "eval_samples": 11985,
+    "eval_samples_per_second": 38.621,
+    "eval_steps_per_second": 4.83,
+    "eval_total": 11873,
+    "total_flos": 6.833694078916608e+16,
+    "train_loss": 1.1028300029987763,
+    "train_runtime": 22205.1611,
+    "train_samples": 130765,
+    "train_samples_per_second": 11.778,
+    "train_steps_per_second": 0.982
 }

eval_results.json

@@ -1,20 +1,20 @@
 {
     "epoch": 2.0,
-    "eval_HasAns_exact": 2.2222222222222223,
-    "eval_HasAns_f1": 8.371200331984646,
-    "eval_HasAns_total": 45,
-    "eval_NoAns_exact": 0.0,
-    "eval_NoAns_f1": 0.0,
-    "eval_NoAns_total": 55,
-    "eval_best_exact": 55.0,
+    "eval_HasAns_exact": 69.87179487179488,
+    "eval_HasAns_f1": 76.91826514384819,
+    "eval_HasAns_total": 5928,
+    "eval_NoAns_exact": 76.16484440706476,
+    "eval_NoAns_f1": 76.16484440706476,
+    "eval_NoAns_total": 5945,
+    "eval_best_exact": 73.02282489682473,
     "eval_best_exact_thresh": 0.0,
-    "eval_best_f1": 55.517647058823535,
+    "eval_best_f1": 76.54101539398066,
     "eval_best_f1_thresh": 0.0,
-    "eval_exact": 1.0,
-    "eval_f1": 3.7670401493930905,
-    "eval_runtime": 2.3757,
-    "eval_samples": 100,
-    "eval_samples_per_second": 42.093,
-    "eval_steps_per_second": 5.472,
-    "eval_total": 100
+    "eval_exact": 73.02282489682473,
+    "eval_f1": 76.54101539398063,
+    "eval_runtime": 310.3268,
+    "eval_samples": 11985,
+    "eval_samples_per_second": 38.621,
+    "eval_steps_per_second": 4.83,
+    "eval_total": 11873
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 2.0,
-    "total_flos": 51214164320256.0,
-    "train_loss": 4.990250057644314,
-    "train_runtime": 14.83,
-    "train_samples": 100,
-    "train_samples_per_second": 13.486,
-    "train_steps_per_second": 1.214
+    "total_flos": 6.833694078916608e+16,
+    "train_loss": 1.1028300029987763,
+    "train_runtime": 22205.1611,
+    "train_samples": 130765,
+    "train_samples_per_second": 11.778,
+    "train_steps_per_second": 0.982
 }

trainer_state.json

@@ -3,26 +3,327 @@
   "best_model_checkpoint": null,
   "epoch": 2.0,
   "eval_steps": 500,
-  "global_step": 18,
+  "global_step": 21796,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 0.04587997797761057,
+      "grad_norm": 14.238367080688477,
+      "learning_rate": 2.9311800330335842e-05,
+      "loss": 2.6043,
+      "step": 500
+    },
+    {
+      "epoch": 0.09175995595522114,
+      "grad_norm": 14.928800582885742,
+      "learning_rate": 2.8623600660671684e-05,
+      "loss": 1.9062,
+      "step": 1000
+    },
+    {
+      "epoch": 0.13763993393283172,
+      "grad_norm": 20.992238998413086,
+      "learning_rate": 2.7935400991007525e-05,
+      "loss": 1.7256,
+      "step": 1500
+    },
+    {
+      "epoch": 0.18351991191044228,
+      "grad_norm": 17.107088088989258,
+      "learning_rate": 2.7247201321343367e-05,
+      "loss": 1.5579,
+      "step": 2000
+    },
+    {
+      "epoch": 0.22939988988805285,
+      "grad_norm": 17.52288246154785,
+      "learning_rate": 2.6559001651679208e-05,
+      "loss": 1.5227,
+      "step": 2500
+    },
+    {
+      "epoch": 0.27527986786566344,
+      "grad_norm": 15.443458557128906,
+      "learning_rate": 2.587080198201505e-05,
+      "loss": 1.4445,
+      "step": 3000
+    },
+    {
+      "epoch": 0.321159845843274,
+      "grad_norm": 15.13930892944336,
+      "learning_rate": 2.518260231235089e-05,
+      "loss": 1.4353,
+      "step": 3500
+    },
+    {
+      "epoch": 0.36703982382088457,
+      "grad_norm": 21.271406173706055,
+      "learning_rate": 2.4494402642686732e-05,
+      "loss": 1.3741,
+      "step": 4000
+    },
+    {
+      "epoch": 0.41291980179849513,
+      "grad_norm": 14.242935180664062,
+      "learning_rate": 2.3806202973022574e-05,
+      "loss": 1.3533,
+      "step": 4500
+    },
+    {
+      "epoch": 0.4587997797761057,
+      "grad_norm": 19.635059356689453,
+      "learning_rate": 2.3118003303358415e-05,
+      "loss": 1.3076,
+      "step": 5000
+    },
+    {
+      "epoch": 0.5046797577537163,
+      "grad_norm": 22.73372459411621,
+      "learning_rate": 2.2429803633694257e-05,
+      "loss": 1.2723,
+      "step": 5500
+    },
+    {
+      "epoch": 0.5505597357313269,
+      "grad_norm": 20.632278442382812,
+      "learning_rate": 2.17416039640301e-05,
+      "loss": 1.2534,
+      "step": 6000
+    },
+    {
+      "epoch": 0.5964397137089374,
+      "grad_norm": 24.003353118896484,
+      "learning_rate": 2.105340429436594e-05,
+      "loss": 1.2556,
+      "step": 6500
+    },
+    {
+      "epoch": 0.642319691686548,
+      "grad_norm": 20.14078712463379,
+      "learning_rate": 2.036520462470178e-05,
+      "loss": 1.2472,
+      "step": 7000
+    },
+    {
+      "epoch": 0.6881996696641586,
+      "grad_norm": 11.762835502624512,
+      "learning_rate": 1.9677004955037623e-05,
+      "loss": 1.1952,
+      "step": 7500
+    },
+    {
+      "epoch": 0.7340796476417691,
+      "grad_norm": 19.036245346069336,
+      "learning_rate": 1.8988805285373464e-05,
+      "loss": 1.1887,
+      "step": 8000
+    },
+    {
+      "epoch": 0.7799596256193797,
+      "grad_norm": 23.435789108276367,
+      "learning_rate": 1.8300605615709306e-05,
+      "loss": 1.2012,
+      "step": 8500
+    },
+    {
+      "epoch": 0.8258396035969903,
+      "grad_norm": 8.716887474060059,
+      "learning_rate": 1.7612405946045144e-05,
+      "loss": 1.1651,
+      "step": 9000
+    },
+    {
+      "epoch": 0.8717195815746008,
+      "grad_norm": 18.400320053100586,
+      "learning_rate": 1.692420627638099e-05,
+      "loss": 1.1494,
+      "step": 9500
+    },
+    {
+      "epoch": 0.9175995595522114,
+      "grad_norm": 13.751968383789062,
+      "learning_rate": 1.6236006606716827e-05,
+      "loss": 1.1467,
+      "step": 10000
+    },
+    {
+      "epoch": 0.963479537529822,
+      "grad_norm": 17.547643661499023,
+      "learning_rate": 1.554780693705267e-05,
+      "loss": 1.1369,
+      "step": 10500
+    },
+    {
+      "epoch": 1.0093595155074326,
+      "grad_norm": 10.978132247924805,
+      "learning_rate": 1.4859607267388513e-05,
+      "loss": 1.0624,
+      "step": 11000
+    },
+    {
+      "epoch": 1.055239493485043,
+      "grad_norm": 13.976412773132324,
+      "learning_rate": 1.4171407597724355e-05,
+      "loss": 0.8489,
+      "step": 11500
+    },
+    {
+      "epoch": 1.1011194714626538,
+      "grad_norm": 20.82090950012207,
+      "learning_rate": 1.3483207928060196e-05,
+      "loss": 0.8338,
+      "step": 12000
+    },
+    {
+      "epoch": 1.1469994494402642,
+      "grad_norm": 15.114026069641113,
+      "learning_rate": 1.2795008258396037e-05,
+      "loss": 0.8326,
+      "step": 12500
+    },
+    {
+      "epoch": 1.192879427417875,
+      "grad_norm": 12.581435203552246,
+      "learning_rate": 1.2106808588731877e-05,
+      "loss": 0.8298,
+      "step": 13000
+    },
+    {
+      "epoch": 1.2387594053954853,
+      "grad_norm": 7.8120856285095215,
+      "learning_rate": 1.1418608919067719e-05,
+      "loss": 0.8512,
+      "step": 13500
+    },
+    {
+      "epoch": 1.284639383373096,
+      "grad_norm": 8.480123519897461,
+      "learning_rate": 1.073040924940356e-05,
+      "loss": 0.8186,
+      "step": 14000
+    },
+    {
+      "epoch": 1.3305193613507065,
+      "grad_norm": 23.09768295288086,
+      "learning_rate": 1.0042209579739402e-05,
+      "loss": 0.8084,
+      "step": 14500
+    },
+    {
+      "epoch": 1.3763993393283172,
+      "grad_norm": 13.01577091217041,
+      "learning_rate": 9.354009910075243e-06,
+      "loss": 0.8521,
+      "step": 15000
+    },
+    {
+      "epoch": 1.4222793173059278,
+      "grad_norm": 17.708642959594727,
+      "learning_rate": 8.665810240411085e-06,
+      "loss": 0.8355,
+      "step": 15500
+    },
+    {
+      "epoch": 1.4681592952835383,
+      "grad_norm": 18.191207885742188,
+      "learning_rate": 7.977610570746926e-06,
+      "loss": 0.8216,
+      "step": 16000
+    },
+    {
+      "epoch": 1.5140392732611487,
+      "grad_norm": 22.47287940979004,
+      "learning_rate": 7.2894109010827675e-06,
+      "loss": 0.8332,
+      "step": 16500
+    },
+    {
+      "epoch": 1.5599192512387594,
+      "grad_norm": 17.28014373779297,
+      "learning_rate": 6.601211231418609e-06,
+      "loss": 0.7964,
+      "step": 17000
+    },
+    {
+      "epoch": 1.60579922921637,
+      "grad_norm": 15.099440574645996,
+      "learning_rate": 5.9130115617544504e-06,
+      "loss": 0.8361,
+      "step": 17500
+    },
+    {
+      "epoch": 1.6516792071939805,
+      "grad_norm": 15.106084823608398,
+      "learning_rate": 5.224811892090292e-06,
+      "loss": 0.8145,
+      "step": 18000
+    },
+    {
+      "epoch": 1.697559185171591,
+      "grad_norm": 12.694562911987305,
+      "learning_rate": 4.536612222426133e-06,
+      "loss": 0.784,
+      "step": 18500
+    },
+    {
+      "epoch": 1.7434391631492017,
+      "grad_norm": 17.011945724487305,
+      "learning_rate": 3.848412552761975e-06,
+      "loss": 0.7694,
+      "step": 19000
+    },
+    {
+      "epoch": 1.7893191411268123,
+      "grad_norm": 8.237266540527344,
+      "learning_rate": 3.1602128830978163e-06,
+      "loss": 0.7693,
+      "step": 19500
+    },
+    {
+      "epoch": 1.8351991191044228,
+      "grad_norm": 12.428207397460938,
+      "learning_rate": 2.4720132134336578e-06,
+      "loss": 0.824,
+      "step": 20000
+    },
+    {
+      "epoch": 1.8810790970820332,
+      "grad_norm": 21.18424415588379,
+      "learning_rate": 1.783813543769499e-06,
+      "loss": 0.7838,
+      "step": 20500
+    },
+    {
+      "epoch": 1.926959075059644,
+      "grad_norm": 13.418033599853516,
+      "learning_rate": 1.0956138741053403e-06,
+      "loss": 0.7877,
+      "step": 21000
+    },
+    {
+      "epoch": 1.9728390530372546,
+      "grad_norm": 11.871217727661133,
+      "learning_rate": 4.074142044411819e-07,
+      "loss": 0.7867,
+      "step": 21500
+    },
     {
       "epoch": 2.0,
-      "step": 18,
-      "total_flos": 51214164320256.0,
-      "train_loss": 4.990250057644314,
-      "train_runtime": 14.83,
-      "train_samples_per_second": 13.486,
-      "train_steps_per_second": 1.214
+      "step": 21796,
+      "total_flos": 6.833694078916608e+16,
+      "train_loss": 1.1028300029987763,
+      "train_runtime": 22205.1611,
+      "train_samples_per_second": 11.778,
+      "train_steps_per_second": 0.982
     }
   ],
   "logging_steps": 500,
-  "max_steps": 18,
+  "max_steps": 21796,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 2,
-  "save_steps": 500,
+  "save_steps": 2000,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {
@@ -35,7 +336,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 51214164320256.0,
+  "total_flos": 6.833694078916608e+16,
   "train_batch_size": 12,
   "trial_name": null,
   "trial_params": null