Model save

README.md

@@ -21,7 +21,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [DedalusHealthCare/tinybert-mlm-en](https://huggingface.co/DedalusHealthCare/tinybert-mlm-en) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5776
+- Loss: 0.6019
 - Accuracy: 0.9816
 - F1: 0.0
 - Precision: 0.0
@@ -60,15 +60,15 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
 |:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
-| 0.7155        | 0.2857 | 1    | 0.7256          | 0.1969   | 0.0437 | 0.0224    | 1.0    |
-| 0.7155        | 0.5714 | 2    | 0.7204          | 0.2283   | 0.0455 | 0.0233    | 1.0    |
-| 0.7155        | 0.8571 | 3    | 0.7102          | 0.3333   | 0.0451 | 0.0232    | 0.8571 |
-| 0.7155        | 1.1429 | 4    | 0.6954          | 0.4803   | 0.0388 | 0.0201    | 0.5714 |
-| 0.7155        | 1.4286 | 5    | 0.6763          | 0.7008   | 0.0500 | 0.0265    | 0.4286 |
-| 0.7155        | 1.7143 | 6    | 0.6533          | 0.8530   | 0.0345 | 0.0196    | 0.1429 |
-| 0.7155        | 2.0    | 7    | 0.6268          | 0.9580   | 0.0    | 0.0       | 0.0    |
-| 0.7155        | 2.2857 | 8    | 0.6016          | 0.9816   | 0.0    | 0.0       | 0.0    |
-| 0.7155        | 2.5714 | 9    | 0.5776          | 0.9816   | 0.0    | 0.0       | 0.0    |
+| 0.7327        | 0.2857 | 1    | 0.7513          | 0.0341   | 0.0366 | 0.0187    | 1.0    |
+| 0.7327        | 0.5714 | 2    | 0.7464          | 0.0525   | 0.0373 | 0.0190    | 1.0    |
+| 0.7327        | 0.8571 | 3    | 0.7368          | 0.0892   | 0.0388 | 0.0198    | 1.0    |
+| 0.7327        | 1.1429 | 4    | 0.7225          | 0.1916   | 0.0375 | 0.0192    | 0.8571 |
+| 0.7327        | 1.4286 | 5    | 0.7037          | 0.3885   | 0.0251 | 0.0129    | 0.4286 |
+| 0.7327        | 1.7143 | 6    | 0.6806          | 0.6325   | 0.0    | 0.0       | 0.0    |
+| 0.7327        | 2.0    | 7    | 0.6538          | 0.8688   | 0.0    | 0.0       | 0.0    |
+| 0.7327        | 2.2857 | 8    | 0.6274          | 0.9738   | 0.0    | 0.0       | 0.0    |
+| 0.7327        | 2.5714 | 9    | 0.6019          | 0.9816   | 0.0    | 0.0       | 0.0    |
 
 
 ### Framework versions

logs/events.out.tfevents.1760000689.ip-172-31-12-22.60300.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:182fcdd48e40b74e85680105f7799127104601f667992658d7c8a1a39d446766
-size 9102
+oid sha256:0771b82a0b1c920e332fdcef27121fdda29a391886f7203c786bce77fc69b851
+size 9913

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73a4e01c323b04a5fbeb6f60d411268d744db3cf2916655f4ff3675db95f491f
+oid sha256:8aacff2c15f2be28bc6a28dbe6b917ce1a64cb9b4ea7b535bffebc91aed3c342
 size 46634832

training_artifacts/training_history.csv

@@ -1,12 +1,12 @@
 loss;grad_norm;learning_rate;epoch;step;eval_loss;eval_accuracy;eval_f1;eval_precision;eval_recall;eval_runtime;eval_samples_per_second;eval_steps_per_second;train_runtime;train_samples_per_second;train_steps_per_second;total_flos;train_loss
-0.7155;3.170656681060791;3.3333333333333333e-06;0.2857142857142857;1;;;;;;;;;;;;;
-;;;0.2857142857142857;1;0.7255510687828064;0.1968503937007874;0.04375;0.022364217252396165;1.0;7.3525;51.819;3.264;;;;;
-;;;0.5714285714285714;2;0.7203730940818787;0.2283464566929134;0.04545454545454545;0.023255813953488372;1.0;7.5704;50.328;3.17;;;;;
-;;;0.8571428571428571;3;0.7101789712905884;0.3333333333333333;0.04511278195488722;0.023166023166023165;0.8571428571428571;7.7608;49.093;3.092;;;;;
-;;;1.1428571428571428;4;0.6954057216644287;0.48031496062992124;0.038834951456310676;0.020100502512562814;0.5714285714285714;7.798;48.859;3.078;;;;;
-;;;1.4285714285714286;5;0.6762979626655579;0.7007874015748031;0.049999999999999996;0.02654867256637168;0.42857142857142855;8.2567;46.144;2.907;;;;;
-;;;1.7142857142857144;6;0.6532518267631531;0.8530183727034121;0.034482758620689655;0.0196078431372549;0.14285714285714285;7.127;53.459;3.367;;;;;
-;;;2.0;7;0.6268358826637268;0.958005249343832;0.0;0.0;0.0;7.3211;52.041;3.278;;;;;
-;;;2.2857142857142856;8;0.6016007661819458;0.9816272965879265;0.0;0.0;0.0;8.4314;45.188;2.847;;;;;
-;;;2.571428571428571;9;0.5776201486587524;0.9816272965879265;0.0;0.0;0.0;7.5633;50.375;3.173;;;;;
-;;;2.571428571428571;9;;;;;;;;;133.4632;31.769;0.45;2156353020864.0;0.68766188621521
+0.7327;2.7726080417633057;3.3333333333333333e-06;0.2857142857142857;1;;;;;;;;;;;;;
+;;;0.2857142857142857;1;0.7512667775154114;0.03412073490813648;0.03664921465968587;0.018666666666666668;1.0;5.3956;70.613;4.448;;;;;
+;;;0.5714285714285714;2;0.7464109659194946;0.05249343832020997;0.037333333333333336;0.019021739130434784;1.0;5.5933;68.117;4.291;;;;;
+;;;0.8571428571428571;3;0.7368069291114807;0.08923884514435695;0.038781163434903045;0.01977401129943503;1.0;5.5621;68.5;4.315;;;;;
+;;;1.1428571428571428;4;0.7224730849266052;0.19160104986876642;0.03749999999999999;0.019169329073482427;0.8571428571428571;5.6279;67.698;4.264;;;;;
+;;;1.4285714285714286;5;0.7036912441253662;0.3884514435695538;0.02510460251046025;0.01293103448275862;0.42857142857142855;5.703;66.807;4.208;;;;;
+;;;1.7142857142857144;6;0.6806334853172302;0.6325459317585301;0.0;0.0;0.0;5.773;65.997;4.157;;;;;
+;;;2.0;7;0.6538448929786682;0.868766404199475;0.0;0.0;0.0;5.6817;67.057;4.224;;;;;
+;;;2.2857142857142856;8;0.6274169683456421;0.973753280839895;0.0;0.0;0.0;5.8624;64.991;4.094;;;;;
+;;;2.571428571428571;9;0.6018883585929871;0.9816272965879265;0.0;0.0;0.0;5.644;67.505;4.252;;;;;
+;;;2.571428571428571;9;;;;;;;;;101.6832;41.698;0.59;2156353020864.0;0.6984957986407809

training_artifacts/training_history.json

@@ -1,125 +1,125 @@
 [
   {
-    "loss": 0.7155,
-    "grad_norm": 3.170656681060791,
+    "loss": 0.7327,
+    "grad_norm": 2.7726080417633057,
     "learning_rate": 3.3333333333333333e-06,
     "epoch": 0.2857142857142857,
     "step": 1
   },
   {
-    "eval_loss": 0.7255510687828064,
-    "eval_accuracy": 0.1968503937007874,
-    "eval_f1": 0.04375,
-    "eval_precision": 0.022364217252396165,
+    "eval_loss": 0.7512667775154114,
+    "eval_accuracy": 0.03412073490813648,
+    "eval_f1": 0.03664921465968587,
+    "eval_precision": 0.018666666666666668,
     "eval_recall": 1.0,
-    "eval_runtime": 7.3525,
-    "eval_samples_per_second": 51.819,
-    "eval_steps_per_second": 3.264,
+    "eval_runtime": 5.3956,
+    "eval_samples_per_second": 70.613,
+    "eval_steps_per_second": 4.448,
     "epoch": 0.2857142857142857,
     "step": 1
   },
   {
-    "eval_loss": 0.7203730940818787,
-    "eval_accuracy": 0.2283464566929134,
-    "eval_f1": 0.04545454545454545,
-    "eval_precision": 0.023255813953488372,
+    "eval_loss": 0.7464109659194946,
+    "eval_accuracy": 0.05249343832020997,
+    "eval_f1": 0.037333333333333336,
+    "eval_precision": 0.019021739130434784,
     "eval_recall": 1.0,
-    "eval_runtime": 7.5704,
-    "eval_samples_per_second": 50.328,
-    "eval_steps_per_second": 3.17,
+    "eval_runtime": 5.5933,
+    "eval_samples_per_second": 68.117,
+    "eval_steps_per_second": 4.291,
     "epoch": 0.5714285714285714,
     "step": 2
   },
   {
-    "eval_loss": 0.7101789712905884,
-    "eval_accuracy": 0.3333333333333333,
-    "eval_f1": 0.04511278195488722,
-    "eval_precision": 0.023166023166023165,
-    "eval_recall": 0.8571428571428571,
-    "eval_runtime": 7.7608,
-    "eval_samples_per_second": 49.093,
-    "eval_steps_per_second": 3.092,
+    "eval_loss": 0.7368069291114807,
+    "eval_accuracy": 0.08923884514435695,
+    "eval_f1": 0.038781163434903045,
+    "eval_precision": 0.01977401129943503,
+    "eval_recall": 1.0,
+    "eval_runtime": 5.5621,
+    "eval_samples_per_second": 68.5,
+    "eval_steps_per_second": 4.315,
     "epoch": 0.8571428571428571,
     "step": 3
   },
   {
-    "eval_loss": 0.6954057216644287,
-    "eval_accuracy": 0.48031496062992124,
-    "eval_f1": 0.038834951456310676,
-    "eval_precision": 0.020100502512562814,
-    "eval_recall": 0.5714285714285714,
-    "eval_runtime": 7.798,
-    "eval_samples_per_second": 48.859,
-    "eval_steps_per_second": 3.078,
+    "eval_loss": 0.7224730849266052,
+    "eval_accuracy": 0.19160104986876642,
+    "eval_f1": 0.03749999999999999,
+    "eval_precision": 0.019169329073482427,
+    "eval_recall": 0.8571428571428571,
+    "eval_runtime": 5.6279,
+    "eval_samples_per_second": 67.698,
+    "eval_steps_per_second": 4.264,
     "epoch": 1.1428571428571428,
     "step": 4
   },
   {
-    "eval_loss": 0.6762979626655579,
-    "eval_accuracy": 0.7007874015748031,
-    "eval_f1": 0.049999999999999996,
-    "eval_precision": 0.02654867256637168,
+    "eval_loss": 0.7036912441253662,
+    "eval_accuracy": 0.3884514435695538,
+    "eval_f1": 0.02510460251046025,
+    "eval_precision": 0.01293103448275862,
     "eval_recall": 0.42857142857142855,
-    "eval_runtime": 8.2567,
-    "eval_samples_per_second": 46.144,
-    "eval_steps_per_second": 2.907,
+    "eval_runtime": 5.703,
+    "eval_samples_per_second": 66.807,
+    "eval_steps_per_second": 4.208,
     "epoch": 1.4285714285714286,
     "step": 5
   },
   {
-    "eval_loss": 0.6532518267631531,
-    "eval_accuracy": 0.8530183727034121,
-    "eval_f1": 0.034482758620689655,
-    "eval_precision": 0.0196078431372549,
-    "eval_recall": 0.14285714285714285,
-    "eval_runtime": 7.127,
-    "eval_samples_per_second": 53.459,
-    "eval_steps_per_second": 3.367,
+    "eval_loss": 0.6806334853172302,
+    "eval_accuracy": 0.6325459317585301,
+    "eval_f1": 0.0,
+    "eval_precision": 0.0,
+    "eval_recall": 0.0,
+    "eval_runtime": 5.773,
+    "eval_samples_per_second": 65.997,
+    "eval_steps_per_second": 4.157,
     "epoch": 1.7142857142857144,
     "step": 6
   },
   {
-    "eval_loss": 0.6268358826637268,
-    "eval_accuracy": 0.958005249343832,
+    "eval_loss": 0.6538448929786682,
+    "eval_accuracy": 0.868766404199475,
     "eval_f1": 0.0,
     "eval_precision": 0.0,
     "eval_recall": 0.0,
-    "eval_runtime": 7.3211,
-    "eval_samples_per_second": 52.041,
-    "eval_steps_per_second": 3.278,
+    "eval_runtime": 5.6817,
+    "eval_samples_per_second": 67.057,
+    "eval_steps_per_second": 4.224,
     "epoch": 2.0,
     "step": 7
   },
   {
-    "eval_loss": 0.6016007661819458,
-    "eval_accuracy": 0.9816272965879265,
+    "eval_loss": 0.6274169683456421,
+    "eval_accuracy": 0.973753280839895,
     "eval_f1": 0.0,
     "eval_precision": 0.0,
     "eval_recall": 0.0,
-    "eval_runtime": 8.4314,
-    "eval_samples_per_second": 45.188,
-    "eval_steps_per_second": 2.847,
+    "eval_runtime": 5.8624,
+    "eval_samples_per_second": 64.991,
+    "eval_steps_per_second": 4.094,
     "epoch": 2.2857142857142856,
     "step": 8
   },
   {
-    "eval_loss": 0.5776201486587524,
+    "eval_loss": 0.6018883585929871,
     "eval_accuracy": 0.9816272965879265,
     "eval_f1": 0.0,
     "eval_precision": 0.0,
     "eval_recall": 0.0,
-    "eval_runtime": 7.5633,
-    "eval_samples_per_second": 50.375,
-    "eval_steps_per_second": 3.173,
+    "eval_runtime": 5.644,
+    "eval_samples_per_second": 67.505,
+    "eval_steps_per_second": 4.252,
     "epoch": 2.571428571428571,
     "step": 9
   },
   {
-    "train_runtime": 133.4632,
-    "train_samples_per_second": 31.769,
-    "train_steps_per_second": 0.45,
+    "train_runtime": 101.6832,
+    "train_samples_per_second": 41.698,
+    "train_steps_per_second": 0.59,
     "total_flos": 2156353020864.0,
-    "train_loss": 0.68766188621521,
+    "train_loss": 0.6984957986407809,
     "epoch": 2.571428571428571,
     "step": 9
   }

training_artifacts/training_summary.json

@@ -1,7 +1,7 @@
 {
   "total_epochs": 2.571428571428571,
   "total_steps": "9",
-  "final_train_loss": 0.7155,
-  "final_eval_loss": 0.5776201486587524,
-  "best_eval_loss": 0.5776201486587524
+  "final_train_loss": 0.7327,
+  "final_eval_loss": 0.6018883585929871,
+  "best_eval_loss": 0.6018883585929871
 }