Model save

README.md

@@ -21,8 +21,8 @@ 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: 0.5007
-- Accuracy: 0.9816
+- Loss: 0.6434
+- Accuracy: 0.9633
 - F1: 0.0
 - Precision: 0.0
 - Recall: 0.0
@@ -58,17 +58,17 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Accuracy | F1  | Precision | Recall |
-|:-------------:|:------:|:----:|:---------------:|:--------:|:---:|:---------:|:------:|
-| 0.6651        | 0.2857 | 1    | 0.6412          | 0.9003   | 0.0 | 0.0       | 0.0    |
-| 0.6651        | 0.5714 | 2    | 0.6363          | 0.9239   | 0.0 | 0.0       | 0.0    |
-| 0.6651        | 0.8571 | 3    | 0.6264          | 0.9475   | 0.0 | 0.0       | 0.0    |
-| 0.6651        | 1.1429 | 4    | 0.6125          | 0.9606   | 0.0 | 0.0       | 0.0    |
-| 0.6651        | 1.4286 | 5    | 0.5944          | 0.9685   | 0.0 | 0.0       | 0.0    |
-| 0.6651        | 1.7143 | 6    | 0.5722          | 0.9738   | 0.0 | 0.0       | 0.0    |
-| 0.6651        | 2.0    | 7    | 0.5467          | 0.9790   | 0.0 | 0.0       | 0.0    |
-| 0.6651        | 2.2857 | 8    | 0.5230          | 0.9816   | 0.0 | 0.0       | 0.0    |
-| 0.6651        | 2.5714 | 9    | 0.5007          | 0.9816   | 0.0 | 0.0       | 0.0    |
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
+| No log        | 0.2857 | 1    | 0.8097          | 0.0236   | 0.0363 | 0.0185    | 1.0    |
+| No log        | 0.5714 | 2    | 0.8041          | 0.0236   | 0.0363 | 0.0185    | 1.0    |
+| No log        | 0.8571 | 3    | 0.7931          | 0.0236   | 0.0363 | 0.0185    | 1.0    |
+| No log        | 1.1429 | 4    | 0.7771          | 0.0315   | 0.0366 | 0.0186    | 1.0    |
+| No log        | 1.4286 | 5    | 0.7562          | 0.0499   | 0.0372 | 0.0190    | 1.0    |
+| No log        | 1.7143 | 6    | 0.7305          | 0.1417   | 0.0411 | 0.0210    | 1.0    |
+| No log        | 2.0    | 7    | 0.7005          | 0.3937   | 0.0494 | 0.0254    | 0.8571 |
+| No log        | 2.2857 | 8    | 0.6714          | 0.7717   | 0.0225 | 0.0122    | 0.1429 |
+| No log        | 2.5714 | 9    | 0.6434          | 0.9633   | 0.0    | 0.0       | 0.0    |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82ee5e3e44d3ad9965610c84000143c847d437724ab0e7c8d6efda1e617fd245
+oid sha256:cf161c14c36d4a8c003c0f05b3deac39339b663c124fe1070d37a7a84fec8f1a
 size 46634832

training_artifacts/training_history.csv

@@ -1,11 +1,11 @@
 eval_loss;eval_accuracy;eval_f1;eval_precision;eval_recall;eval_runtime;eval_samples_per_second;eval_steps_per_second;epoch;step;train_runtime;train_samples_per_second;train_steps_per_second;total_flos;train_loss
-0.6739318370819092;0.8110236220472441;0.02702702702702703;0.014925373134328358;0.14285714285714285;5.4658;69.706;4.391;0.2857142857142857;1;;;;;
-0.6692991852760315;0.8372703412073491;0.0;0.0;0.0;5.4265;70.21;4.423;0.5714285714285714;2;;;;;
-0.659990131855011;0.9081364829396326;0.0;0.0;0.0;5.4075;70.457;4.438;0.8571428571428571;3;;;;;
-0.646578311920166;0.9658792650918635;0.0;0.0;0.0;5.484;69.475;4.376;1.1428571428571428;4;;;;;
-0.6292694211006165;0.979002624671916;0.0;0.0;0.0;5.5523;68.62;4.323;1.4285714285714286;5;;;;;
-0.6079225540161133;0.9816272965879265;0.0;0.0;0.0;5.4019;70.531;4.443;1.7142857142857144;6;;;;;
-0.5826576352119446;0.9816272965879265;0.0;0.0;0.0;5.3839;70.767;4.458;2.0;7;;;;;
-0.5580222606658936;0.9816272965879265;0.0;0.0;0.0;5.6743;67.145;4.23;2.2857142857142856;8;;;;;
-0.5341243743896484;0.9816272965879265;0.0;0.0;0.0;5.4883;69.421;4.373;2.571428571428571;9;;;;;
-;;;;;;;;2.571428571428571;9;93.9944;45.109;0.638;2164459611168.0;0.6654236051771376
+0.8096907734870911;0.023622047244094488;0.036269430051813475;0.018469656992084433;1.0;5.4392;70.048;4.412;0.2857142857142857;1;;;;;
+0.8040798902511597;0.023622047244094488;0.036269430051813475;0.018469656992084433;1.0;5.3267;71.526;4.506;0.5714285714285714;2;;;;;
+0.7930730581283569;0.023622047244094488;0.036269430051813475;0.018469656992084433;1.0;5.5906;68.15;4.293;0.8571428571428571;3;;;;;
+0.7771380543708801;0.031496062992125984;0.03655352480417755;0.018617021276595744;1.0;5.3804;70.812;4.461;1.1428571428571428;4;;;;;
+0.7562002539634705;0.049868766404199474;0.03723404255319149;0.018970189701897018;1.0;5.2972;71.924;4.531;1.4285714285714286;5;;;;;
+0.7305469512939453;0.14173228346456693;0.04105571847507331;0.020958083832335328;1.0;5.459;69.793;4.396;1.7142857142857144;6;;;;;
+0.7005411982536316;0.3937007874015748;0.04938271604938272;0.025423728813559324;0.8571428571428571;5.332;71.455;4.501;2.0;7;;;;;
+0.6713957786560059;0.7716535433070866;0.02247191011235955;0.012195121951219513;0.14285714285714285;5.3463;71.264;4.489;2.2857142857142856;8;;;;;
+0.6433929800987244;0.963254593175853;0.0;0.0;0.0;5.451;69.896;4.403;2.571428571428571;9;;;;;
+;;;;;;;;2.571428571428571;9;94.4266;44.903;0.635;2148246430560.0;0.7422826555040147

training_artifacts/training_history.json

@@ -1,118 +1,118 @@
 [
   {
-    "eval_loss": 0.6739318370819092,
-    "eval_accuracy": 0.8110236220472441,
-    "eval_f1": 0.02702702702702703,
-    "eval_precision": 0.014925373134328358,
-    "eval_recall": 0.14285714285714285,
-    "eval_runtime": 5.4658,
-    "eval_samples_per_second": 69.706,
-    "eval_steps_per_second": 4.391,
+    "eval_loss": 0.8096907734870911,
+    "eval_accuracy": 0.023622047244094488,
+    "eval_f1": 0.036269430051813475,
+    "eval_precision": 0.018469656992084433,
+    "eval_recall": 1.0,
+    "eval_runtime": 5.4392,
+    "eval_samples_per_second": 70.048,
+    "eval_steps_per_second": 4.412,
     "epoch": 0.2857142857142857,
     "step": 1
   },
   {
-    "eval_loss": 0.6692991852760315,
-    "eval_accuracy": 0.8372703412073491,
-    "eval_f1": 0.0,
-    "eval_precision": 0.0,
-    "eval_recall": 0.0,
-    "eval_runtime": 5.4265,
-    "eval_samples_per_second": 70.21,
-    "eval_steps_per_second": 4.423,
+    "eval_loss": 0.8040798902511597,
+    "eval_accuracy": 0.023622047244094488,
+    "eval_f1": 0.036269430051813475,
+    "eval_precision": 0.018469656992084433,
+    "eval_recall": 1.0,
+    "eval_runtime": 5.3267,
+    "eval_samples_per_second": 71.526,
+    "eval_steps_per_second": 4.506,
     "epoch": 0.5714285714285714,
     "step": 2
   },
   {
-    "eval_loss": 0.659990131855011,
-    "eval_accuracy": 0.9081364829396326,
-    "eval_f1": 0.0,
-    "eval_precision": 0.0,
-    "eval_recall": 0.0,
-    "eval_runtime": 5.4075,
-    "eval_samples_per_second": 70.457,
-    "eval_steps_per_second": 4.438,
+    "eval_loss": 0.7930730581283569,
+    "eval_accuracy": 0.023622047244094488,
+    "eval_f1": 0.036269430051813475,
+    "eval_precision": 0.018469656992084433,
+    "eval_recall": 1.0,
+    "eval_runtime": 5.5906,
+    "eval_samples_per_second": 68.15,
+    "eval_steps_per_second": 4.293,
     "epoch": 0.8571428571428571,
     "step": 3
   },
   {
-    "eval_loss": 0.646578311920166,
-    "eval_accuracy": 0.9658792650918635,
-    "eval_f1": 0.0,
-    "eval_precision": 0.0,
-    "eval_recall": 0.0,
-    "eval_runtime": 5.484,
-    "eval_samples_per_second": 69.475,
-    "eval_steps_per_second": 4.376,
+    "eval_loss": 0.7771380543708801,
+    "eval_accuracy": 0.031496062992125984,
+    "eval_f1": 0.03655352480417755,
+    "eval_precision": 0.018617021276595744,
+    "eval_recall": 1.0,
+    "eval_runtime": 5.3804,
+    "eval_samples_per_second": 70.812,
+    "eval_steps_per_second": 4.461,
     "epoch": 1.1428571428571428,
     "step": 4
   },
   {
-    "eval_loss": 0.6292694211006165,
-    "eval_accuracy": 0.979002624671916,
-    "eval_f1": 0.0,
-    "eval_precision": 0.0,
-    "eval_recall": 0.0,
-    "eval_runtime": 5.5523,
-    "eval_samples_per_second": 68.62,
-    "eval_steps_per_second": 4.323,
+    "eval_loss": 0.7562002539634705,
+    "eval_accuracy": 0.049868766404199474,
+    "eval_f1": 0.03723404255319149,
+    "eval_precision": 0.018970189701897018,
+    "eval_recall": 1.0,
+    "eval_runtime": 5.2972,
+    "eval_samples_per_second": 71.924,
+    "eval_steps_per_second": 4.531,
     "epoch": 1.4285714285714286,
     "step": 5
   },
   {
-    "eval_loss": 0.6079225540161133,
-    "eval_accuracy": 0.9816272965879265,
-    "eval_f1": 0.0,
-    "eval_precision": 0.0,
-    "eval_recall": 0.0,
-    "eval_runtime": 5.4019,
-    "eval_samples_per_second": 70.531,
-    "eval_steps_per_second": 4.443,
+    "eval_loss": 0.7305469512939453,
+    "eval_accuracy": 0.14173228346456693,
+    "eval_f1": 0.04105571847507331,
+    "eval_precision": 0.020958083832335328,
+    "eval_recall": 1.0,
+    "eval_runtime": 5.459,
+    "eval_samples_per_second": 69.793,
+    "eval_steps_per_second": 4.396,
     "epoch": 1.7142857142857144,
     "step": 6
   },
   {
-    "eval_loss": 0.5826576352119446,
-    "eval_accuracy": 0.9816272965879265,
-    "eval_f1": 0.0,
-    "eval_precision": 0.0,
-    "eval_recall": 0.0,
-    "eval_runtime": 5.3839,
-    "eval_samples_per_second": 70.767,
-    "eval_steps_per_second": 4.458,
+    "eval_loss": 0.7005411982536316,
+    "eval_accuracy": 0.3937007874015748,
+    "eval_f1": 0.04938271604938272,
+    "eval_precision": 0.025423728813559324,
+    "eval_recall": 0.8571428571428571,
+    "eval_runtime": 5.332,
+    "eval_samples_per_second": 71.455,
+    "eval_steps_per_second": 4.501,
     "epoch": 2.0,
     "step": 7
   },
   {
-    "eval_loss": 0.5580222606658936,
-    "eval_accuracy": 0.9816272965879265,
-    "eval_f1": 0.0,
-    "eval_precision": 0.0,
-    "eval_recall": 0.0,
-    "eval_runtime": 5.6743,
-    "eval_samples_per_second": 67.145,
-    "eval_steps_per_second": 4.23,
+    "eval_loss": 0.6713957786560059,
+    "eval_accuracy": 0.7716535433070866,
+    "eval_f1": 0.02247191011235955,
+    "eval_precision": 0.012195121951219513,
+    "eval_recall": 0.14285714285714285,
+    "eval_runtime": 5.3463,
+    "eval_samples_per_second": 71.264,
+    "eval_steps_per_second": 4.489,
     "epoch": 2.2857142857142856,
     "step": 8
   },
   {
-    "eval_loss": 0.5341243743896484,
-    "eval_accuracy": 0.9816272965879265,
+    "eval_loss": 0.6433929800987244,
+    "eval_accuracy": 0.963254593175853,
     "eval_f1": 0.0,
     "eval_precision": 0.0,
     "eval_recall": 0.0,
-    "eval_runtime": 5.4883,
-    "eval_samples_per_second": 69.421,
-    "eval_steps_per_second": 4.373,
+    "eval_runtime": 5.451,
+    "eval_samples_per_second": 69.896,
+    "eval_steps_per_second": 4.403,
     "epoch": 2.571428571428571,
     "step": 9
   },
   {
-    "train_runtime": 93.9944,
-    "train_samples_per_second": 45.109,
-    "train_steps_per_second": 0.638,
-    "total_flos": 2164459611168.0,
-    "train_loss": 0.6654236051771376,
+    "train_runtime": 94.4266,
+    "train_samples_per_second": 44.903,
+    "train_steps_per_second": 0.635,
+    "total_flos": 2148246430560.0,
+    "train_loss": 0.7422826555040147,
     "epoch": 2.571428571428571,
     "step": 9
   }

training_artifacts/training_summary.json

@@ -1,7 +1,7 @@
 {
   "total_epochs": 2.571428571428571,
   "total_steps": "9",
-  "final_train_loss": 0.6651,
-  "final_eval_loss": 0.5007050633430481,
-  "best_eval_loss": 0.5007050633430481
+  "final_train_loss": null,
+  "final_eval_loss": 0.6433929800987244,
+  "best_eval_loss": 0.6433929800987244
 }