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: 0.6434
-- Accuracy: 0.9633
+- Loss: 0.5185
+- Accuracy: 0.9816
 - 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 |
-|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
-| 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    |
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy | F1  | Precision | Recall |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:---:|:---------:|:------:|
+| No log        | 0.2857 | 1    | 0.6663          | 0.7953   | 0.0 | 0.0       | 0.0    |
+| No log        | 0.5714 | 2    | 0.6612          | 0.8189   | 0.0 | 0.0       | 0.0    |
+| No log        | 0.8571 | 3    | 0.6516          | 0.8766   | 0.0 | 0.0       | 0.0    |
+| No log        | 1.1429 | 4    | 0.6373          | 0.9081   | 0.0 | 0.0       | 0.0    |
+| No log        | 1.4286 | 5    | 0.6185          | 0.9423   | 0.0 | 0.0       | 0.0    |
+| No log        | 1.7143 | 6    | 0.5955          | 0.9685   | 0.0 | 0.0       | 0.0    |
+| No log        | 2.0    | 7    | 0.5687          | 0.9790   | 0.0 | 0.0       | 0.0    |
+| No log        | 2.2857 | 8    | 0.5431          | 0.9816   | 0.0 | 0.0       | 0.0    |
+| No log        | 2.5714 | 9    | 0.5185          | 0.9816   | 0.0 | 0.0       | 0.0    |
 
 
 ### Framework versions

logs/events.out.tfevents.1762172811.ip-172-31-12-22.8994.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd55062dc02f281f269674d23c49a0afceb3d5b1f648b3003437aa56ea66dd87
-size 8897
+oid sha256:e9aec6d7b2885fbc0dde387cddb0fc4fcac1d3db3486dfab1741fcc0adb5ba79
+size 9708

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96535e266d440c97be85668eea30fea6360ff8e318348e10d5a42a877a3ca25a
+oid sha256:78bacd979776c3ccc2a8405ba190da41af6108747bc5e2b98c9263e6dfd60d23
 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
-0.8096907734870911;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
+0.666271448135376;0.7952755905511811;0.0;0.0;0.0;5.2752;72.225;4.55;0.2857142857142857;1;;;;;
+0.6612232327461243;0.8188976377952756;0.0;0.0;0.0;5.3323;71.452;4.501;0.5714285714285714;2;;;;;
+0.6515527367591858;0.8766404199475065;0.0;0.0;0.0;5.3094;71.76;4.52;0.8571428571428571;3;;;;;
+0.6372586488723755;0.9081364829396326;0.0;0.0;0.0;5.3883;70.709;4.454;1.1428571428571428;4;;;;;
+0.6185054183006287;0.9422572178477691;0.0;0.0;0.0;5.3088;71.767;4.521;1.4285714285714286;5;;;;;
+0.5955398082733154;0.968503937007874;0.0;0.0;0.0;5.3343;71.424;4.499;1.7142857142857144;6;;;;;
+0.568749189376831;0.979002624671916;0.0;0.0;0.0;5.3627;71.047;4.475;2.0;7;;;;;
+0.5430901646614075;0.9816272965879265;0.0;0.0;0.0;5.267;72.337;4.557;2.2857142857142856;8;;;;;
+0.5185076594352722;0.9816272965879265;0.0;0.0;0.0;5.2515;72.55;4.57;2.571428571428571;9;;;;;
+;;;;;;;;2.571428571428571;9;90.9475;46.62;0.66;2164459611168.0;0.6569934950934516

training_artifacts/training_history.json

@@ -1,118 +1,118 @@
 [
   {
-    "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,
+    "eval_loss": 0.666271448135376,
+    "eval_accuracy": 0.7952755905511811,
+    "eval_f1": 0.0,
+    "eval_precision": 0.0,
+    "eval_recall": 0.0,
+    "eval_runtime": 5.2752,
+    "eval_samples_per_second": 72.225,
+    "eval_steps_per_second": 4.55,
     "epoch": 0.2857142857142857,
     "step": 1
   },
   {
-    "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,
+    "eval_loss": 0.6612232327461243,
+    "eval_accuracy": 0.8188976377952756,
+    "eval_f1": 0.0,
+    "eval_precision": 0.0,
+    "eval_recall": 0.0,
+    "eval_runtime": 5.3323,
+    "eval_samples_per_second": 71.452,
+    "eval_steps_per_second": 4.501,
     "epoch": 0.5714285714285714,
     "step": 2
   },
   {
-    "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,
+    "eval_loss": 0.6515527367591858,
+    "eval_accuracy": 0.8766404199475065,
+    "eval_f1": 0.0,
+    "eval_precision": 0.0,
+    "eval_recall": 0.0,
+    "eval_runtime": 5.3094,
+    "eval_samples_per_second": 71.76,
+    "eval_steps_per_second": 4.52,
     "epoch": 0.8571428571428571,
     "step": 3
   },
   {
-    "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,
+    "eval_loss": 0.6372586488723755,
+    "eval_accuracy": 0.9081364829396326,
+    "eval_f1": 0.0,
+    "eval_precision": 0.0,
+    "eval_recall": 0.0,
+    "eval_runtime": 5.3883,
+    "eval_samples_per_second": 70.709,
+    "eval_steps_per_second": 4.454,
     "epoch": 1.1428571428571428,
     "step": 4
   },
   {
-    "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,
+    "eval_loss": 0.6185054183006287,
+    "eval_accuracy": 0.9422572178477691,
+    "eval_f1": 0.0,
+    "eval_precision": 0.0,
+    "eval_recall": 0.0,
+    "eval_runtime": 5.3088,
+    "eval_samples_per_second": 71.767,
+    "eval_steps_per_second": 4.521,
     "epoch": 1.4285714285714286,
     "step": 5
   },
   {
-    "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,
+    "eval_loss": 0.5955398082733154,
+    "eval_accuracy": 0.968503937007874,
+    "eval_f1": 0.0,
+    "eval_precision": 0.0,
+    "eval_recall": 0.0,
+    "eval_runtime": 5.3343,
+    "eval_samples_per_second": 71.424,
+    "eval_steps_per_second": 4.499,
     "epoch": 1.7142857142857144,
     "step": 6
   },
   {
-    "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,
+    "eval_loss": 0.568749189376831,
+    "eval_accuracy": 0.979002624671916,
+    "eval_f1": 0.0,
+    "eval_precision": 0.0,
+    "eval_recall": 0.0,
+    "eval_runtime": 5.3627,
+    "eval_samples_per_second": 71.047,
+    "eval_steps_per_second": 4.475,
     "epoch": 2.0,
     "step": 7
   },
   {
-    "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,
+    "eval_loss": 0.5430901646614075,
+    "eval_accuracy": 0.9816272965879265,
+    "eval_f1": 0.0,
+    "eval_precision": 0.0,
+    "eval_recall": 0.0,
+    "eval_runtime": 5.267,
+    "eval_samples_per_second": 72.337,
+    "eval_steps_per_second": 4.557,
     "epoch": 2.2857142857142856,
     "step": 8
   },
   {
-    "eval_loss": 0.6433929800987244,
-    "eval_accuracy": 0.963254593175853,
+    "eval_loss": 0.5185076594352722,
+    "eval_accuracy": 0.9816272965879265,
     "eval_f1": 0.0,
     "eval_precision": 0.0,
     "eval_recall": 0.0,
-    "eval_runtime": 5.451,
-    "eval_samples_per_second": 69.896,
-    "eval_steps_per_second": 4.403,
+    "eval_runtime": 5.2515,
+    "eval_samples_per_second": 72.55,
+    "eval_steps_per_second": 4.57,
     "epoch": 2.571428571428571,
     "step": 9
   },
   {
-    "train_runtime": 94.4266,
-    "train_samples_per_second": 44.903,
-    "train_steps_per_second": 0.635,
-    "total_flos": 2148246430560.0,
-    "train_loss": 0.7422826555040147,
+    "train_runtime": 90.9475,
+    "train_samples_per_second": 46.62,
+    "train_steps_per_second": 0.66,
+    "total_flos": 2164459611168.0,
+    "train_loss": 0.6569934950934516,
     "epoch": 2.571428571428571,
     "step": 9
   }

training_artifacts/training_summary.json

@@ -2,6 +2,6 @@
   "total_epochs": 2.571428571428571,
   "total_steps": "9",
   "final_train_loss": null,
-  "final_eval_loss": 0.6433929800987244,
-  "best_eval_loss": 0.6433929800987244
+  "final_eval_loss": 0.5185076594352722,
+  "best_eval_loss": 0.5185076594352722
 }