End of training

all_results.json

@@ -1,17 +1,17 @@
 {
-    "epoch": 2.13,
-    "eval_accuracy": 0.909112825458052,
-    "eval_f1": 0.8837125963089052,
-    "eval_loss": 0.6461763978004456,
-    "eval_precision": 0.879144385026738,
-    "eval_recall": 0.888328530259366,
-    "eval_runtime": 15.0345,
+    "epoch": 21.28,
+    "eval_accuracy": 0.9908389585342333,
+    "eval_f1": 0.9868823000898472,
+    "eval_loss": 0.05581057444214821,
+    "eval_precision": 0.9845822875582646,
+    "eval_recall": 0.989193083573487,
+    "eval_runtime": 15.2739,
     "eval_samples": 189,
-    "eval_samples_per_second": 12.571,
-    "eval_steps_per_second": 0.798,
-    "train_loss": 1.3162074279785156,
-    "train_runtime": 179.095,
+    "eval_samples_per_second": 12.374,
+    "eval_steps_per_second": 0.786,
+    "train_loss": 0.1608761215209961,
+    "train_runtime": 1880.5995,
     "train_samples": 752,
-    "train_samples_per_second": 8.934,
-    "train_steps_per_second": 0.558
+    "train_samples_per_second": 8.508,
+    "train_steps_per_second": 0.532
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 2.13,
-    "eval_accuracy": 0.909112825458052,
-    "eval_f1": 0.8837125963089052,
-    "eval_loss": 0.6461763978004456,
-    "eval_precision": 0.879144385026738,
-    "eval_recall": 0.888328530259366,
-    "eval_runtime": 15.0345,
+    "epoch": 21.28,
+    "eval_accuracy": 0.9908389585342333,
+    "eval_f1": 0.9868823000898472,
+    "eval_loss": 0.05581057444214821,
+    "eval_precision": 0.9845822875582646,
+    "eval_recall": 0.989193083573487,
+    "eval_runtime": 15.2739,
     "eval_samples": 189,
-    "eval_samples_per_second": 12.571,
-    "eval_steps_per_second": 0.798
+    "eval_samples_per_second": 12.374,
+    "eval_steps_per_second": 0.786
 }

runs/Dec20_17-18-14_pod-pasha/events.out.tfevents.1671558810.pod-pasha.541684.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df831f67092f73ac658ca971c347d14c5331db4b1283f747dd58af680fa232e8
+size 512

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.13,
-    "train_loss": 1.3162074279785156,
-    "train_runtime": 179.095,
+    "epoch": 21.28,
+    "train_loss": 0.1608761215209961,
+    "train_runtime": 1880.5995,
     "train_samples": 752,
-    "train_samples_per_second": 8.934,
-    "train_steps_per_second": 0.558
+    "train_samples_per_second": 8.508,
+    "train_steps_per_second": 0.532
 }

trainer_state.json

@@ -1,37 +1,157 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 2.127659574468085,
-  "global_step": 100,
+  "best_metric": 0.9868823000898472,
+  "best_model_checkpoint": "pasha/checkpoint-1000",
+  "epoch": 21.27659574468085,
+  "global_step": 1000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 2.13,
-      "eval_accuracy": 0.909112825458052,
-      "eval_f1": 0.8837125963089052,
-      "eval_loss": 0.6461763978004456,
-      "eval_precision": 0.879144385026738,
-      "eval_recall": 0.888328530259366,
-      "eval_runtime": 15.0755,
-      "eval_samples_per_second": 12.537,
-      "eval_steps_per_second": 0.796,
+      "eval_accuracy": 0.9566055930568949,
+      "eval_f1": 0.9482633863965269,
+      "eval_loss": 0.2661653161048889,
+      "eval_precision": 0.9523982558139535,
+      "eval_recall": 0.944164265129683,
+      "eval_runtime": 15.4131,
+      "eval_samples_per_second": 12.262,
+      "eval_steps_per_second": 0.779,
       "step": 100
     },
     {
-      "epoch": 2.13,
-      "step": 100,
-      "total_flos": 424705445068800.0,
-      "train_loss": 1.3162074279785156,
-      "train_runtime": 179.095,
-      "train_samples_per_second": 8.934,
-      "train_steps_per_second": 0.558
+      "epoch": 4.26,
+      "eval_accuracy": 0.9850530376084861,
+      "eval_f1": 0.9795185052102047,
+      "eval_loss": 0.1026068776845932,
+      "eval_precision": 0.9770609318996416,
+      "eval_recall": 0.9819884726224783,
+      "eval_runtime": 15.204,
+      "eval_samples_per_second": 12.431,
+      "eval_steps_per_second": 0.789,
+      "step": 200
+    },
+    {
+      "epoch": 6.38,
+      "eval_accuracy": 0.9884281581485053,
+      "eval_f1": 0.9849137931034483,
+      "eval_loss": 0.07217290997505188,
+      "eval_precision": 0.9820916905444126,
+      "eval_recall": 0.9877521613832853,
+      "eval_runtime": 15.2143,
+      "eval_samples_per_second": 12.423,
+      "eval_steps_per_second": 0.789,
+      "step": 300
+    },
+    {
+      "epoch": 8.51,
+      "eval_accuracy": 0.9891513982642237,
+      "eval_f1": 0.9857785778577858,
+      "eval_loss": 0.060767240822315216,
+      "eval_precision": 0.9852464915437208,
+      "eval_recall": 0.9863112391930836,
+      "eval_runtime": 15.3275,
+      "eval_samples_per_second": 12.331,
+      "eval_steps_per_second": 0.783,
+      "step": 400
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 5e-06,
+      "loss": 0.2962,
+      "step": 500
+    },
+    {
+      "epoch": 10.64,
+      "eval_accuracy": 0.9889103182256509,
+      "eval_f1": 0.9854185418541853,
+      "eval_loss": 0.060581281781196594,
+      "eval_precision": 0.9848866498740554,
+      "eval_recall": 0.9859510086455331,
+      "eval_runtime": 14.8027,
+      "eval_samples_per_second": 12.768,
+      "eval_steps_per_second": 0.811,
+      "step": 500
+    },
+    {
+      "epoch": 12.77,
+      "eval_accuracy": 0.9920443587270974,
+      "eval_f1": 0.988501616960115,
+      "eval_loss": 0.0517994724214077,
+      "eval_precision": 0.986021505376344,
+      "eval_recall": 0.9909942363112392,
+      "eval_runtime": 14.7335,
+      "eval_samples_per_second": 12.828,
+      "eval_steps_per_second": 0.814,
+      "step": 600
+    },
+    {
+      "epoch": 14.89,
+      "eval_accuracy": 0.9922854387656702,
+      "eval_f1": 0.988679245283019,
+      "eval_loss": 0.052589546889066696,
+      "eval_precision": 0.9863750448189316,
+      "eval_recall": 0.9909942363112392,
+      "eval_runtime": 14.9339,
+      "eval_samples_per_second": 12.656,
+      "eval_steps_per_second": 0.804,
+      "step": 700
+    },
+    {
+      "epoch": 17.02,
+      "eval_accuracy": 0.991321118611379,
+      "eval_f1": 0.9872416891284815,
+      "eval_loss": 0.05428989231586456,
+      "eval_precision": 0.984940839010398,
+      "eval_recall": 0.9895533141210374,
+      "eval_runtime": 14.8022,
+      "eval_samples_per_second": 12.768,
+      "eval_steps_per_second": 0.811,
+      "step": 800
+    },
+    {
+      "epoch": 19.15,
+      "eval_accuracy": 0.9910800385728061,
+      "eval_f1": 0.9867002156721782,
+      "eval_loss": 0.05573796480894089,
+      "eval_precision": 0.9845767575322812,
+      "eval_recall": 0.9888328530259366,
+      "eval_runtime": 14.7741,
+      "eval_samples_per_second": 12.793,
+      "eval_steps_per_second": 0.812,
+      "step": 900
+    },
+    {
+      "epoch": 21.28,
+      "learning_rate": 0.0,
+      "loss": 0.0255,
+      "step": 1000
+    },
+    {
+      "epoch": 21.28,
+      "eval_accuracy": 0.9908389585342333,
+      "eval_f1": 0.9868823000898472,
+      "eval_loss": 0.05581057444214821,
+      "eval_precision": 0.9845822875582646,
+      "eval_recall": 0.989193083573487,
+      "eval_runtime": 15.103,
+      "eval_samples_per_second": 12.514,
+      "eval_steps_per_second": 0.795,
+      "step": 1000
+    },
+    {
+      "epoch": 21.28,
+      "step": 1000,
+      "total_flos": 4247054450688000.0,
+      "train_loss": 0.1608761215209961,
+      "train_runtime": 1880.5995,
+      "train_samples_per_second": 8.508,
+      "train_steps_per_second": 0.532
     }
   ],
-  "max_steps": 100,
-  "num_train_epochs": 3,
-  "total_flos": 424705445068800.0,
+  "max_steps": 1000,
+  "num_train_epochs": 22,
+  "total_flos": 4247054450688000.0,
   "trial_name": null,
   "trial_params": null
 }