End of training

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 25.0,
+    "epoch": 30.0,
     "eval_loss": 0.8710034489631653,
-    "eval_runtime": 324.5791,
-    "eval_samples_per_second": 1.577,
-    "eval_steps_per_second": 0.049,
+    "eval_runtime": 337.4912,
+    "eval_samples_per_second": 1.517,
+    "eval_steps_per_second": 0.047,
     "eval_wer": 60.05599273607748,
-    "train_loss": 0.16543263983726503,
-    "train_runtime": 2579.5458,
-    "train_samples_per_second": 24.811,
-    "train_steps_per_second": 0.388
+    "train_loss": 0.04157245059808095,
+    "train_runtime": 1332.7289,
+    "train_samples_per_second": 57.626,
+    "train_steps_per_second": 0.9
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 25.0,
+    "epoch": 30.0,
     "eval_loss": 0.8710034489631653,
-    "eval_runtime": 324.5791,
-    "eval_samples_per_second": 1.577,
-    "eval_steps_per_second": 0.049,
+    "eval_runtime": 337.4912,
+    "eval_samples_per_second": 1.517,
+    "eval_steps_per_second": 0.047,
     "eval_wer": 60.05599273607748
 }

runs/Dec20_19-12-39_129-146-32-172/events.out.tfevents.1671565316.129-146-32-172.141824.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec56a28eb4e411c0adfb0babd31b28a184b5ee3604d6eb7ab0a277729f38d64b
+size 358

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 25.0,
-    "train_loss": 0.16543263983726503,
-    "train_runtime": 2579.5458,
-    "train_samples_per_second": 24.811,
-    "train_steps_per_second": 0.388
+    "epoch": 30.0,
+    "train_loss": 0.04157245059808095,
+    "train_runtime": 1332.7289,
+    "train_samples_per_second": 57.626,
+    "train_steps_per_second": 0.9
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": 0.8710034489631653,
   "best_model_checkpoint": "./checkpoint-400",
-  "epoch": 25.0,
-  "global_step": 1000,
+  "epoch": 30.0,
+  "global_step": 1200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -698,18 +698,156 @@
       "step": 1000
     },
     {
-      "epoch": 25.0,
-      "step": 1000,
-      "total_flos": 4.048885260288e+18,
-      "train_loss": 0.16543263983726503,
-      "train_runtime": 2579.5458,
-      "train_samples_per_second": 24.811,
-      "train_steps_per_second": 0.388
+      "epoch": 25.25,
+      "learning_rate": 1.6581196581196582e-06,
+      "loss": 0.2704,
+      "step": 1010
+    },
+    {
+      "epoch": 25.5,
+      "learning_rate": 1.5726495726495727e-06,
+      "loss": 0.2582,
+      "step": 1020
+    },
+    {
+      "epoch": 25.75,
+      "learning_rate": 1.4871794871794873e-06,
+      "loss": 0.2634,
+      "step": 1030
+    },
+    {
+      "epoch": 26.0,
+      "learning_rate": 1.401709401709402e-06,
+      "loss": 0.2451,
+      "step": 1040
+    },
+    {
+      "epoch": 26.25,
+      "learning_rate": 1.3162393162393163e-06,
+      "loss": 0.2546,
+      "step": 1050
+    },
+    {
+      "epoch": 26.5,
+      "learning_rate": 1.230769230769231e-06,
+      "loss": 0.2604,
+      "step": 1060
+    },
+    {
+      "epoch": 26.75,
+      "learning_rate": 1.1452991452991454e-06,
+      "loss": 0.2459,
+      "step": 1070
+    },
+    {
+      "epoch": 27.0,
+      "learning_rate": 1.0598290598290598e-06,
+      "loss": 0.255,
+      "step": 1080
+    },
+    {
+      "epoch": 27.25,
+      "learning_rate": 9.743589743589745e-07,
+      "loss": 0.2505,
+      "step": 1090
+    },
+    {
+      "epoch": 27.5,
+      "learning_rate": 8.88888888888889e-07,
+      "loss": 0.2454,
+      "step": 1100
+    },
+    {
+      "epoch": 27.5,
+      "eval_loss": 0.9438697099685669,
+      "eval_runtime": 297.6469,
+      "eval_samples_per_second": 1.72,
+      "eval_steps_per_second": 0.054,
+      "eval_wer": 59.193401937046005,
+      "step": 1100
+    },
+    {
+      "epoch": 27.75,
+      "learning_rate": 8.034188034188035e-07,
+      "loss": 0.2506,
+      "step": 1110
+    },
+    {
+      "epoch": 28.0,
+      "learning_rate": 7.179487179487179e-07,
+      "loss": 0.2497,
+      "step": 1120
+    },
+    {
+      "epoch": 28.25,
+      "learning_rate": 6.324786324786325e-07,
+      "loss": 0.2477,
+      "step": 1130
+    },
+    {
+      "epoch": 28.5,
+      "learning_rate": 5.470085470085471e-07,
+      "loss": 0.2489,
+      "step": 1140
+    },
+    {
+      "epoch": 28.75,
+      "learning_rate": 4.615384615384616e-07,
+      "loss": 0.2428,
+      "step": 1150
+    },
+    {
+      "epoch": 29.0,
+      "learning_rate": 3.760683760683761e-07,
+      "loss": 0.2395,
+      "step": 1160
+    },
+    {
+      "epoch": 29.25,
+      "learning_rate": 2.905982905982906e-07,
+      "loss": 0.2438,
+      "step": 1170
+    },
+    {
+      "epoch": 29.5,
+      "learning_rate": 2.0512820512820514e-07,
+      "loss": 0.2403,
+      "step": 1180
+    },
+    {
+      "epoch": 29.75,
+      "learning_rate": 1.1965811965811967e-07,
+      "loss": 0.2469,
+      "step": 1190
+    },
+    {
+      "epoch": 30.0,
+      "learning_rate": 3.418803418803419e-08,
+      "loss": 0.2297,
+      "step": 1200
+    },
+    {
+      "epoch": 30.0,
+      "eval_loss": 0.9485259652137756,
+      "eval_runtime": 355.4454,
+      "eval_samples_per_second": 1.44,
+      "eval_steps_per_second": 0.045,
+      "eval_wer": 59.042070217917676,
+      "step": 1200
+    },
+    {
+      "epoch": 30.0,
+      "step": 1200,
+      "total_flos": 4.8586623123456e+18,
+      "train_loss": 0.04157245059808095,
+      "train_runtime": 1332.7289,
+      "train_samples_per_second": 57.626,
+      "train_steps_per_second": 0.9
     }
   ],
-  "max_steps": 1000,
-  "num_train_epochs": 25,
-  "total_flos": 4.048885260288e+18,
+  "max_steps": 1200,
+  "num_train_epochs": 30,
+  "total_flos": 4.8586623123456e+18,
   "trial_name": null,
   "trial_params": null
 }