End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 1.0,
-    "eval_loss": 0.4685819447040558,
-    "eval_runtime": 16.4606,
+    "epoch": 0.33,
+    "eval_loss": 0.43880951404571533,
+    "eval_runtime": 16.7298,
     "eval_samples": 64,
-    "eval_samples_per_second": 3.888,
-    "eval_steps_per_second": 0.122,
-    "eval_wer": 51.28205128205128,
-    "train_loss": 0.8012711083889008,
-    "train_runtime": 2053.4009,
-    "train_samples_per_second": 3.117,
-    "train_steps_per_second": 0.097
+    "eval_samples_per_second": 3.825,
+    "eval_steps_per_second": 0.12,
+    "eval_wer": 46.52014652014652,
+    "train_loss": 0.10500287771224975,
+    "train_runtime": 1208.0467,
+    "train_samples_per_second": 7.947,
+    "train_steps_per_second": 0.248
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 1.0,
-    "eval_loss": 0.4685819447040558,
-    "eval_runtime": 16.4606,
+    "epoch": 0.33,
+    "eval_loss": 0.43880951404571533,
+    "eval_runtime": 16.7298,
     "eval_samples": 64,
-    "eval_samples_per_second": 3.888,
-    "eval_steps_per_second": 0.122,
-    "eval_wer": 51.28205128205128
+    "eval_samples_per_second": 3.825,
+    "eval_steps_per_second": 0.12,
+    "eval_wer": 46.52014652014652
 }

runs/Dec13_14-14-43_d7f040c448a8/events.out.tfevents.1670942499.d7f040c448a8.61677.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:017da71b5fc49a675e692d2511d60d4312881697dcd047d467bbf1e3e0b6cbf5
+size 358

train.log

@@ -244,3 +244,9 @@
 {'loss': 0.2921, 'learning_rate': 1.0344827586206898e-06, 'epoch': 0.33}
 {'eval_loss': 0.4282010793685913, 'eval_wer': 46.7032967032967, 'eval_runtime': 18.1178, 'eval_samples_per_second': 3.532, 'eval_steps_per_second': 0.11, 'epoch': 0.33}
 {'train_runtime': 1208.0467, 'train_samples_per_second': 7.947, 'train_steps_per_second': 0.248, 'train_loss': 0.10500287771224975, 'epoch': 0.33}
+***** train metrics *****
+  epoch                    =       0.33
+  train_loss               =      0.105
+  train_runtime            = 0:20:08.04
+  train_samples_per_second =      7.947
+  train_steps_per_second   =      0.248

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 1.0,
-    "train_loss": 0.8012711083889008,
-    "train_runtime": 2053.4009,
-    "train_samples_per_second": 3.117,
-    "train_steps_per_second": 0.097
+    "epoch": 0.33,
+    "train_loss": 0.10500287771224975,
+    "train_runtime": 1208.0467,
+    "train_samples_per_second": 7.947,
+    "train_steps_per_second": 0.248
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 51.28205128205128,
-  "best_model_checkpoint": "./checkpoint-190",
-  "epoch": 1.0,
-  "global_step": 200,
+  "best_metric": 46.52014652014652,
+  "best_model_checkpoint": "./checkpoint-280",
+  "epoch": 0.3333333333333333,
+  "global_step": 300,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -314,18 +314,168 @@
       "step": 200
     },
     {
-      "epoch": 1.0,
-      "step": 200,
-      "total_flos": 1.57560864768e+17,
-      "train_loss": 0.8012711083889008,
-      "train_runtime": 2053.4009,
-      "train_samples_per_second": 3.117,
-      "train_steps_per_second": 0.097
+      "epoch": 0.03,
+      "learning_rate": 3.206896551724138e-05,
+      "loss": 0.3956,
+      "step": 210
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 0.46904227137565613,
+      "eval_runtime": 16.6622,
+      "eval_samples_per_second": 3.841,
+      "eval_steps_per_second": 0.12,
+      "eval_wer": 52.01465201465202,
+      "step": 210
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 2.8620689655172417e-05,
+      "loss": 0.3719,
+      "step": 220
+    },
+    {
+      "epoch": 0.07,
+      "eval_loss": 0.4673417806625366,
+      "eval_runtime": 17.4548,
+      "eval_samples_per_second": 3.667,
+      "eval_steps_per_second": 0.115,
+      "eval_wer": 52.74725274725275,
+      "step": 220
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2.517241379310345e-05,
+      "loss": 0.3168,
+      "step": 230
+    },
+    {
+      "epoch": 0.1,
+      "eval_loss": 0.44994863867759705,
+      "eval_runtime": 17.0948,
+      "eval_samples_per_second": 3.744,
+      "eval_steps_per_second": 0.117,
+      "eval_wer": 51.46520146520146,
+      "step": 230
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 2.1724137931034484e-05,
+      "loss": 0.3582,
+      "step": 240
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.4524703323841095,
+      "eval_runtime": 17.7891,
+      "eval_samples_per_second": 3.598,
+      "eval_steps_per_second": 0.112,
+      "eval_wer": 46.88644688644688,
+      "step": 240
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.827586206896552e-05,
+      "loss": 0.2475,
+      "step": 250
+    },
+    {
+      "epoch": 0.17,
+      "eval_loss": 0.46121472120285034,
+      "eval_runtime": 17.7044,
+      "eval_samples_per_second": 3.615,
+      "eval_steps_per_second": 0.113,
+      "eval_wer": 52.38095238095239,
+      "step": 250
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.482758620689655e-05,
+      "loss": 0.2988,
+      "step": 260
+    },
+    {
+      "epoch": 0.2,
+      "eval_loss": 0.4346223473548889,
+      "eval_runtime": 19.6665,
+      "eval_samples_per_second": 3.254,
+      "eval_steps_per_second": 0.102,
+      "eval_wer": 49.81684981684982,
+      "step": 260
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.1379310344827587e-05,
+      "loss": 0.2749,
+      "step": 270
+    },
+    {
+      "epoch": 0.23,
+      "eval_loss": 0.42485401034355164,
+      "eval_runtime": 17.6854,
+      "eval_samples_per_second": 3.619,
+      "eval_steps_per_second": 0.113,
+      "eval_wer": 48.9010989010989,
+      "step": 270
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 7.93103448275862e-06,
+      "loss": 0.3368,
+      "step": 280
+    },
+    {
+      "epoch": 0.27,
+      "eval_loss": 0.43880951404571533,
+      "eval_runtime": 18.1285,
+      "eval_samples_per_second": 3.53,
+      "eval_steps_per_second": 0.11,
+      "eval_wer": 46.52014652014652,
+      "step": 280
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.482758620689655e-06,
+      "loss": 0.2574,
+      "step": 290
+    },
+    {
+      "epoch": 0.3,
+      "eval_loss": 0.43085092306137085,
+      "eval_runtime": 18.1023,
+      "eval_samples_per_second": 3.535,
+      "eval_steps_per_second": 0.11,
+      "eval_wer": 46.7032967032967,
+      "step": 290
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.0344827586206898e-06,
+      "loss": 0.2921,
+      "step": 300
+    },
+    {
+      "epoch": 0.33,
+      "eval_loss": 0.4282010793685913,
+      "eval_runtime": 18.1178,
+      "eval_samples_per_second": 3.532,
+      "eval_steps_per_second": 0.11,
+      "eval_wer": 46.7032967032967,
+      "step": 300
+    },
+    {
+      "epoch": 0.33,
+      "step": 300,
+      "total_flos": 2.36341297152e+17,
+      "train_loss": 0.10500287771224975,
+      "train_runtime": 1208.0467,
+      "train_samples_per_second": 7.947,
+      "train_steps_per_second": 0.248
     }
   ],
-  "max_steps": 200,
+  "max_steps": 300,
   "num_train_epochs": 9223372036854775807,
-  "total_flos": 1.57560864768e+17,
+  "total_flos": 2.36341297152e+17,
   "trial_name": null,
   "trial_params": null
 }