End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 15.0,
+    "eval_loss": 0.3793739378452301,
+    "eval_runtime": 78.0857,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 21.092,
+    "eval_steps_per_second": 2.638,
+    "eval_wer": 0.3446021856807272,
+    "train_loss": 1.0822357591868177,
+    "train_runtime": 3765.0907,
+    "train_samples": 3478,
+    "train_samples_per_second": 13.856,
+    "train_steps_per_second": 0.434
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 15.0,
+    "eval_loss": 0.3793739378452301,
+    "eval_runtime": 78.0857,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 21.092,
+    "eval_steps_per_second": 2.638,
+    "eval_wer": 0.3446021856807272
+}

runs/Dec19_10-36-20_gpu08/events.out.tfevents.1671452678.gpu08.217435.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e05381a497f1315e5aa6576d20638e142ae71f31ac454a9d2f44fe4f3cc2092
+size 358

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 15.0,
+    "train_loss": 1.0822357591868177,
+    "train_runtime": 3765.0907,
+    "train_samples": 3478,
+    "train_samples_per_second": 13.856,
+    "train_steps_per_second": 0.434
+}

trainer_state.json

@@ -0,0 +1,187 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 15.0,
+  "global_step": 1635,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.92,
+      "eval_loss": 3.595559597015381,
+      "eval_runtime": 78.9286,
+      "eval_samples_per_second": 20.867,
+      "eval_steps_per_second": 2.61,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 1.83,
+      "eval_loss": 3.0269057750701904,
+      "eval_runtime": 78.2182,
+      "eval_samples_per_second": 21.056,
+      "eval_steps_per_second": 2.634,
+      "eval_wer": 0.9998978653865795,
+      "step": 200
+    },
+    {
+      "epoch": 2.75,
+      "eval_loss": 0.9827112555503845,
+      "eval_runtime": 77.904,
+      "eval_samples_per_second": 21.141,
+      "eval_steps_per_second": 2.644,
+      "eval_wer": 0.8110509651720968,
+      "step": 300
+    },
+    {
+      "epoch": 3.67,
+      "eval_loss": 0.6235527396202087,
+      "eval_runtime": 78.104,
+      "eval_samples_per_second": 21.087,
+      "eval_steps_per_second": 2.638,
+      "eval_wer": 0.6303748340312532,
+      "step": 400
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 0.0002988,
+      "loss": 3.1866,
+      "step": 500
+    },
+    {
+      "epoch": 4.59,
+      "eval_loss": 0.5015883445739746,
+      "eval_runtime": 78.1295,
+      "eval_samples_per_second": 21.08,
+      "eval_steps_per_second": 2.637,
+      "eval_wer": 0.5264017975691963,
+      "step": 500
+    },
+    {
+      "epoch": 5.5,
+      "eval_loss": 0.4523301422595978,
+      "eval_runtime": 77.6138,
+      "eval_samples_per_second": 21.22,
+      "eval_steps_per_second": 2.654,
+      "eval_wer": 0.493514452047799,
+      "step": 600
+    },
+    {
+      "epoch": 6.42,
+      "eval_loss": 0.4305676221847534,
+      "eval_runtime": 78.2496,
+      "eval_samples_per_second": 21.048,
+      "eval_steps_per_second": 2.633,
+      "eval_wer": 0.4527627412930242,
+      "step": 700
+    },
+    {
+      "epoch": 7.34,
+      "eval_loss": 0.43279117345809937,
+      "eval_runtime": 78.3317,
+      "eval_samples_per_second": 21.026,
+      "eval_steps_per_second": 2.63,
+      "eval_wer": 0.43294862628944947,
+      "step": 800
+    },
+    {
+      "epoch": 8.26,
+      "eval_loss": 0.40258118510246277,
+      "eval_runtime": 77.3151,
+      "eval_samples_per_second": 21.302,
+      "eval_steps_per_second": 2.664,
+      "eval_wer": 0.4104790113369421,
+      "step": 900
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 0.00016837004405286342,
+      "loss": 0.227,
+      "step": 1000
+    },
+    {
+      "epoch": 9.17,
+      "eval_loss": 0.409604012966156,
+      "eval_runtime": 77.7312,
+      "eval_samples_per_second": 21.188,
+      "eval_steps_per_second": 2.65,
+      "eval_wer": 0.4080277806148504,
+      "step": 1000
+    },
+    {
+      "epoch": 10.09,
+      "eval_loss": 0.3921073079109192,
+      "eval_runtime": 78.0257,
+      "eval_samples_per_second": 21.108,
+      "eval_steps_per_second": 2.64,
+      "eval_wer": 0.3914819732407313,
+      "step": 1100
+    },
+    {
+      "epoch": 11.01,
+      "eval_loss": 0.3830062448978424,
+      "eval_runtime": 77.5372,
+      "eval_samples_per_second": 21.241,
+      "eval_steps_per_second": 2.657,
+      "eval_wer": 0.37779593504238584,
+      "step": 1200
+    },
+    {
+      "epoch": 11.93,
+      "eval_loss": 0.38462668657302856,
+      "eval_runtime": 77.1044,
+      "eval_samples_per_second": 21.361,
+      "eval_steps_per_second": 2.672,
+      "eval_wer": 0.36155653150852823,
+      "step": 1300
+    },
+    {
+      "epoch": 12.84,
+      "eval_loss": 0.3887845277786255,
+      "eval_runtime": 76.824,
+      "eval_samples_per_second": 21.439,
+      "eval_steps_per_second": 2.681,
+      "eval_wer": 0.3618629353487897,
+      "step": 1400
+    },
+    {
+      "epoch": 13.76,
+      "learning_rate": 3.621145374449339e-05,
+      "loss": 0.1046,
+      "step": 1500
+    },
+    {
+      "epoch": 13.76,
+      "eval_loss": 0.38607490062713623,
+      "eval_runtime": 78.3544,
+      "eval_samples_per_second": 21.02,
+      "eval_steps_per_second": 2.629,
+      "eval_wer": 0.3509345317127975,
+      "step": 1500
+    },
+    {
+      "epoch": 14.68,
+      "eval_loss": 0.3798215985298157,
+      "eval_runtime": 77.659,
+      "eval_samples_per_second": 21.208,
+      "eval_steps_per_second": 2.653,
+      "eval_wer": 0.3455213972015116,
+      "step": 1600
+    },
+    {
+      "epoch": 15.0,
+      "step": 1635,
+      "total_flos": 6.464559143942451e+18,
+      "train_loss": 1.0822357591868177,
+      "train_runtime": 3765.0907,
+      "train_samples_per_second": 13.856,
+      "train_steps_per_second": 0.434
+    }
+  ],
+  "max_steps": 1635,
+  "num_train_epochs": 15,
+  "total_flos": 6.464559143942451e+18,
+  "trial_name": null,
+  "trial_params": null
+}