End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 15.0,
+    "eval_loss": 0.38286134600639343,
+    "eval_runtime": 65.8164,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 25.024,
+    "eval_steps_per_second": 3.13,
+    "eval_wer": 0.35777755081197016,
+    "train_loss": 1.0907053682053125,
+    "train_runtime": 3864.236,
+    "train_samples": 3478,
+    "train_samples_per_second": 13.501,
+    "train_steps_per_second": 0.423
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 15.0,
+    "eval_loss": 0.38286134600639343,
+    "eval_runtime": 65.8164,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 25.024,
+    "eval_steps_per_second": 3.13,
+    "eval_wer": 0.35777755081197016
+}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98aef17b53ec204c9f17fe1d08ee9444a89e4ecb069d535a81f5c7d1d36d6bdc
+oid sha256:9fa1d1cdcc4025c387c5270539b29f441c2b820329cc3c52df861070c78f81c8
 size 1262091761

runs/Dec11_08-56-09_cuda11-renwei-docker/events.out.tfevents.1639184319.cuda11-renwei-docker.1237.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b9eec3786b3784d93f1b48d2c21d18ce5349b6001524f6fad9afa1eb9cf6313
-size 6169
+oid sha256:0ca5b0c3367a683deb2f185831a15972ce362334035756c269e522403ef30265
+size 10810

runs/Dec11_08-56-09_cuda11-renwei-docker/events.out.tfevents.1639188251.cuda11-renwei-docker.1237.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:857a79f5545a422ef945f496b82359b6c4538e474fb873b5ccefaf92103b3462
+size 358

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 15.0,
+    "train_loss": 1.0907053682053125,
+    "train_runtime": 3864.236,
+    "train_samples": 3478,
+    "train_samples_per_second": 13.501,
+    "train_steps_per_second": 0.423
+}

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": 3.585611343383789,
+      "eval_runtime": 69.1996,
+      "eval_samples_per_second": 23.801,
+      "eval_steps_per_second": 2.977,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 1.83,
+      "eval_loss": 3.0358903408050537,
+      "eval_runtime": 70.0289,
+      "eval_samples_per_second": 23.519,
+      "eval_steps_per_second": 2.942,
+      "eval_wer": 0.9998978653865795,
+      "step": 200
+    },
+    {
+      "epoch": 2.75,
+      "eval_loss": 0.9341107606887817,
+      "eval_runtime": 69.5975,
+      "eval_samples_per_second": 23.665,
+      "eval_steps_per_second": 2.96,
+      "eval_wer": 0.7737718312736186,
+      "step": 300
+    },
+    {
+      "epoch": 3.67,
+      "eval_loss": 0.5874470472335815,
+      "eval_runtime": 69.8546,
+      "eval_samples_per_second": 23.578,
+      "eval_steps_per_second": 2.949,
+      "eval_wer": 0.6386477377183127,
+      "step": 400
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 0.0002988,
+      "loss": 3.214,
+      "step": 500
+    },
+    {
+      "epoch": 4.59,
+      "eval_loss": 0.5316070914268494,
+      "eval_runtime": 69.5825,
+      "eval_samples_per_second": 23.67,
+      "eval_steps_per_second": 2.961,
+      "eval_wer": 0.5525482586048411,
+      "step": 500
+    },
+    {
+      "epoch": 5.5,
+      "eval_loss": 0.45860984921455383,
+      "eval_runtime": 68.7814,
+      "eval_samples_per_second": 23.945,
+      "eval_steps_per_second": 2.995,
+      "eval_wer": 0.519252374629762,
+      "step": 600
+    },
+    {
+      "epoch": 6.42,
+      "eval_loss": 0.45034125447273254,
+      "eval_runtime": 67.077,
+      "eval_samples_per_second": 24.554,
+      "eval_steps_per_second": 3.071,
+      "eval_wer": 0.4605249719129813,
+      "step": 700
+    },
+    {
+      "epoch": 7.34,
+      "eval_loss": 0.41716983914375305,
+      "eval_runtime": 68.1176,
+      "eval_samples_per_second": 24.179,
+      "eval_steps_per_second": 3.024,
+      "eval_wer": 0.4390767030946788,
+      "step": 800
+    },
+    {
+      "epoch": 8.26,
+      "eval_loss": 0.4093872606754303,
+      "eval_runtime": 69.7848,
+      "eval_samples_per_second": 23.601,
+      "eval_steps_per_second": 2.952,
+      "eval_wer": 0.41854764579716064,
+      "step": 900
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 0.00016837004405286342,
+      "loss": 0.2284,
+      "step": 1000
+    },
+    {
+      "epoch": 9.17,
+      "eval_loss": 0.4091317355632782,
+      "eval_runtime": 70.1477,
+      "eval_samples_per_second": 23.479,
+      "eval_steps_per_second": 2.937,
+      "eval_wer": 0.39985701154121134,
+      "step": 1000
+    },
+    {
+      "epoch": 10.09,
+      "eval_loss": 0.39890098571777344,
+      "eval_runtime": 69.3262,
+      "eval_samples_per_second": 23.757,
+      "eval_steps_per_second": 2.971,
+      "eval_wer": 0.40404453069145135,
+      "step": 1100
+    },
+    {
+      "epoch": 11.01,
+      "eval_loss": 0.4066716432571411,
+      "eval_runtime": 69.7662,
+      "eval_samples_per_second": 23.607,
+      "eval_steps_per_second": 2.953,
+      "eval_wer": 0.3930139924420386,
+      "step": 1200
+    },
+    {
+      "epoch": 11.93,
+      "eval_loss": 0.40022560954093933,
+      "eval_runtime": 67.6908,
+      "eval_samples_per_second": 24.331,
+      "eval_steps_per_second": 3.043,
+      "eval_wer": 0.3775916658155449,
+      "step": 1300
+    },
+    {
+      "epoch": 12.84,
+      "eval_loss": 0.3982270061969757,
+      "eval_runtime": 66.5115,
+      "eval_samples_per_second": 24.763,
+      "eval_steps_per_second": 3.097,
+      "eval_wer": 0.36492697375140437,
+      "step": 1400
+    },
+    {
+      "epoch": 13.76,
+      "learning_rate": 3.621145374449339e-05,
+      "loss": 0.1029,
+      "step": 1500
+    },
+    {
+      "epoch": 13.76,
+      "eval_loss": 0.38969308137893677,
+      "eval_runtime": 65.75,
+      "eval_samples_per_second": 25.049,
+      "eval_steps_per_second": 3.133,
+      "eval_wer": 0.3703401082626902,
+      "step": 1500
+    },
+    {
+      "epoch": 14.68,
+      "eval_loss": 0.3836365342140198,
+      "eval_runtime": 65.8183,
+      "eval_samples_per_second": 25.023,
+      "eval_steps_per_second": 3.13,
+      "eval_wer": 0.3566540700643448,
+      "step": 1600
+    },
+    {
+      "epoch": 15.0,
+      "step": 1635,
+      "total_flos": 6.429866751096261e+18,
+      "train_loss": 1.0907053682053125,
+      "train_runtime": 3864.236,
+      "train_samples_per_second": 13.501,
+      "train_steps_per_second": 0.423
+    }
+  ],
+  "max_steps": 1635,
+  "num_train_epochs": 15,
+  "total_flos": 6.429866751096261e+18,
+  "trial_name": null,
+  "trial_params": null
+}