End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 49.99,
|
3 |
+
"eval_loss": 0.41795814037323,
|
4 |
+
"eval_runtime": 62.4704,
|
5 |
+
"eval_samples": 561,
|
6 |
+
"eval_samples_per_second": 8.98,
|
7 |
+
"eval_steps_per_second": 2.257,
|
8 |
+
"eval_wer": 0.143314443915733,
|
9 |
+
"train_loss": 0.35037224591642185,
|
10 |
+
"train_runtime": 11843.2621,
|
11 |
+
"train_samples": 1116,
|
12 |
+
"train_samples_per_second": 4.712,
|
13 |
+
"train_steps_per_second": 0.291
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 49.99,
|
3 |
+
"eval_loss": 0.41795814037323,
|
4 |
+
"eval_runtime": 62.4704,
|
5 |
+
"eval_samples": 561,
|
6 |
+
"eval_samples_per_second": 8.98,
|
7 |
+
"eval_steps_per_second": 2.257,
|
8 |
+
"eval_wer": 0.143314443915733
|
9 |
+
}
|
runs/Nov30_06-03-08_83be21523c57/events.out.tfevents.1669806548.83be21523c57.21519.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3c0f8117caac0da2bdb1ce360f054bf974c2c61ba330957ccea5411674e81094
|
3 |
+
size 358
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 49.99,
|
3 |
+
"train_loss": 0.35037224591642185,
|
4 |
+
"train_runtime": 11843.2621,
|
5 |
+
"train_samples": 1116,
|
6 |
+
"train_samples_per_second": 4.712,
|
7 |
+
"train_steps_per_second": 0.291
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2545 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.41795814037323,
|
3 |
+
"best_model_checkpoint": "ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-13k_onset-drums_fold_2/checkpoint-1863",
|
4 |
+
"epoch": 49.98924731182796,
|
5 |
+
"global_step": 3450,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.14,
|
12 |
+
"learning_rate": 2.9999999999999997e-05,
|
13 |
+
"loss": 0.3312,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.29,
|
18 |
+
"learning_rate": 5.6999999999999996e-05,
|
19 |
+
"loss": 0.7196,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.43,
|
24 |
+
"learning_rate": 8.699999999999999e-05,
|
25 |
+
"loss": 0.3342,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.57,
|
30 |
+
"learning_rate": 0.000117,
|
31 |
+
"loss": 0.5121,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.72,
|
36 |
+
"learning_rate": 0.000147,
|
37 |
+
"loss": 0.6408,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.86,
|
42 |
+
"learning_rate": 0.00017699999999999997,
|
43 |
+
"loss": 0.3238,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.99,
|
48 |
+
"eval_loss": 0.45810404419898987,
|
49 |
+
"eval_runtime": 62.3289,
|
50 |
+
"eval_samples_per_second": 9.001,
|
51 |
+
"eval_steps_per_second": 2.262,
|
52 |
+
"eval_wer": 0.20811523986676472,
|
53 |
+
"step": 69
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 1.01,
|
57 |
+
"learning_rate": 0.00020699999999999996,
|
58 |
+
"loss": 0.8089,
|
59 |
+
"step": 70
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.16,
|
63 |
+
"learning_rate": 0.000237,
|
64 |
+
"loss": 0.2476,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 1.3,
|
69 |
+
"learning_rate": 0.000267,
|
70 |
+
"loss": 0.7357,
|
71 |
+
"step": 90
|
72 |
+
},
|
73 |
+
{
|
74 |
+
"epoch": 1.44,
|
75 |
+
"learning_rate": 0.00029699999999999996,
|
76 |
+
"loss": 0.4624,
|
77 |
+
"step": 100
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 1.59,
|
81 |
+
"learning_rate": 0.00029919402985074625,
|
82 |
+
"loss": 0.4802,
|
83 |
+
"step": 110
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 1.73,
|
87 |
+
"learning_rate": 0.0002982985074626866,
|
88 |
+
"loss": 0.7866,
|
89 |
+
"step": 120
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 1.87,
|
93 |
+
"learning_rate": 0.00029740298507462685,
|
94 |
+
"loss": 0.275,
|
95 |
+
"step": 130
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 1.99,
|
99 |
+
"eval_loss": 0.6493655443191528,
|
100 |
+
"eval_runtime": 63.3598,
|
101 |
+
"eval_samples_per_second": 8.854,
|
102 |
+
"eval_steps_per_second": 2.225,
|
103 |
+
"eval_wer": 0.33425617510922695,
|
104 |
+
"step": 138
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 2.03,
|
108 |
+
"learning_rate": 0.0002965074626865671,
|
109 |
+
"loss": 0.7862,
|
110 |
+
"step": 140
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 2.17,
|
114 |
+
"learning_rate": 0.00029561194029850745,
|
115 |
+
"loss": 0.3001,
|
116 |
+
"step": 150
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 2.32,
|
120 |
+
"learning_rate": 0.0002947164179104477,
|
121 |
+
"loss": 0.6909,
|
122 |
+
"step": 160
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 2.46,
|
126 |
+
"learning_rate": 0.00029382089552238805,
|
127 |
+
"loss": 0.4748,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 2.6,
|
132 |
+
"learning_rate": 0.0002929253731343283,
|
133 |
+
"loss": 0.4152,
|
134 |
+
"step": 180
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 2.75,
|
138 |
+
"learning_rate": 0.00029202985074626864,
|
139 |
+
"loss": 0.7598,
|
140 |
+
"step": 190
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 2.89,
|
144 |
+
"learning_rate": 0.0002911343283582089,
|
145 |
+
"loss": 0.2965,
|
146 |
+
"step": 200
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 2.99,
|
150 |
+
"eval_loss": 0.6193479895591736,
|
151 |
+
"eval_runtime": 62.3049,
|
152 |
+
"eval_samples_per_second": 9.004,
|
153 |
+
"eval_steps_per_second": 2.263,
|
154 |
+
"eval_wer": 0.2274516589522862,
|
155 |
+
"step": 207
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 3.04,
|
159 |
+
"learning_rate": 0.0002903283582089552,
|
160 |
+
"loss": 0.8123,
|
161 |
+
"step": 210
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 3.19,
|
165 |
+
"learning_rate": 0.00028943283582089546,
|
166 |
+
"loss": 0.4209,
|
167 |
+
"step": 220
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 3.33,
|
171 |
+
"learning_rate": 0.0002885373134328358,
|
172 |
+
"loss": 0.604,
|
173 |
+
"step": 230
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 3.47,
|
177 |
+
"learning_rate": 0.0002876417910447761,
|
178 |
+
"loss": 0.5433,
|
179 |
+
"step": 240
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 3.62,
|
183 |
+
"learning_rate": 0.0002867462686567164,
|
184 |
+
"loss": 0.3383,
|
185 |
+
"step": 250
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 3.76,
|
189 |
+
"learning_rate": 0.0002858507462686567,
|
190 |
+
"loss": 0.7398,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 3.9,
|
195 |
+
"learning_rate": 0.000284955223880597,
|
196 |
+
"loss": 0.3406,
|
197 |
+
"step": 270
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 3.99,
|
201 |
+
"eval_loss": 0.6934247612953186,
|
202 |
+
"eval_runtime": 61.6757,
|
203 |
+
"eval_samples_per_second": 9.096,
|
204 |
+
"eval_steps_per_second": 2.286,
|
205 |
+
"eval_wer": 0.2614958688411126,
|
206 |
+
"step": 276
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 4.06,
|
210 |
+
"learning_rate": 0.00028405970149253726,
|
211 |
+
"loss": 0.7566,
|
212 |
+
"step": 280
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 4.2,
|
216 |
+
"learning_rate": 0.0002831641791044776,
|
217 |
+
"loss": 0.4127,
|
218 |
+
"step": 290
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 4.34,
|
222 |
+
"learning_rate": 0.0002822686567164179,
|
223 |
+
"loss": 0.5787,
|
224 |
+
"step": 300
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 4.49,
|
228 |
+
"learning_rate": 0.0002813731343283582,
|
229 |
+
"loss": 0.8365,
|
230 |
+
"step": 310
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 4.63,
|
234 |
+
"learning_rate": 0.00028047761194029846,
|
235 |
+
"loss": 0.2287,
|
236 |
+
"step": 320
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 4.77,
|
240 |
+
"learning_rate": 0.0002795820895522388,
|
241 |
+
"loss": 0.6623,
|
242 |
+
"step": 330
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 4.92,
|
246 |
+
"learning_rate": 0.00027868656716417906,
|
247 |
+
"loss": 0.3906,
|
248 |
+
"step": 340
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 4.99,
|
252 |
+
"eval_loss": 0.6265394687652588,
|
253 |
+
"eval_runtime": 61.9233,
|
254 |
+
"eval_samples_per_second": 9.06,
|
255 |
+
"eval_steps_per_second": 2.277,
|
256 |
+
"eval_wer": 0.1835445775835965,
|
257 |
+
"step": 345
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 5.07,
|
261 |
+
"learning_rate": 0.0002777910447761194,
|
262 |
+
"loss": 0.5863,
|
263 |
+
"step": 350
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 5.22,
|
267 |
+
"learning_rate": 0.00027689552238805966,
|
268 |
+
"loss": 0.4848,
|
269 |
+
"step": 360
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 5.36,
|
273 |
+
"learning_rate": 0.000276,
|
274 |
+
"loss": 0.4226,
|
275 |
+
"step": 370
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 5.5,
|
279 |
+
"learning_rate": 0.00027510447761194026,
|
280 |
+
"loss": 0.6544,
|
281 |
+
"step": 380
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 5.65,
|
285 |
+
"learning_rate": 0.0002742089552238806,
|
286 |
+
"loss": 0.2422,
|
287 |
+
"step": 390
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 5.79,
|
291 |
+
"learning_rate": 0.00027331343283582086,
|
292 |
+
"loss": 0.6015,
|
293 |
+
"step": 400
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 5.93,
|
297 |
+
"learning_rate": 0.0002724179104477612,
|
298 |
+
"loss": 0.4643,
|
299 |
+
"step": 410
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 5.99,
|
303 |
+
"eval_loss": 0.5878893733024597,
|
304 |
+
"eval_runtime": 61.4594,
|
305 |
+
"eval_samples_per_second": 9.128,
|
306 |
+
"eval_steps_per_second": 2.294,
|
307 |
+
"eval_wer": 0.18990353419561362,
|
308 |
+
"step": 414
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 6.09,
|
312 |
+
"learning_rate": 0.00027152238805970146,
|
313 |
+
"loss": 0.61,
|
314 |
+
"step": 420
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 6.23,
|
318 |
+
"learning_rate": 0.00027062686567164173,
|
319 |
+
"loss": 0.5311,
|
320 |
+
"step": 430
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 6.37,
|
324 |
+
"learning_rate": 0.00026973134328358206,
|
325 |
+
"loss": 0.2503,
|
326 |
+
"step": 440
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 6.52,
|
330 |
+
"learning_rate": 0.0002688358208955224,
|
331 |
+
"loss": 0.7559,
|
332 |
+
"step": 450
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 6.66,
|
336 |
+
"learning_rate": 0.00026794029850746266,
|
337 |
+
"loss": 0.314,
|
338 |
+
"step": 460
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 6.8,
|
342 |
+
"learning_rate": 0.00026704477611940293,
|
343 |
+
"loss": 0.6115,
|
344 |
+
"step": 470
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 6.95,
|
348 |
+
"learning_rate": 0.00026614925373134326,
|
349 |
+
"loss": 0.4652,
|
350 |
+
"step": 480
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 6.99,
|
354 |
+
"eval_loss": 0.4961091876029968,
|
355 |
+
"eval_runtime": 61.4725,
|
356 |
+
"eval_samples_per_second": 9.126,
|
357 |
+
"eval_steps_per_second": 2.294,
|
358 |
+
"eval_wer": 0.16044469438075876,
|
359 |
+
"step": 483
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 7.1,
|
363 |
+
"learning_rate": 0.0002652537313432836,
|
364 |
+
"loss": 0.4757,
|
365 |
+
"step": 490
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 7.24,
|
369 |
+
"learning_rate": 0.00026435820895522386,
|
370 |
+
"loss": 0.694,
|
371 |
+
"step": 500
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 7.39,
|
375 |
+
"learning_rate": 0.0002634626865671642,
|
376 |
+
"loss": 0.1975,
|
377 |
+
"step": 510
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 7.53,
|
381 |
+
"learning_rate": 0.00026256716417910446,
|
382 |
+
"loss": 0.6958,
|
383 |
+
"step": 520
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 7.67,
|
387 |
+
"learning_rate": 0.00026167164179104473,
|
388 |
+
"loss": 0.3152,
|
389 |
+
"step": 530
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 7.82,
|
393 |
+
"learning_rate": 0.00026077611940298506,
|
394 |
+
"loss": 0.4814,
|
395 |
+
"step": 540
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 7.96,
|
399 |
+
"learning_rate": 0.0002598805970149254,
|
400 |
+
"loss": 0.4512,
|
401 |
+
"step": 550
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 7.99,
|
405 |
+
"eval_loss": 0.571247935295105,
|
406 |
+
"eval_runtime": 62.044,
|
407 |
+
"eval_samples_per_second": 9.042,
|
408 |
+
"eval_steps_per_second": 2.273,
|
409 |
+
"eval_wer": 0.2800536401782238,
|
410 |
+
"step": 552
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 8.11,
|
414 |
+
"learning_rate": 0.00025898507462686566,
|
415 |
+
"loss": 0.3612,
|
416 |
+
"step": 560
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 8.26,
|
420 |
+
"learning_rate": 0.00025808955223880593,
|
421 |
+
"loss": 0.7199,
|
422 |
+
"step": 570
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 8.4,
|
426 |
+
"learning_rate": 0.00025719402985074626,
|
427 |
+
"loss": 0.2195,
|
428 |
+
"step": 580
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 8.54,
|
432 |
+
"learning_rate": 0.00025629850746268653,
|
433 |
+
"loss": 0.673,
|
434 |
+
"step": 590
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 8.69,
|
438 |
+
"learning_rate": 0.00025540298507462685,
|
439 |
+
"loss": 0.3611,
|
440 |
+
"step": 600
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 8.83,
|
444 |
+
"learning_rate": 0.00025450746268656713,
|
445 |
+
"loss": 0.3776,
|
446 |
+
"step": 610
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 8.97,
|
450 |
+
"learning_rate": 0.00025361194029850745,
|
451 |
+
"loss": 0.5321,
|
452 |
+
"step": 620
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 8.99,
|
456 |
+
"eval_loss": 0.6898208260536194,
|
457 |
+
"eval_runtime": 62.8462,
|
458 |
+
"eval_samples_per_second": 8.927,
|
459 |
+
"eval_steps_per_second": 2.244,
|
460 |
+
"eval_wer": 0.29363671756715837,
|
461 |
+
"step": 621
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 9.13,
|
465 |
+
"learning_rate": 0.00025271641791044773,
|
466 |
+
"loss": 0.2462,
|
467 |
+
"step": 630
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 9.27,
|
471 |
+
"learning_rate": 0.00025182089552238805,
|
472 |
+
"loss": 0.5621,
|
473 |
+
"step": 640
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 9.42,
|
477 |
+
"learning_rate": 0.0002509253731343283,
|
478 |
+
"loss": 0.2758,
|
479 |
+
"step": 650
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 9.56,
|
483 |
+
"learning_rate": 0.00025002985074626865,
|
484 |
+
"loss": 0.6596,
|
485 |
+
"step": 660
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 9.7,
|
489 |
+
"learning_rate": 0.0002491343283582089,
|
490 |
+
"loss": 0.3786,
|
491 |
+
"step": 670
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 9.85,
|
495 |
+
"learning_rate": 0.0002482388059701492,
|
496 |
+
"loss": 0.4499,
|
497 |
+
"step": 680
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 9.99,
|
501 |
+
"learning_rate": 0.0002473432835820895,
|
502 |
+
"loss": 0.64,
|
503 |
+
"step": 690
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 9.99,
|
507 |
+
"eval_loss": 0.5916096568107605,
|
508 |
+
"eval_runtime": 62.3005,
|
509 |
+
"eval_samples_per_second": 9.005,
|
510 |
+
"eval_steps_per_second": 2.263,
|
511 |
+
"eval_wer": 0.26482675087597873,
|
512 |
+
"step": 690
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 10.14,
|
516 |
+
"learning_rate": 0.00024644776119402985,
|
517 |
+
"loss": 0.241,
|
518 |
+
"step": 700
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 10.29,
|
522 |
+
"learning_rate": 0.0002455522388059701,
|
523 |
+
"loss": 0.5644,
|
524 |
+
"step": 710
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 10.43,
|
528 |
+
"learning_rate": 0.0002446567164179104,
|
529 |
+
"loss": 0.2882,
|
530 |
+
"step": 720
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 10.57,
|
534 |
+
"learning_rate": 0.00024376119402985073,
|
535 |
+
"loss": 0.4583,
|
536 |
+
"step": 730
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 10.72,
|
540 |
+
"learning_rate": 0.00024286567164179103,
|
541 |
+
"loss": 0.5087,
|
542 |
+
"step": 740
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 10.86,
|
546 |
+
"learning_rate": 0.00024197014925373132,
|
547 |
+
"loss": 0.2959,
|
548 |
+
"step": 750
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 10.99,
|
552 |
+
"eval_loss": 0.5573664307594299,
|
553 |
+
"eval_runtime": 62.7291,
|
554 |
+
"eval_samples_per_second": 8.943,
|
555 |
+
"eval_steps_per_second": 2.248,
|
556 |
+
"eval_wer": 0.17450361206038845,
|
557 |
+
"step": 759
|
558 |
+
},
|
559 |
+
{
|
560 |
+
"epoch": 11.01,
|
561 |
+
"learning_rate": 0.0002410746268656716,
|
562 |
+
"loss": 0.6544,
|
563 |
+
"step": 760
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 11.16,
|
567 |
+
"learning_rate": 0.00024017910447761192,
|
568 |
+
"loss": 0.1627,
|
569 |
+
"step": 770
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 11.3,
|
573 |
+
"learning_rate": 0.00023928358208955222,
|
574 |
+
"loss": 0.5087,
|
575 |
+
"step": 780
|
576 |
+
},
|
577 |
+
{
|
578 |
+
"epoch": 11.44,
|
579 |
+
"learning_rate": 0.0002383880597014925,
|
580 |
+
"loss": 0.4087,
|
581 |
+
"step": 790
|
582 |
+
},
|
583 |
+
{
|
584 |
+
"epoch": 11.59,
|
585 |
+
"learning_rate": 0.00023749253731343282,
|
586 |
+
"loss": 0.3695,
|
587 |
+
"step": 800
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 11.73,
|
591 |
+
"learning_rate": 0.00023659701492537312,
|
592 |
+
"loss": 0.5399,
|
593 |
+
"step": 810
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 11.87,
|
597 |
+
"learning_rate": 0.0002357014925373134,
|
598 |
+
"loss": 0.2053,
|
599 |
+
"step": 820
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 11.99,
|
603 |
+
"eval_loss": 0.5216460824012756,
|
604 |
+
"eval_runtime": 62.584,
|
605 |
+
"eval_samples_per_second": 8.964,
|
606 |
+
"eval_steps_per_second": 2.253,
|
607 |
+
"eval_wer": 0.20093437729809233,
|
608 |
+
"step": 828
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 12.03,
|
612 |
+
"learning_rate": 0.0002348059701492537,
|
613 |
+
"loss": 0.6779,
|
614 |
+
"step": 830
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 12.17,
|
618 |
+
"learning_rate": 0.00023391044776119402,
|
619 |
+
"loss": 0.2308,
|
620 |
+
"step": 840
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 12.32,
|
624 |
+
"learning_rate": 0.0002330149253731343,
|
625 |
+
"loss": 0.4464,
|
626 |
+
"step": 850
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 12.46,
|
630 |
+
"learning_rate": 0.0002321194029850746,
|
631 |
+
"loss": 0.3519,
|
632 |
+
"step": 860
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 12.6,
|
636 |
+
"learning_rate": 0.00023122388059701492,
|
637 |
+
"loss": 0.3585,
|
638 |
+
"step": 870
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 12.75,
|
642 |
+
"learning_rate": 0.00023032835820895522,
|
643 |
+
"loss": 0.5939,
|
644 |
+
"step": 880
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 12.89,
|
648 |
+
"learning_rate": 0.0002294328358208955,
|
649 |
+
"loss": 0.2433,
|
650 |
+
"step": 890
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 12.99,
|
654 |
+
"eval_loss": 0.4737621247768402,
|
655 |
+
"eval_runtime": 62.102,
|
656 |
+
"eval_samples_per_second": 9.034,
|
657 |
+
"eval_steps_per_second": 2.27,
|
658 |
+
"eval_wer": 0.16425141670632,
|
659 |
+
"step": 897
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 13.04,
|
663 |
+
"learning_rate": 0.0002285373134328358,
|
664 |
+
"loss": 0.6217,
|
665 |
+
"step": 900
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 13.19,
|
669 |
+
"learning_rate": 0.00022764179104477612,
|
670 |
+
"loss": 0.2514,
|
671 |
+
"step": 910
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 13.33,
|
675 |
+
"learning_rate": 0.0002267462686567164,
|
676 |
+
"loss": 0.4339,
|
677 |
+
"step": 920
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 13.47,
|
681 |
+
"learning_rate": 0.0002258507462686567,
|
682 |
+
"loss": 0.459,
|
683 |
+
"step": 930
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 13.62,
|
687 |
+
"learning_rate": 0.00022495522388059697,
|
688 |
+
"loss": 0.2808,
|
689 |
+
"step": 940
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 13.76,
|
693 |
+
"learning_rate": 0.0002240597014925373,
|
694 |
+
"loss": 0.5249,
|
695 |
+
"step": 950
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 13.9,
|
699 |
+
"learning_rate": 0.0002231641791044776,
|
700 |
+
"loss": 0.2036,
|
701 |
+
"step": 960
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 13.99,
|
705 |
+
"eval_loss": 0.5062561631202698,
|
706 |
+
"eval_runtime": 62.2056,
|
707 |
+
"eval_samples_per_second": 9.018,
|
708 |
+
"eval_steps_per_second": 2.267,
|
709 |
+
"eval_wer": 0.1650733226629753,
|
710 |
+
"step": 966
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 14.06,
|
714 |
+
"learning_rate": 0.0002222686567164179,
|
715 |
+
"loss": 0.54,
|
716 |
+
"step": 970
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 14.2,
|
720 |
+
"learning_rate": 0.0002213731343283582,
|
721 |
+
"loss": 0.2911,
|
722 |
+
"step": 980
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 14.34,
|
726 |
+
"learning_rate": 0.0002204776119402985,
|
727 |
+
"loss": 0.3306,
|
728 |
+
"step": 990
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 14.49,
|
732 |
+
"learning_rate": 0.0002195820895522388,
|
733 |
+
"loss": 0.5251,
|
734 |
+
"step": 1000
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 14.63,
|
738 |
+
"learning_rate": 0.00021868656716417907,
|
739 |
+
"loss": 0.1507,
|
740 |
+
"step": 1010
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 14.77,
|
744 |
+
"learning_rate": 0.0002177910447761194,
|
745 |
+
"loss": 0.5758,
|
746 |
+
"step": 1020
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 14.92,
|
750 |
+
"learning_rate": 0.0002168955223880597,
|
751 |
+
"loss": 0.2654,
|
752 |
+
"step": 1030
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 14.99,
|
756 |
+
"eval_loss": 0.4903617799282074,
|
757 |
+
"eval_runtime": 62.3504,
|
758 |
+
"eval_samples_per_second": 8.998,
|
759 |
+
"eval_steps_per_second": 2.261,
|
760 |
+
"eval_wer": 0.15114417960808063,
|
761 |
+
"step": 1035
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 15.07,
|
765 |
+
"learning_rate": 0.00021599999999999996,
|
766 |
+
"loss": 0.505,
|
767 |
+
"step": 1040
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 15.22,
|
771 |
+
"learning_rate": 0.0002151044776119403,
|
772 |
+
"loss": 0.362,
|
773 |
+
"step": 1050
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 15.36,
|
777 |
+
"learning_rate": 0.0002142089552238806,
|
778 |
+
"loss": 0.3129,
|
779 |
+
"step": 1060
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 15.5,
|
783 |
+
"learning_rate": 0.00021331343283582086,
|
784 |
+
"loss": 0.5513,
|
785 |
+
"step": 1070
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 15.65,
|
789 |
+
"learning_rate": 0.00021241791044776116,
|
790 |
+
"loss": 0.1605,
|
791 |
+
"step": 1080
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 15.79,
|
795 |
+
"learning_rate": 0.0002115223880597015,
|
796 |
+
"loss": 0.4756,
|
797 |
+
"step": 1090
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 15.93,
|
801 |
+
"learning_rate": 0.00021062686567164176,
|
802 |
+
"loss": 0.3641,
|
803 |
+
"step": 1100
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 15.99,
|
807 |
+
"eval_loss": 0.4660147726535797,
|
808 |
+
"eval_runtime": 62.4193,
|
809 |
+
"eval_samples_per_second": 8.988,
|
810 |
+
"eval_steps_per_second": 2.259,
|
811 |
+
"eval_wer": 0.1668901674092659,
|
812 |
+
"step": 1104
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 16.09,
|
816 |
+
"learning_rate": 0.00020973134328358206,
|
817 |
+
"loss": 0.3724,
|
818 |
+
"step": 1110
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 16.23,
|
822 |
+
"learning_rate": 0.00020883582089552236,
|
823 |
+
"loss": 0.3679,
|
824 |
+
"step": 1120
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 16.37,
|
828 |
+
"learning_rate": 0.0002079402985074627,
|
829 |
+
"loss": 0.2284,
|
830 |
+
"step": 1130
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 16.52,
|
834 |
+
"learning_rate": 0.00020704477611940296,
|
835 |
+
"loss": 0.5271,
|
836 |
+
"step": 1140
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 16.66,
|
840 |
+
"learning_rate": 0.00020614925373134326,
|
841 |
+
"loss": 0.2279,
|
842 |
+
"step": 1150
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 16.8,
|
846 |
+
"learning_rate": 0.0002052537313432836,
|
847 |
+
"loss": 0.4453,
|
848 |
+
"step": 1160
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 16.95,
|
852 |
+
"learning_rate": 0.00020435820895522386,
|
853 |
+
"loss": 0.373,
|
854 |
+
"step": 1170
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 16.99,
|
858 |
+
"eval_loss": 0.5132745504379272,
|
859 |
+
"eval_runtime": 63.5973,
|
860 |
+
"eval_samples_per_second": 8.821,
|
861 |
+
"eval_steps_per_second": 2.217,
|
862 |
+
"eval_wer": 0.21058095773673055,
|
863 |
+
"step": 1173
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 17.1,
|
867 |
+
"learning_rate": 0.00020346268656716416,
|
868 |
+
"loss": 0.3844,
|
869 |
+
"step": 1180
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 17.24,
|
873 |
+
"learning_rate": 0.00020256716417910443,
|
874 |
+
"loss": 0.4336,
|
875 |
+
"step": 1190
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 17.39,
|
879 |
+
"learning_rate": 0.00020167164179104476,
|
880 |
+
"loss": 0.1497,
|
881 |
+
"step": 1200
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 17.53,
|
885 |
+
"learning_rate": 0.00020077611940298506,
|
886 |
+
"loss": 0.4863,
|
887 |
+
"step": 1210
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 17.67,
|
891 |
+
"learning_rate": 0.00019988059701492536,
|
892 |
+
"loss": 0.2108,
|
893 |
+
"step": 1220
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 17.82,
|
897 |
+
"learning_rate": 0.00019898507462686566,
|
898 |
+
"loss": 0.3513,
|
899 |
+
"step": 1230
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 17.96,
|
903 |
+
"learning_rate": 0.00019808955223880596,
|
904 |
+
"loss": 0.4715,
|
905 |
+
"step": 1240
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 17.99,
|
909 |
+
"eval_loss": 0.5312526226043701,
|
910 |
+
"eval_runtime": 62.3777,
|
911 |
+
"eval_samples_per_second": 8.994,
|
912 |
+
"eval_steps_per_second": 2.26,
|
913 |
+
"eval_wer": 0.19120128044296406,
|
914 |
+
"step": 1242
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 18.11,
|
918 |
+
"learning_rate": 0.00019719402985074626,
|
919 |
+
"loss": 0.2865,
|
920 |
+
"step": 1250
|
921 |
+
},
|
922 |
+
{
|
923 |
+
"epoch": 18.26,
|
924 |
+
"learning_rate": 0.00019629850746268653,
|
925 |
+
"loss": 0.5362,
|
926 |
+
"step": 1260
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 18.4,
|
930 |
+
"learning_rate": 0.00019540298507462686,
|
931 |
+
"loss": 0.178,
|
932 |
+
"step": 1270
|
933 |
+
},
|
934 |
+
{
|
935 |
+
"epoch": 18.54,
|
936 |
+
"learning_rate": 0.00019450746268656716,
|
937 |
+
"loss": 0.4632,
|
938 |
+
"step": 1280
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 18.69,
|
942 |
+
"learning_rate": 0.00019361194029850743,
|
943 |
+
"loss": 0.2589,
|
944 |
+
"step": 1290
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 18.83,
|
948 |
+
"learning_rate": 0.00019271641791044773,
|
949 |
+
"loss": 0.346,
|
950 |
+
"step": 1300
|
951 |
+
},
|
952 |
+
{
|
953 |
+
"epoch": 18.97,
|
954 |
+
"learning_rate": 0.00019182089552238806,
|
955 |
+
"loss": 0.4893,
|
956 |
+
"step": 1310
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 18.99,
|
960 |
+
"eval_loss": 0.5152482390403748,
|
961 |
+
"eval_runtime": 62.4572,
|
962 |
+
"eval_samples_per_second": 8.982,
|
963 |
+
"eval_steps_per_second": 2.258,
|
964 |
+
"eval_wer": 0.17117273002552233,
|
965 |
+
"step": 1311
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 19.13,
|
969 |
+
"learning_rate": 0.00019092537313432833,
|
970 |
+
"loss": 0.2141,
|
971 |
+
"step": 1320
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 19.27,
|
975 |
+
"learning_rate": 0.00019002985074626863,
|
976 |
+
"loss": 0.5869,
|
977 |
+
"step": 1330
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 19.42,
|
981 |
+
"learning_rate": 0.00018913432835820896,
|
982 |
+
"loss": 0.2126,
|
983 |
+
"step": 1340
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 19.56,
|
987 |
+
"learning_rate": 0.00018823880597014923,
|
988 |
+
"loss": 0.4274,
|
989 |
+
"step": 1350
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 19.7,
|
993 |
+
"learning_rate": 0.00018734328358208953,
|
994 |
+
"loss": 0.3929,
|
995 |
+
"step": 1360
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 19.85,
|
999 |
+
"learning_rate": 0.00018644776119402983,
|
1000 |
+
"loss": 0.259,
|
1001 |
+
"step": 1370
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 19.99,
|
1005 |
+
"learning_rate": 0.00018555223880597013,
|
1006 |
+
"loss": 0.4875,
|
1007 |
+
"step": 1380
|
1008 |
+
},
|
1009 |
+
{
|
1010 |
+
"epoch": 19.99,
|
1011 |
+
"eval_loss": 0.5481781959533691,
|
1012 |
+
"eval_runtime": 62.6832,
|
1013 |
+
"eval_samples_per_second": 8.95,
|
1014 |
+
"eval_steps_per_second": 2.249,
|
1015 |
+
"eval_wer": 0.17182160314919756,
|
1016 |
+
"step": 1380
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 20.14,
|
1020 |
+
"learning_rate": 0.00018465671641791043,
|
1021 |
+
"loss": 0.1949,
|
1022 |
+
"step": 1390
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 20.29,
|
1026 |
+
"learning_rate": 0.00018376119402985073,
|
1027 |
+
"loss": 0.4376,
|
1028 |
+
"step": 1400
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 20.43,
|
1032 |
+
"learning_rate": 0.00018286567164179106,
|
1033 |
+
"loss": 0.2543,
|
1034 |
+
"step": 1410
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 20.57,
|
1038 |
+
"learning_rate": 0.00018197014925373133,
|
1039 |
+
"loss": 0.4001,
|
1040 |
+
"step": 1420
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 20.72,
|
1044 |
+
"learning_rate": 0.00018107462686567163,
|
1045 |
+
"loss": 0.3915,
|
1046 |
+
"step": 1430
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 20.86,
|
1050 |
+
"learning_rate": 0.0001801791044776119,
|
1051 |
+
"loss": 0.1971,
|
1052 |
+
"step": 1440
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 20.99,
|
1056 |
+
"eval_loss": 0.45662185549736023,
|
1057 |
+
"eval_runtime": 63.0963,
|
1058 |
+
"eval_samples_per_second": 8.891,
|
1059 |
+
"eval_steps_per_second": 2.235,
|
1060 |
+
"eval_wer": 0.14487173941255352,
|
1061 |
+
"step": 1449
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 21.01,
|
1065 |
+
"learning_rate": 0.00017928358208955223,
|
1066 |
+
"loss": 0.6252,
|
1067 |
+
"step": 1450
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 21.16,
|
1071 |
+
"learning_rate": 0.00017838805970149253,
|
1072 |
+
"loss": 0.1249,
|
1073 |
+
"step": 1460
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 21.3,
|
1077 |
+
"learning_rate": 0.0001774925373134328,
|
1078 |
+
"loss": 0.3906,
|
1079 |
+
"step": 1470
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 21.44,
|
1083 |
+
"learning_rate": 0.0001765970149253731,
|
1084 |
+
"loss": 0.3267,
|
1085 |
+
"step": 1480
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 21.59,
|
1089 |
+
"learning_rate": 0.00017570149253731343,
|
1090 |
+
"loss": 0.3302,
|
1091 |
+
"step": 1490
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 21.73,
|
1095 |
+
"learning_rate": 0.00017480597014925373,
|
1096 |
+
"loss": 0.4891,
|
1097 |
+
"step": 1500
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 21.87,
|
1101 |
+
"learning_rate": 0.000173910447761194,
|
1102 |
+
"loss": 0.1286,
|
1103 |
+
"step": 1510
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 21.99,
|
1107 |
+
"eval_loss": 0.45152103900909424,
|
1108 |
+
"eval_runtime": 62.9105,
|
1109 |
+
"eval_samples_per_second": 8.917,
|
1110 |
+
"eval_steps_per_second": 2.241,
|
1111 |
+
"eval_wer": 0.1478132975732145,
|
1112 |
+
"step": 1518
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 22.03,
|
1116 |
+
"learning_rate": 0.00017301492537313433,
|
1117 |
+
"loss": 0.5264,
|
1118 |
+
"step": 1520
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 22.17,
|
1122 |
+
"learning_rate": 0.00017211940298507463,
|
1123 |
+
"loss": 0.1714,
|
1124 |
+
"step": 1530
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 22.32,
|
1128 |
+
"learning_rate": 0.0001712238805970149,
|
1129 |
+
"loss": 0.3947,
|
1130 |
+
"step": 1540
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 22.46,
|
1134 |
+
"learning_rate": 0.0001703283582089552,
|
1135 |
+
"loss": 0.3371,
|
1136 |
+
"step": 1550
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 22.6,
|
1140 |
+
"learning_rate": 0.00016943283582089553,
|
1141 |
+
"loss": 0.2423,
|
1142 |
+
"step": 1560
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 22.75,
|
1146 |
+
"learning_rate": 0.0001685373134328358,
|
1147 |
+
"loss": 0.5367,
|
1148 |
+
"step": 1570
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 22.89,
|
1152 |
+
"learning_rate": 0.0001676417910447761,
|
1153 |
+
"loss": 0.1472,
|
1154 |
+
"step": 1580
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 22.99,
|
1158 |
+
"eval_loss": 0.5059288144111633,
|
1159 |
+
"eval_runtime": 62.4349,
|
1160 |
+
"eval_samples_per_second": 8.985,
|
1161 |
+
"eval_steps_per_second": 2.258,
|
1162 |
+
"eval_wer": 0.1418436648354025,
|
1163 |
+
"step": 1587
|
1164 |
+
},
|
1165 |
+
{
|
1166 |
+
"epoch": 23.04,
|
1167 |
+
"learning_rate": 0.00016674626865671643,
|
1168 |
+
"loss": 0.4567,
|
1169 |
+
"step": 1590
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 23.19,
|
1173 |
+
"learning_rate": 0.0001658507462686567,
|
1174 |
+
"loss": 0.2266,
|
1175 |
+
"step": 1600
|
1176 |
+
},
|
1177 |
+
{
|
1178 |
+
"epoch": 23.33,
|
1179 |
+
"learning_rate": 0.000164955223880597,
|
1180 |
+
"loss": 0.3705,
|
1181 |
+
"step": 1610
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 23.47,
|
1185 |
+
"learning_rate": 0.0001640597014925373,
|
1186 |
+
"loss": 0.3832,
|
1187 |
+
"step": 1620
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 23.62,
|
1191 |
+
"learning_rate": 0.0001631641791044776,
|
1192 |
+
"loss": 0.2372,
|
1193 |
+
"step": 1630
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 23.76,
|
1197 |
+
"learning_rate": 0.0001622686567164179,
|
1198 |
+
"loss": 0.4775,
|
1199 |
+
"step": 1640
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 23.9,
|
1203 |
+
"learning_rate": 0.0001613731343283582,
|
1204 |
+
"loss": 0.1917,
|
1205 |
+
"step": 1650
|
1206 |
+
},
|
1207 |
+
{
|
1208 |
+
"epoch": 23.99,
|
1209 |
+
"eval_loss": 0.5583454966545105,
|
1210 |
+
"eval_runtime": 62.2591,
|
1211 |
+
"eval_samples_per_second": 9.011,
|
1212 |
+
"eval_steps_per_second": 2.265,
|
1213 |
+
"eval_wer": 0.1456936453692088,
|
1214 |
+
"step": 1656
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 24.06,
|
1218 |
+
"learning_rate": 0.00016047761194029847,
|
1219 |
+
"loss": 0.4329,
|
1220 |
+
"step": 1660
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 24.2,
|
1224 |
+
"learning_rate": 0.0001595820895522388,
|
1225 |
+
"loss": 0.2719,
|
1226 |
+
"step": 1670
|
1227 |
+
},
|
1228 |
+
{
|
1229 |
+
"epoch": 24.34,
|
1230 |
+
"learning_rate": 0.0001586865671641791,
|
1231 |
+
"loss": 0.3019,
|
1232 |
+
"step": 1680
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 24.49,
|
1236 |
+
"learning_rate": 0.00015779104477611937,
|
1237 |
+
"loss": 0.3512,
|
1238 |
+
"step": 1690
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 24.63,
|
1242 |
+
"learning_rate": 0.0001568955223880597,
|
1243 |
+
"loss": 0.157,
|
1244 |
+
"step": 1700
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 24.77,
|
1248 |
+
"learning_rate": 0.000156,
|
1249 |
+
"loss": 0.5036,
|
1250 |
+
"step": 1710
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 24.92,
|
1254 |
+
"learning_rate": 0.00015510447761194027,
|
1255 |
+
"loss": 0.2874,
|
1256 |
+
"step": 1720
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 24.99,
|
1260 |
+
"eval_loss": 0.5195438861846924,
|
1261 |
+
"eval_runtime": 62.6596,
|
1262 |
+
"eval_samples_per_second": 8.953,
|
1263 |
+
"eval_steps_per_second": 2.25,
|
1264 |
+
"eval_wer": 0.15027901544318034,
|
1265 |
+
"step": 1725
|
1266 |
+
},
|
1267 |
+
{
|
1268 |
+
"epoch": 25.07,
|
1269 |
+
"learning_rate": 0.00015420895522388057,
|
1270 |
+
"loss": 0.4331,
|
1271 |
+
"step": 1730
|
1272 |
+
},
|
1273 |
+
{
|
1274 |
+
"epoch": 25.22,
|
1275 |
+
"learning_rate": 0.0001533134328358209,
|
1276 |
+
"loss": 0.3291,
|
1277 |
+
"step": 1740
|
1278 |
+
},
|
1279 |
+
{
|
1280 |
+
"epoch": 25.36,
|
1281 |
+
"learning_rate": 0.00015241791044776117,
|
1282 |
+
"loss": 0.2418,
|
1283 |
+
"step": 1750
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 25.5,
|
1287 |
+
"learning_rate": 0.00015152238805970147,
|
1288 |
+
"loss": 0.3786,
|
1289 |
+
"step": 1760
|
1290 |
+
},
|
1291 |
+
{
|
1292 |
+
"epoch": 25.65,
|
1293 |
+
"learning_rate": 0.0001506268656716418,
|
1294 |
+
"loss": 0.1725,
|
1295 |
+
"step": 1770
|
1296 |
+
},
|
1297 |
+
{
|
1298 |
+
"epoch": 25.79,
|
1299 |
+
"learning_rate": 0.0001497313432835821,
|
1300 |
+
"loss": 0.3772,
|
1301 |
+
"step": 1780
|
1302 |
+
},
|
1303 |
+
{
|
1304 |
+
"epoch": 25.93,
|
1305 |
+
"learning_rate": 0.00014883582089552237,
|
1306 |
+
"loss": 0.2252,
|
1307 |
+
"step": 1790
|
1308 |
+
},
|
1309 |
+
{
|
1310 |
+
"epoch": 25.99,
|
1311 |
+
"eval_loss": 0.4408688247203827,
|
1312 |
+
"eval_runtime": 62.9542,
|
1313 |
+
"eval_samples_per_second": 8.911,
|
1314 |
+
"eval_steps_per_second": 2.24,
|
1315 |
+
"eval_wer": 0.15062508110914047,
|
1316 |
+
"step": 1794
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 26.09,
|
1320 |
+
"learning_rate": 0.00014794029850746267,
|
1321 |
+
"loss": 0.3815,
|
1322 |
+
"step": 1800
|
1323 |
+
},
|
1324 |
+
{
|
1325 |
+
"epoch": 26.23,
|
1326 |
+
"learning_rate": 0.00014704477611940297,
|
1327 |
+
"loss": 0.3281,
|
1328 |
+
"step": 1810
|
1329 |
+
},
|
1330 |
+
{
|
1331 |
+
"epoch": 26.37,
|
1332 |
+
"learning_rate": 0.00014614925373134327,
|
1333 |
+
"loss": 0.1803,
|
1334 |
+
"step": 1820
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 26.52,
|
1338 |
+
"learning_rate": 0.00014534328358208954,
|
1339 |
+
"loss": 0.4612,
|
1340 |
+
"step": 1830
|
1341 |
+
},
|
1342 |
+
{
|
1343 |
+
"epoch": 26.66,
|
1344 |
+
"learning_rate": 0.00014444776119402984,
|
1345 |
+
"loss": 0.1874,
|
1346 |
+
"step": 1840
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 26.8,
|
1350 |
+
"learning_rate": 0.00014355223880597014,
|
1351 |
+
"loss": 0.4485,
|
1352 |
+
"step": 1850
|
1353 |
+
},
|
1354 |
+
{
|
1355 |
+
"epoch": 26.95,
|
1356 |
+
"learning_rate": 0.00014265671641791044,
|
1357 |
+
"loss": 0.3142,
|
1358 |
+
"step": 1860
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 26.99,
|
1362 |
+
"eval_loss": 0.41795814037323,
|
1363 |
+
"eval_runtime": 62.551,
|
1364 |
+
"eval_samples_per_second": 8.969,
|
1365 |
+
"eval_steps_per_second": 2.254,
|
1366 |
+
"eval_wer": 0.143314443915733,
|
1367 |
+
"step": 1863
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 27.1,
|
1371 |
+
"learning_rate": 0.00014176119402985074,
|
1372 |
+
"loss": 0.2691,
|
1373 |
+
"step": 1870
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 27.24,
|
1377 |
+
"learning_rate": 0.00014086567164179104,
|
1378 |
+
"loss": 0.4776,
|
1379 |
+
"step": 1880
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 27.39,
|
1383 |
+
"learning_rate": 0.00013997014925373134,
|
1384 |
+
"loss": 0.1103,
|
1385 |
+
"step": 1890
|
1386 |
+
},
|
1387 |
+
{
|
1388 |
+
"epoch": 27.53,
|
1389 |
+
"learning_rate": 0.0001390746268656716,
|
1390 |
+
"loss": 0.4259,
|
1391 |
+
"step": 1900
|
1392 |
+
},
|
1393 |
+
{
|
1394 |
+
"epoch": 27.67,
|
1395 |
+
"learning_rate": 0.00013817910447761194,
|
1396 |
+
"loss": 0.223,
|
1397 |
+
"step": 1910
|
1398 |
+
},
|
1399 |
+
{
|
1400 |
+
"epoch": 27.82,
|
1401 |
+
"learning_rate": 0.0001372835820895522,
|
1402 |
+
"loss": 0.297,
|
1403 |
+
"step": 1920
|
1404 |
+
},
|
1405 |
+
{
|
1406 |
+
"epoch": 27.96,
|
1407 |
+
"learning_rate": 0.0001363880597014925,
|
1408 |
+
"loss": 0.385,
|
1409 |
+
"step": 1930
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 27.99,
|
1413 |
+
"eval_loss": 0.470846951007843,
|
1414 |
+
"eval_runtime": 62.9775,
|
1415 |
+
"eval_samples_per_second": 8.908,
|
1416 |
+
"eval_steps_per_second": 2.239,
|
1417 |
+
"eval_wer": 0.13673919626249081,
|
1418 |
+
"step": 1932
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 28.11,
|
1422 |
+
"learning_rate": 0.00013549253731343284,
|
1423 |
+
"loss": 0.2514,
|
1424 |
+
"step": 1940
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 28.26,
|
1428 |
+
"learning_rate": 0.0001345970149253731,
|
1429 |
+
"loss": 0.4299,
|
1430 |
+
"step": 1950
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 28.4,
|
1434 |
+
"learning_rate": 0.00013370149253731344,
|
1435 |
+
"loss": 0.1096,
|
1436 |
+
"step": 1960
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 28.54,
|
1440 |
+
"learning_rate": 0.0001328059701492537,
|
1441 |
+
"loss": 0.3887,
|
1442 |
+
"step": 1970
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 28.69,
|
1446 |
+
"learning_rate": 0.000131910447761194,
|
1447 |
+
"loss": 0.2654,
|
1448 |
+
"step": 1980
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 28.83,
|
1452 |
+
"learning_rate": 0.0001310149253731343,
|
1453 |
+
"loss": 0.3682,
|
1454 |
+
"step": 1990
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 28.97,
|
1458 |
+
"learning_rate": 0.0001301194029850746,
|
1459 |
+
"loss": 0.4296,
|
1460 |
+
"step": 2000
|
1461 |
+
},
|
1462 |
+
{
|
1463 |
+
"epoch": 28.99,
|
1464 |
+
"eval_loss": 0.4739513695240021,
|
1465 |
+
"eval_runtime": 63.1343,
|
1466 |
+
"eval_samples_per_second": 8.886,
|
1467 |
+
"eval_steps_per_second": 2.233,
|
1468 |
+
"eval_wer": 0.15058182290089545,
|
1469 |
+
"step": 2001
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 29.13,
|
1473 |
+
"learning_rate": 0.0001292238805970149,
|
1474 |
+
"loss": 0.17,
|
1475 |
+
"step": 2010
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 29.27,
|
1479 |
+
"learning_rate": 0.0001283283582089552,
|
1480 |
+
"loss": 0.4158,
|
1481 |
+
"step": 2020
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 29.42,
|
1485 |
+
"learning_rate": 0.0001274328358208955,
|
1486 |
+
"loss": 0.1852,
|
1487 |
+
"step": 2030
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 29.56,
|
1491 |
+
"learning_rate": 0.0001265373134328358,
|
1492 |
+
"loss": 0.3744,
|
1493 |
+
"step": 2040
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 29.7,
|
1497 |
+
"learning_rate": 0.0001256417910447761,
|
1498 |
+
"loss": 0.2737,
|
1499 |
+
"step": 2050
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 29.85,
|
1503 |
+
"learning_rate": 0.0001247462686567164,
|
1504 |
+
"loss": 0.2033,
|
1505 |
+
"step": 2060
|
1506 |
+
},
|
1507 |
+
{
|
1508 |
+
"epoch": 29.99,
|
1509 |
+
"learning_rate": 0.0001238507462686567,
|
1510 |
+
"loss": 0.4404,
|
1511 |
+
"step": 2070
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 29.99,
|
1515 |
+
"eval_loss": 0.46518003940582275,
|
1516 |
+
"eval_runtime": 63.0114,
|
1517 |
+
"eval_samples_per_second": 8.903,
|
1518 |
+
"eval_steps_per_second": 2.238,
|
1519 |
+
"eval_wer": 0.16455422416403512,
|
1520 |
+
"step": 2070
|
1521 |
+
},
|
1522 |
+
{
|
1523 |
+
"epoch": 30.14,
|
1524 |
+
"learning_rate": 0.000122955223880597,
|
1525 |
+
"loss": 0.1586,
|
1526 |
+
"step": 2080
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 30.29,
|
1530 |
+
"learning_rate": 0.00012205970149253731,
|
1531 |
+
"loss": 0.3666,
|
1532 |
+
"step": 2090
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 30.43,
|
1536 |
+
"learning_rate": 0.0001211641791044776,
|
1537 |
+
"loss": 0.2325,
|
1538 |
+
"step": 2100
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 30.57,
|
1542 |
+
"learning_rate": 0.0001202686567164179,
|
1543 |
+
"loss": 0.2943,
|
1544 |
+
"step": 2110
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 30.72,
|
1548 |
+
"learning_rate": 0.00011937313432835821,
|
1549 |
+
"loss": 0.3407,
|
1550 |
+
"step": 2120
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 30.86,
|
1554 |
+
"learning_rate": 0.0001184776119402985,
|
1555 |
+
"loss": 0.2466,
|
1556 |
+
"step": 2130
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 30.99,
|
1560 |
+
"eval_loss": 0.5013484954833984,
|
1561 |
+
"eval_runtime": 64.0675,
|
1562 |
+
"eval_samples_per_second": 8.756,
|
1563 |
+
"eval_steps_per_second": 2.201,
|
1564 |
+
"eval_wer": 0.1528312497296362,
|
1565 |
+
"step": 2139
|
1566 |
+
},
|
1567 |
+
{
|
1568 |
+
"epoch": 31.01,
|
1569 |
+
"learning_rate": 0.00011758208955223881,
|
1570 |
+
"loss": 0.4194,
|
1571 |
+
"step": 2140
|
1572 |
+
},
|
1573 |
+
{
|
1574 |
+
"epoch": 31.16,
|
1575 |
+
"learning_rate": 0.0001166865671641791,
|
1576 |
+
"loss": 0.1414,
|
1577 |
+
"step": 2150
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 31.3,
|
1581 |
+
"learning_rate": 0.00011579104477611939,
|
1582 |
+
"loss": 0.3346,
|
1583 |
+
"step": 2160
|
1584 |
+
},
|
1585 |
+
{
|
1586 |
+
"epoch": 31.44,
|
1587 |
+
"learning_rate": 0.00011489552238805968,
|
1588 |
+
"loss": 0.2872,
|
1589 |
+
"step": 2170
|
1590 |
+
},
|
1591 |
+
{
|
1592 |
+
"epoch": 31.59,
|
1593 |
+
"learning_rate": 0.00011399999999999999,
|
1594 |
+
"loss": 0.3151,
|
1595 |
+
"step": 2180
|
1596 |
+
},
|
1597 |
+
{
|
1598 |
+
"epoch": 31.73,
|
1599 |
+
"learning_rate": 0.00011310447761194028,
|
1600 |
+
"loss": 0.3952,
|
1601 |
+
"step": 2190
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 31.87,
|
1605 |
+
"learning_rate": 0.00011220895522388059,
|
1606 |
+
"loss": 0.1017,
|
1607 |
+
"step": 2200
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 31.99,
|
1611 |
+
"eval_loss": 0.45783624053001404,
|
1612 |
+
"eval_runtime": 64.1463,
|
1613 |
+
"eval_samples_per_second": 8.746,
|
1614 |
+
"eval_steps_per_second": 2.198,
|
1615 |
+
"eval_wer": 0.155210451183112,
|
1616 |
+
"step": 2208
|
1617 |
+
},
|
1618 |
+
{
|
1619 |
+
"epoch": 32.03,
|
1620 |
+
"learning_rate": 0.00011131343283582089,
|
1621 |
+
"loss": 0.4051,
|
1622 |
+
"step": 2210
|
1623 |
+
},
|
1624 |
+
{
|
1625 |
+
"epoch": 32.17,
|
1626 |
+
"learning_rate": 0.00011041791044776118,
|
1627 |
+
"loss": 0.1706,
|
1628 |
+
"step": 2220
|
1629 |
+
},
|
1630 |
+
{
|
1631 |
+
"epoch": 32.32,
|
1632 |
+
"learning_rate": 0.00010952238805970149,
|
1633 |
+
"loss": 0.3782,
|
1634 |
+
"step": 2230
|
1635 |
+
},
|
1636 |
+
{
|
1637 |
+
"epoch": 32.46,
|
1638 |
+
"learning_rate": 0.00010862686567164178,
|
1639 |
+
"loss": 0.2823,
|
1640 |
+
"step": 2240
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 32.6,
|
1644 |
+
"learning_rate": 0.00010773134328358208,
|
1645 |
+
"loss": 0.2133,
|
1646 |
+
"step": 2250
|
1647 |
+
},
|
1648 |
+
{
|
1649 |
+
"epoch": 32.75,
|
1650 |
+
"learning_rate": 0.00010683582089552238,
|
1651 |
+
"loss": 0.3748,
|
1652 |
+
"step": 2260
|
1653 |
+
},
|
1654 |
+
{
|
1655 |
+
"epoch": 32.89,
|
1656 |
+
"learning_rate": 0.00010594029850746268,
|
1657 |
+
"loss": 0.1383,
|
1658 |
+
"step": 2270
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 32.99,
|
1662 |
+
"eval_loss": 0.5026096105575562,
|
1663 |
+
"eval_runtime": 64.8383,
|
1664 |
+
"eval_samples_per_second": 8.652,
|
1665 |
+
"eval_steps_per_second": 2.175,
|
1666 |
+
"eval_wer": 0.14188692304364753,
|
1667 |
+
"step": 2277
|
1668 |
+
},
|
1669 |
+
{
|
1670 |
+
"epoch": 33.04,
|
1671 |
+
"learning_rate": 0.00010504477611940296,
|
1672 |
+
"loss": 0.4434,
|
1673 |
+
"step": 2280
|
1674 |
+
},
|
1675 |
+
{
|
1676 |
+
"epoch": 33.19,
|
1677 |
+
"learning_rate": 0.00010414925373134328,
|
1678 |
+
"loss": 0.1895,
|
1679 |
+
"step": 2290
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 33.33,
|
1683 |
+
"learning_rate": 0.00010325373134328358,
|
1684 |
+
"loss": 0.2909,
|
1685 |
+
"step": 2300
|
1686 |
+
},
|
1687 |
+
{
|
1688 |
+
"epoch": 33.47,
|
1689 |
+
"learning_rate": 0.00010235820895522386,
|
1690 |
+
"loss": 0.3125,
|
1691 |
+
"step": 2310
|
1692 |
+
},
|
1693 |
+
{
|
1694 |
+
"epoch": 33.62,
|
1695 |
+
"learning_rate": 0.00010146268656716418,
|
1696 |
+
"loss": 0.1565,
|
1697 |
+
"step": 2320
|
1698 |
+
},
|
1699 |
+
{
|
1700 |
+
"epoch": 33.76,
|
1701 |
+
"learning_rate": 0.00010056716417910446,
|
1702 |
+
"loss": 0.5154,
|
1703 |
+
"step": 2330
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 33.9,
|
1707 |
+
"learning_rate": 9.967164179104478e-05,
|
1708 |
+
"loss": 0.1719,
|
1709 |
+
"step": 2340
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 33.99,
|
1713 |
+
"eval_loss": 0.46511512994766235,
|
1714 |
+
"eval_runtime": 64.671,
|
1715 |
+
"eval_samples_per_second": 8.675,
|
1716 |
+
"eval_steps_per_second": 2.18,
|
1717 |
+
"eval_wer": 0.14422286628887832,
|
1718 |
+
"step": 2346
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 34.06,
|
1722 |
+
"learning_rate": 9.877611940298506e-05,
|
1723 |
+
"loss": 0.4138,
|
1724 |
+
"step": 2350
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 34.2,
|
1728 |
+
"learning_rate": 9.788059701492536e-05,
|
1729 |
+
"loss": 0.2165,
|
1730 |
+
"step": 2360
|
1731 |
+
},
|
1732 |
+
{
|
1733 |
+
"epoch": 34.34,
|
1734 |
+
"learning_rate": 9.698507462686566e-05,
|
1735 |
+
"loss": 0.2969,
|
1736 |
+
"step": 2370
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 34.49,
|
1740 |
+
"learning_rate": 9.608955223880596e-05,
|
1741 |
+
"loss": 0.4197,
|
1742 |
+
"step": 2380
|
1743 |
+
},
|
1744 |
+
{
|
1745 |
+
"epoch": 34.63,
|
1746 |
+
"learning_rate": 9.519402985074626e-05,
|
1747 |
+
"loss": 0.1103,
|
1748 |
+
"step": 2390
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 34.77,
|
1752 |
+
"learning_rate": 9.429850746268656e-05,
|
1753 |
+
"loss": 0.3708,
|
1754 |
+
"step": 2400
|
1755 |
+
},
|
1756 |
+
{
|
1757 |
+
"epoch": 34.92,
|
1758 |
+
"learning_rate": 9.340298507462686e-05,
|
1759 |
+
"loss": 0.1808,
|
1760 |
+
"step": 2410
|
1761 |
+
},
|
1762 |
+
{
|
1763 |
+
"epoch": 34.99,
|
1764 |
+
"eval_loss": 0.4499177634716034,
|
1765 |
+
"eval_runtime": 64.5038,
|
1766 |
+
"eval_samples_per_second": 8.697,
|
1767 |
+
"eval_steps_per_second": 2.186,
|
1768 |
+
"eval_wer": 0.14123804991997232,
|
1769 |
+
"step": 2415
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 35.07,
|
1773 |
+
"learning_rate": 9.250746268656715e-05,
|
1774 |
+
"loss": 0.3266,
|
1775 |
+
"step": 2420
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 35.22,
|
1779 |
+
"learning_rate": 9.161194029850746e-05,
|
1780 |
+
"loss": 0.2852,
|
1781 |
+
"step": 2430
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 35.36,
|
1785 |
+
"learning_rate": 9.071641791044775e-05,
|
1786 |
+
"loss": 0.294,
|
1787 |
+
"step": 2440
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 35.5,
|
1791 |
+
"learning_rate": 8.982089552238805e-05,
|
1792 |
+
"loss": 0.4613,
|
1793 |
+
"step": 2450
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 35.65,
|
1797 |
+
"learning_rate": 8.892537313432835e-05,
|
1798 |
+
"loss": 0.0998,
|
1799 |
+
"step": 2460
|
1800 |
+
},
|
1801 |
+
{
|
1802 |
+
"epoch": 35.79,
|
1803 |
+
"learning_rate": 8.802985074626865e-05,
|
1804 |
+
"loss": 0.3487,
|
1805 |
+
"step": 2470
|
1806 |
+
},
|
1807 |
+
{
|
1808 |
+
"epoch": 35.93,
|
1809 |
+
"learning_rate": 8.713432835820896e-05,
|
1810 |
+
"loss": 0.2429,
|
1811 |
+
"step": 2480
|
1812 |
+
},
|
1813 |
+
{
|
1814 |
+
"epoch": 35.99,
|
1815 |
+
"eval_loss": 0.4522528052330017,
|
1816 |
+
"eval_runtime": 64.4305,
|
1817 |
+
"eval_samples_per_second": 8.707,
|
1818 |
+
"eval_steps_per_second": 2.188,
|
1819 |
+
"eval_wer": 0.1471644244495393,
|
1820 |
+
"step": 2484
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 36.09,
|
1824 |
+
"learning_rate": 8.623880597014925e-05,
|
1825 |
+
"loss": 0.3206,
|
1826 |
+
"step": 2490
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 36.23,
|
1830 |
+
"learning_rate": 8.534328358208955e-05,
|
1831 |
+
"loss": 0.2824,
|
1832 |
+
"step": 2500
|
1833 |
+
},
|
1834 |
+
{
|
1835 |
+
"epoch": 36.37,
|
1836 |
+
"learning_rate": 8.444776119402985e-05,
|
1837 |
+
"loss": 0.156,
|
1838 |
+
"step": 2510
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 36.52,
|
1842 |
+
"learning_rate": 8.355223880597015e-05,
|
1843 |
+
"loss": 0.4632,
|
1844 |
+
"step": 2520
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 36.66,
|
1848 |
+
"learning_rate": 8.265671641791043e-05,
|
1849 |
+
"loss": 0.2011,
|
1850 |
+
"step": 2530
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 36.8,
|
1854 |
+
"learning_rate": 8.176119402985074e-05,
|
1855 |
+
"loss": 0.3352,
|
1856 |
+
"step": 2540
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 36.95,
|
1860 |
+
"learning_rate": 8.086567164179103e-05,
|
1861 |
+
"loss": 0.2651,
|
1862 |
+
"step": 2550
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 36.99,
|
1866 |
+
"eval_loss": 0.454355388879776,
|
1867 |
+
"eval_runtime": 63.0194,
|
1868 |
+
"eval_samples_per_second": 8.902,
|
1869 |
+
"eval_steps_per_second": 2.237,
|
1870 |
+
"eval_wer": 0.1397240126313968,
|
1871 |
+
"step": 2553
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 37.1,
|
1875 |
+
"learning_rate": 7.997014925373133e-05,
|
1876 |
+
"loss": 0.2746,
|
1877 |
+
"step": 2560
|
1878 |
+
},
|
1879 |
+
{
|
1880 |
+
"epoch": 37.24,
|
1881 |
+
"learning_rate": 7.907462686567164e-05,
|
1882 |
+
"loss": 0.4286,
|
1883 |
+
"step": 2570
|
1884 |
+
},
|
1885 |
+
{
|
1886 |
+
"epoch": 37.39,
|
1887 |
+
"learning_rate": 7.817910447761193e-05,
|
1888 |
+
"loss": 0.1063,
|
1889 |
+
"step": 2580
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 37.53,
|
1893 |
+
"learning_rate": 7.728358208955224e-05,
|
1894 |
+
"loss": 0.4127,
|
1895 |
+
"step": 2590
|
1896 |
+
},
|
1897 |
+
{
|
1898 |
+
"epoch": 37.67,
|
1899 |
+
"learning_rate": 7.638805970149253e-05,
|
1900 |
+
"loss": 0.1612,
|
1901 |
+
"step": 2600
|
1902 |
+
},
|
1903 |
+
{
|
1904 |
+
"epoch": 37.82,
|
1905 |
+
"learning_rate": 7.549253731343283e-05,
|
1906 |
+
"loss": 0.3096,
|
1907 |
+
"step": 2610
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 37.96,
|
1911 |
+
"learning_rate": 7.459701492537313e-05,
|
1912 |
+
"loss": 0.2748,
|
1913 |
+
"step": 2620
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 37.99,
|
1917 |
+
"eval_loss": 0.41805797815322876,
|
1918 |
+
"eval_runtime": 63.0354,
|
1919 |
+
"eval_samples_per_second": 8.9,
|
1920 |
+
"eval_steps_per_second": 2.237,
|
1921 |
+
"eval_wer": 0.1385560410087814,
|
1922 |
+
"step": 2622
|
1923 |
+
},
|
1924 |
+
{
|
1925 |
+
"epoch": 38.11,
|
1926 |
+
"learning_rate": 7.370149253731343e-05,
|
1927 |
+
"loss": 0.2088,
|
1928 |
+
"step": 2630
|
1929 |
+
},
|
1930 |
+
{
|
1931 |
+
"epoch": 38.26,
|
1932 |
+
"learning_rate": 7.280597014925373e-05,
|
1933 |
+
"loss": 0.344,
|
1934 |
+
"step": 2640
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 38.4,
|
1938 |
+
"learning_rate": 7.191044776119403e-05,
|
1939 |
+
"loss": 0.1293,
|
1940 |
+
"step": 2650
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 38.54,
|
1944 |
+
"learning_rate": 7.101492537313432e-05,
|
1945 |
+
"loss": 0.2947,
|
1946 |
+
"step": 2660
|
1947 |
+
},
|
1948 |
+
{
|
1949 |
+
"epoch": 38.69,
|
1950 |
+
"learning_rate": 7.011940298507462e-05,
|
1951 |
+
"loss": 0.2007,
|
1952 |
+
"step": 2670
|
1953 |
+
},
|
1954 |
+
{
|
1955 |
+
"epoch": 38.83,
|
1956 |
+
"learning_rate": 6.922388059701491e-05,
|
1957 |
+
"loss": 0.2795,
|
1958 |
+
"step": 2680
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 38.97,
|
1962 |
+
"learning_rate": 6.832835820895521e-05,
|
1963 |
+
"loss": 0.4171,
|
1964 |
+
"step": 2690
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 38.99,
|
1968 |
+
"eval_loss": 0.43853336572647095,
|
1969 |
+
"eval_runtime": 63.2599,
|
1970 |
+
"eval_samples_per_second": 8.868,
|
1971 |
+
"eval_steps_per_second": 2.229,
|
1972 |
+
"eval_wer": 0.1334083142276247,
|
1973 |
+
"step": 2691
|
1974 |
+
},
|
1975 |
+
{
|
1976 |
+
"epoch": 39.13,
|
1977 |
+
"learning_rate": 6.743283582089551e-05,
|
1978 |
+
"loss": 0.1256,
|
1979 |
+
"step": 2700
|
1980 |
+
},
|
1981 |
+
{
|
1982 |
+
"epoch": 39.27,
|
1983 |
+
"learning_rate": 6.653731343283581e-05,
|
1984 |
+
"loss": 0.383,
|
1985 |
+
"step": 2710
|
1986 |
+
},
|
1987 |
+
{
|
1988 |
+
"epoch": 39.42,
|
1989 |
+
"learning_rate": 6.564179104477611e-05,
|
1990 |
+
"loss": 0.1139,
|
1991 |
+
"step": 2720
|
1992 |
+
},
|
1993 |
+
{
|
1994 |
+
"epoch": 39.56,
|
1995 |
+
"learning_rate": 6.474626865671641e-05,
|
1996 |
+
"loss": 0.2979,
|
1997 |
+
"step": 2730
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 39.7,
|
2001 |
+
"learning_rate": 6.385074626865671e-05,
|
2002 |
+
"loss": 0.2812,
|
2003 |
+
"step": 2740
|
2004 |
+
},
|
2005 |
+
{
|
2006 |
+
"epoch": 39.85,
|
2007 |
+
"learning_rate": 6.295522388059701e-05,
|
2008 |
+
"loss": 0.251,
|
2009 |
+
"step": 2750
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 39.99,
|
2013 |
+
"learning_rate": 6.20597014925373e-05,
|
2014 |
+
"loss": 0.4119,
|
2015 |
+
"step": 2760
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 39.99,
|
2019 |
+
"eval_loss": 0.45682770013809204,
|
2020 |
+
"eval_runtime": 63.015,
|
2021 |
+
"eval_samples_per_second": 8.903,
|
2022 |
+
"eval_steps_per_second": 2.238,
|
2023 |
+
"eval_wer": 0.15040879006791538,
|
2024 |
+
"step": 2760
|
2025 |
+
},
|
2026 |
+
{
|
2027 |
+
"epoch": 40.14,
|
2028 |
+
"learning_rate": 6.11641791044776e-05,
|
2029 |
+
"loss": 0.1252,
|
2030 |
+
"step": 2770
|
2031 |
+
},
|
2032 |
+
{
|
2033 |
+
"epoch": 40.29,
|
2034 |
+
"learning_rate": 6.02686567164179e-05,
|
2035 |
+
"loss": 0.3647,
|
2036 |
+
"step": 2780
|
2037 |
+
},
|
2038 |
+
{
|
2039 |
+
"epoch": 40.43,
|
2040 |
+
"learning_rate": 5.9373134328358206e-05,
|
2041 |
+
"loss": 0.2004,
|
2042 |
+
"step": 2790
|
2043 |
+
},
|
2044 |
+
{
|
2045 |
+
"epoch": 40.57,
|
2046 |
+
"learning_rate": 5.8477611940298506e-05,
|
2047 |
+
"loss": 0.2465,
|
2048 |
+
"step": 2800
|
2049 |
+
},
|
2050 |
+
{
|
2051 |
+
"epoch": 40.72,
|
2052 |
+
"learning_rate": 5.7582089552238806e-05,
|
2053 |
+
"loss": 0.3278,
|
2054 |
+
"step": 2810
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 40.86,
|
2058 |
+
"learning_rate": 5.66865671641791e-05,
|
2059 |
+
"loss": 0.1453,
|
2060 |
+
"step": 2820
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 40.99,
|
2064 |
+
"eval_loss": 0.44247445464134216,
|
2065 |
+
"eval_runtime": 63.0109,
|
2066 |
+
"eval_samples_per_second": 8.903,
|
2067 |
+
"eval_steps_per_second": 2.238,
|
2068 |
+
"eval_wer": 0.14309815287450794,
|
2069 |
+
"step": 2829
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 41.01,
|
2073 |
+
"learning_rate": 5.57910447761194e-05,
|
2074 |
+
"loss": 0.4307,
|
2075 |
+
"step": 2830
|
2076 |
+
},
|
2077 |
+
{
|
2078 |
+
"epoch": 41.16,
|
2079 |
+
"learning_rate": 5.48955223880597e-05,
|
2080 |
+
"loss": 0.1089,
|
2081 |
+
"step": 2840
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 41.3,
|
2085 |
+
"learning_rate": 5.399999999999999e-05,
|
2086 |
+
"loss": 0.3548,
|
2087 |
+
"step": 2850
|
2088 |
+
},
|
2089 |
+
{
|
2090 |
+
"epoch": 41.44,
|
2091 |
+
"learning_rate": 5.310447761194029e-05,
|
2092 |
+
"loss": 0.197,
|
2093 |
+
"step": 2860
|
2094 |
+
},
|
2095 |
+
{
|
2096 |
+
"epoch": 41.59,
|
2097 |
+
"learning_rate": 5.220895522388059e-05,
|
2098 |
+
"loss": 0.2719,
|
2099 |
+
"step": 2870
|
2100 |
+
},
|
2101 |
+
{
|
2102 |
+
"epoch": 41.73,
|
2103 |
+
"learning_rate": 5.13134328358209e-05,
|
2104 |
+
"loss": 0.4077,
|
2105 |
+
"step": 2880
|
2106 |
+
},
|
2107 |
+
{
|
2108 |
+
"epoch": 41.87,
|
2109 |
+
"learning_rate": 5.041791044776119e-05,
|
2110 |
+
"loss": 0.105,
|
2111 |
+
"step": 2890
|
2112 |
+
},
|
2113 |
+
{
|
2114 |
+
"epoch": 41.99,
|
2115 |
+
"eval_loss": 0.43674367666244507,
|
2116 |
+
"eval_runtime": 64.6292,
|
2117 |
+
"eval_samples_per_second": 8.68,
|
2118 |
+
"eval_steps_per_second": 2.182,
|
2119 |
+
"eval_wer": 0.13526841718216032,
|
2120 |
+
"step": 2898
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 42.03,
|
2124 |
+
"learning_rate": 4.952238805970149e-05,
|
2125 |
+
"loss": 0.3982,
|
2126 |
+
"step": 2900
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 42.17,
|
2130 |
+
"learning_rate": 4.862686567164179e-05,
|
2131 |
+
"loss": 0.1811,
|
2132 |
+
"step": 2910
|
2133 |
+
},
|
2134 |
+
{
|
2135 |
+
"epoch": 42.32,
|
2136 |
+
"learning_rate": 4.773134328358208e-05,
|
2137 |
+
"loss": 0.3297,
|
2138 |
+
"step": 2920
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 42.46,
|
2142 |
+
"learning_rate": 4.683582089552238e-05,
|
2143 |
+
"loss": 0.2095,
|
2144 |
+
"step": 2930
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 42.6,
|
2148 |
+
"learning_rate": 4.594029850746268e-05,
|
2149 |
+
"loss": 0.1968,
|
2150 |
+
"step": 2940
|
2151 |
+
},
|
2152 |
+
{
|
2153 |
+
"epoch": 42.75,
|
2154 |
+
"learning_rate": 4.5044776119402976e-05,
|
2155 |
+
"loss": 0.3637,
|
2156 |
+
"step": 2950
|
2157 |
+
},
|
2158 |
+
{
|
2159 |
+
"epoch": 42.89,
|
2160 |
+
"learning_rate": 4.4149253731343275e-05,
|
2161 |
+
"loss": 0.1205,
|
2162 |
+
"step": 2960
|
2163 |
+
},
|
2164 |
+
{
|
2165 |
+
"epoch": 42.99,
|
2166 |
+
"eval_loss": 0.44175830483436584,
|
2167 |
+
"eval_runtime": 64.641,
|
2168 |
+
"eval_samples_per_second": 8.679,
|
2169 |
+
"eval_steps_per_second": 2.181,
|
2170 |
+
"eval_wer": 0.1340139291430549,
|
2171 |
+
"step": 2967
|
2172 |
+
},
|
2173 |
+
{
|
2174 |
+
"epoch": 43.04,
|
2175 |
+
"learning_rate": 4.325373134328358e-05,
|
2176 |
+
"loss": 0.3979,
|
2177 |
+
"step": 2970
|
2178 |
+
},
|
2179 |
+
{
|
2180 |
+
"epoch": 43.19,
|
2181 |
+
"learning_rate": 4.235820895522388e-05,
|
2182 |
+
"loss": 0.1674,
|
2183 |
+
"step": 2980
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 43.33,
|
2187 |
+
"learning_rate": 4.1462686567164175e-05,
|
2188 |
+
"loss": 0.3019,
|
2189 |
+
"step": 2990
|
2190 |
+
},
|
2191 |
+
{
|
2192 |
+
"epoch": 43.47,
|
2193 |
+
"learning_rate": 4.0567164179104475e-05,
|
2194 |
+
"loss": 0.2778,
|
2195 |
+
"step": 3000
|
2196 |
+
},
|
2197 |
+
{
|
2198 |
+
"epoch": 43.62,
|
2199 |
+
"learning_rate": 3.9671641791044774e-05,
|
2200 |
+
"loss": 0.1561,
|
2201 |
+
"step": 3010
|
2202 |
+
},
|
2203 |
+
{
|
2204 |
+
"epoch": 43.76,
|
2205 |
+
"learning_rate": 3.877611940298507e-05,
|
2206 |
+
"loss": 0.3686,
|
2207 |
+
"step": 3020
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 43.9,
|
2211 |
+
"learning_rate": 3.788059701492537e-05,
|
2212 |
+
"loss": 0.2039,
|
2213 |
+
"step": 3030
|
2214 |
+
},
|
2215 |
+
{
|
2216 |
+
"epoch": 43.99,
|
2217 |
+
"eval_loss": 0.4585602283477783,
|
2218 |
+
"eval_runtime": 62.823,
|
2219 |
+
"eval_samples_per_second": 8.93,
|
2220 |
+
"eval_steps_per_second": 2.244,
|
2221 |
+
"eval_wer": 0.13786390967686118,
|
2222 |
+
"step": 3036
|
2223 |
+
},
|
2224 |
+
{
|
2225 |
+
"epoch": 44.06,
|
2226 |
+
"learning_rate": 3.698507462686567e-05,
|
2227 |
+
"loss": 0.3361,
|
2228 |
+
"step": 3040
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 44.2,
|
2232 |
+
"learning_rate": 3.608955223880597e-05,
|
2233 |
+
"loss": 0.185,
|
2234 |
+
"step": 3050
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 44.34,
|
2238 |
+
"learning_rate": 3.519402985074627e-05,
|
2239 |
+
"loss": 0.2579,
|
2240 |
+
"step": 3060
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 44.49,
|
2244 |
+
"learning_rate": 3.429850746268656e-05,
|
2245 |
+
"loss": 0.3879,
|
2246 |
+
"step": 3070
|
2247 |
+
},
|
2248 |
+
{
|
2249 |
+
"epoch": 44.63,
|
2250 |
+
"learning_rate": 3.3402985074626866e-05,
|
2251 |
+
"loss": 0.1033,
|
2252 |
+
"step": 3080
|
2253 |
+
},
|
2254 |
+
{
|
2255 |
+
"epoch": 44.77,
|
2256 |
+
"learning_rate": 3.250746268656716e-05,
|
2257 |
+
"loss": 0.3663,
|
2258 |
+
"step": 3090
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 44.92,
|
2262 |
+
"learning_rate": 3.161194029850746e-05,
|
2263 |
+
"loss": 0.1773,
|
2264 |
+
"step": 3100
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 44.99,
|
2268 |
+
"eval_loss": 0.46862030029296875,
|
2269 |
+
"eval_runtime": 63.8696,
|
2270 |
+
"eval_samples_per_second": 8.784,
|
2271 |
+
"eval_steps_per_second": 2.208,
|
2272 |
+
"eval_wer": 0.1390751395077216,
|
2273 |
+
"step": 3105
|
2274 |
+
},
|
2275 |
+
{
|
2276 |
+
"epoch": 45.07,
|
2277 |
+
"learning_rate": 3.071641791044776e-05,
|
2278 |
+
"loss": 0.3077,
|
2279 |
+
"step": 3110
|
2280 |
+
},
|
2281 |
+
{
|
2282 |
+
"epoch": 45.22,
|
2283 |
+
"learning_rate": 2.9820895522388055e-05,
|
2284 |
+
"loss": 0.2927,
|
2285 |
+
"step": 3120
|
2286 |
+
},
|
2287 |
+
{
|
2288 |
+
"epoch": 45.36,
|
2289 |
+
"learning_rate": 2.892537313432836e-05,
|
2290 |
+
"loss": 0.2373,
|
2291 |
+
"step": 3130
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 45.5,
|
2295 |
+
"learning_rate": 2.8029850746268655e-05,
|
2296 |
+
"loss": 0.3392,
|
2297 |
+
"step": 3140
|
2298 |
+
},
|
2299 |
+
{
|
2300 |
+
"epoch": 45.65,
|
2301 |
+
"learning_rate": 2.713432835820895e-05,
|
2302 |
+
"loss": 0.0841,
|
2303 |
+
"step": 3150
|
2304 |
+
},
|
2305 |
+
{
|
2306 |
+
"epoch": 45.79,
|
2307 |
+
"learning_rate": 2.623880597014925e-05,
|
2308 |
+
"loss": 0.3028,
|
2309 |
+
"step": 3160
|
2310 |
+
},
|
2311 |
+
{
|
2312 |
+
"epoch": 45.93,
|
2313 |
+
"learning_rate": 2.534328358208955e-05,
|
2314 |
+
"loss": 0.2186,
|
2315 |
+
"step": 3170
|
2316 |
+
},
|
2317 |
+
{
|
2318 |
+
"epoch": 45.99,
|
2319 |
+
"eval_loss": 0.49753618240356445,
|
2320 |
+
"eval_runtime": 63.1583,
|
2321 |
+
"eval_samples_per_second": 8.882,
|
2322 |
+
"eval_steps_per_second": 2.232,
|
2323 |
+
"eval_wer": 0.14456893195483844,
|
2324 |
+
"step": 3174
|
2325 |
+
},
|
2326 |
+
{
|
2327 |
+
"epoch": 46.09,
|
2328 |
+
"learning_rate": 2.444776119402985e-05,
|
2329 |
+
"loss": 0.2526,
|
2330 |
+
"step": 3180
|
2331 |
+
},
|
2332 |
+
{
|
2333 |
+
"epoch": 46.23,
|
2334 |
+
"learning_rate": 2.3552238805970147e-05,
|
2335 |
+
"loss": 0.3354,
|
2336 |
+
"step": 3190
|
2337 |
+
},
|
2338 |
+
{
|
2339 |
+
"epoch": 46.37,
|
2340 |
+
"learning_rate": 2.2656716417910444e-05,
|
2341 |
+
"loss": 0.1822,
|
2342 |
+
"step": 3200
|
2343 |
+
},
|
2344 |
+
{
|
2345 |
+
"epoch": 46.52,
|
2346 |
+
"learning_rate": 2.1761194029850743e-05,
|
2347 |
+
"loss": 0.3111,
|
2348 |
+
"step": 3210
|
2349 |
+
},
|
2350 |
+
{
|
2351 |
+
"epoch": 46.66,
|
2352 |
+
"learning_rate": 2.0865671641791043e-05,
|
2353 |
+
"loss": 0.1602,
|
2354 |
+
"step": 3220
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 46.8,
|
2358 |
+
"learning_rate": 1.9970149253731343e-05,
|
2359 |
+
"loss": 0.3159,
|
2360 |
+
"step": 3230
|
2361 |
+
},
|
2362 |
+
{
|
2363 |
+
"epoch": 46.95,
|
2364 |
+
"learning_rate": 1.907462686567164e-05,
|
2365 |
+
"loss": 0.2358,
|
2366 |
+
"step": 3240
|
2367 |
+
},
|
2368 |
+
{
|
2369 |
+
"epoch": 46.99,
|
2370 |
+
"eval_loss": 0.4886210858821869,
|
2371 |
+
"eval_runtime": 63.1228,
|
2372 |
+
"eval_samples_per_second": 8.887,
|
2373 |
+
"eval_steps_per_second": 2.234,
|
2374 |
+
"eval_wer": 0.14482848120430852,
|
2375 |
+
"step": 3243
|
2376 |
+
},
|
2377 |
+
{
|
2378 |
+
"epoch": 47.1,
|
2379 |
+
"learning_rate": 1.817910447761194e-05,
|
2380 |
+
"loss": 0.2315,
|
2381 |
+
"step": 3250
|
2382 |
+
},
|
2383 |
+
{
|
2384 |
+
"epoch": 47.24,
|
2385 |
+
"learning_rate": 1.7283582089552236e-05,
|
2386 |
+
"loss": 0.4232,
|
2387 |
+
"step": 3260
|
2388 |
+
},
|
2389 |
+
{
|
2390 |
+
"epoch": 47.39,
|
2391 |
+
"learning_rate": 1.6388059701492535e-05,
|
2392 |
+
"loss": 0.0988,
|
2393 |
+
"step": 3270
|
2394 |
+
},
|
2395 |
+
{
|
2396 |
+
"epoch": 47.53,
|
2397 |
+
"learning_rate": 1.5492537313432835e-05,
|
2398 |
+
"loss": 0.4457,
|
2399 |
+
"step": 3280
|
2400 |
+
},
|
2401 |
+
{
|
2402 |
+
"epoch": 47.67,
|
2403 |
+
"learning_rate": 1.4597014925373132e-05,
|
2404 |
+
"loss": 0.1544,
|
2405 |
+
"step": 3290
|
2406 |
+
},
|
2407 |
+
{
|
2408 |
+
"epoch": 47.82,
|
2409 |
+
"learning_rate": 1.3701492537313431e-05,
|
2410 |
+
"loss": 0.2426,
|
2411 |
+
"step": 3300
|
2412 |
+
},
|
2413 |
+
{
|
2414 |
+
"epoch": 47.96,
|
2415 |
+
"learning_rate": 1.280597014925373e-05,
|
2416 |
+
"loss": 0.3525,
|
2417 |
+
"step": 3310
|
2418 |
+
},
|
2419 |
+
{
|
2420 |
+
"epoch": 47.99,
|
2421 |
+
"eval_loss": 0.47062501311302185,
|
2422 |
+
"eval_runtime": 62.9566,
|
2423 |
+
"eval_samples_per_second": 8.911,
|
2424 |
+
"eval_steps_per_second": 2.24,
|
2425 |
+
"eval_wer": 0.13981052904788685,
|
2426 |
+
"step": 3312
|
2427 |
+
},
|
2428 |
+
{
|
2429 |
+
"epoch": 48.11,
|
2430 |
+
"learning_rate": 1.191044776119403e-05,
|
2431 |
+
"loss": 0.255,
|
2432 |
+
"step": 3320
|
2433 |
+
},
|
2434 |
+
{
|
2435 |
+
"epoch": 48.26,
|
2436 |
+
"learning_rate": 1.1014925373134327e-05,
|
2437 |
+
"loss": 0.3479,
|
2438 |
+
"step": 3330
|
2439 |
+
},
|
2440 |
+
{
|
2441 |
+
"epoch": 48.4,
|
2442 |
+
"learning_rate": 1.0119402985074627e-05,
|
2443 |
+
"loss": 0.1079,
|
2444 |
+
"step": 3340
|
2445 |
+
},
|
2446 |
+
{
|
2447 |
+
"epoch": 48.54,
|
2448 |
+
"learning_rate": 9.223880597014925e-06,
|
2449 |
+
"loss": 0.3069,
|
2450 |
+
"step": 3350
|
2451 |
+
},
|
2452 |
+
{
|
2453 |
+
"epoch": 48.69,
|
2454 |
+
"learning_rate": 8.328358208955223e-06,
|
2455 |
+
"loss": 0.2212,
|
2456 |
+
"step": 3360
|
2457 |
+
},
|
2458 |
+
{
|
2459 |
+
"epoch": 48.83,
|
2460 |
+
"learning_rate": 7.4328358208955215e-06,
|
2461 |
+
"loss": 0.2339,
|
2462 |
+
"step": 3370
|
2463 |
+
},
|
2464 |
+
{
|
2465 |
+
"epoch": 48.97,
|
2466 |
+
"learning_rate": 6.5373134328358204e-06,
|
2467 |
+
"loss": 0.3713,
|
2468 |
+
"step": 3380
|
2469 |
+
},
|
2470 |
+
{
|
2471 |
+
"epoch": 48.99,
|
2472 |
+
"eval_loss": 0.47125667333602905,
|
2473 |
+
"eval_runtime": 62.5816,
|
2474 |
+
"eval_samples_per_second": 8.964,
|
2475 |
+
"eval_steps_per_second": 2.253,
|
2476 |
+
"eval_wer": 0.1387723320500065,
|
2477 |
+
"step": 3381
|
2478 |
+
},
|
2479 |
+
{
|
2480 |
+
"epoch": 49.13,
|
2481 |
+
"learning_rate": 5.641791044776119e-06,
|
2482 |
+
"loss": 0.1483,
|
2483 |
+
"step": 3390
|
2484 |
+
},
|
2485 |
+
{
|
2486 |
+
"epoch": 49.27,
|
2487 |
+
"learning_rate": 4.746268656716418e-06,
|
2488 |
+
"loss": 0.3856,
|
2489 |
+
"step": 3400
|
2490 |
+
},
|
2491 |
+
{
|
2492 |
+
"epoch": 49.42,
|
2493 |
+
"learning_rate": 3.8507462686567165e-06,
|
2494 |
+
"loss": 0.1105,
|
2495 |
+
"step": 3410
|
2496 |
+
},
|
2497 |
+
{
|
2498 |
+
"epoch": 49.56,
|
2499 |
+
"learning_rate": 2.955223880597015e-06,
|
2500 |
+
"loss": 0.2724,
|
2501 |
+
"step": 3420
|
2502 |
+
},
|
2503 |
+
{
|
2504 |
+
"epoch": 49.7,
|
2505 |
+
"learning_rate": 2.0597014925373135e-06,
|
2506 |
+
"loss": 0.2536,
|
2507 |
+
"step": 3430
|
2508 |
+
},
|
2509 |
+
{
|
2510 |
+
"epoch": 49.85,
|
2511 |
+
"learning_rate": 1.1641791044776118e-06,
|
2512 |
+
"loss": 0.2202,
|
2513 |
+
"step": 3440
|
2514 |
+
},
|
2515 |
+
{
|
2516 |
+
"epoch": 49.99,
|
2517 |
+
"learning_rate": 2.686567164179104e-07,
|
2518 |
+
"loss": 0.3543,
|
2519 |
+
"step": 3450
|
2520 |
+
},
|
2521 |
+
{
|
2522 |
+
"epoch": 49.99,
|
2523 |
+
"eval_loss": 0.47204339504241943,
|
2524 |
+
"eval_runtime": 63.0724,
|
2525 |
+
"eval_samples_per_second": 8.895,
|
2526 |
+
"eval_steps_per_second": 2.236,
|
2527 |
+
"eval_wer": 0.1388155902582515,
|
2528 |
+
"step": 3450
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 49.99,
|
2532 |
+
"step": 3450,
|
2533 |
+
"total_flos": 8.165293514893254e+18,
|
2534 |
+
"train_loss": 0.35037224591642185,
|
2535 |
+
"train_runtime": 11843.2621,
|
2536 |
+
"train_samples_per_second": 4.712,
|
2537 |
+
"train_steps_per_second": 0.291
|
2538 |
+
}
|
2539 |
+
],
|
2540 |
+
"max_steps": 3450,
|
2541 |
+
"num_train_epochs": 50,
|
2542 |
+
"total_flos": 8.165293514893254e+18,
|
2543 |
+
"trial_name": null,
|
2544 |
+
"trial_params": null
|
2545 |
+
}
|