End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 99.99,
|
3 |
+
"eval_loss": 0.6378720998764038,
|
4 |
+
"eval_runtime": 9.0914,
|
5 |
+
"eval_samples": 63,
|
6 |
+
"eval_samples_per_second": 6.93,
|
7 |
+
"eval_steps_per_second": 1.76,
|
8 |
+
"eval_wer": 0.37916490932096164,
|
9 |
+
"train_loss": 0.4800486422266279,
|
10 |
+
"train_runtime": 12025.3803,
|
11 |
+
"train_samples": 563,
|
12 |
+
"train_samples_per_second": 4.682,
|
13 |
+
"train_steps_per_second": 0.291
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 99.99,
|
3 |
+
"eval_loss": 0.6378720998764038,
|
4 |
+
"eval_runtime": 9.0914,
|
5 |
+
"eval_samples": 63,
|
6 |
+
"eval_samples_per_second": 6.93,
|
7 |
+
"eval_steps_per_second": 1.76,
|
8 |
+
"eval_wer": 0.37916490932096164
|
9 |
+
}
|
runs/Nov25_07-58-12_4b2aca464fe8/events.out.tfevents.1669375820.4b2aca464fe8.12485.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ab171b2008a775b049057dced237084109837a48fa372326d84a24e5ba1209eb
|
3 |
+
size 358
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 99.99,
|
3 |
+
"train_loss": 0.4800486422266279,
|
4 |
+
"train_runtime": 12025.3803,
|
5 |
+
"train_samples": 563,
|
6 |
+
"train_samples_per_second": 4.682,
|
7 |
+
"train_steps_per_second": 0.291
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,3025 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.6378720998764038,
|
3 |
+
"best_model_checkpoint": "ai-light-dance_drums_ft_pretrain_wav2vec2-base-new_onset-idmt-mdb-enst-2/checkpoint-700",
|
4 |
+
"epoch": 99.99290780141844,
|
5 |
+
"global_step": 3500,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.28,
|
12 |
+
"learning_rate": 7e-05,
|
13 |
+
"loss": 2.6822,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.57,
|
18 |
+
"learning_rate": 0.00016999999999999999,
|
19 |
+
"loss": 1.5659,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.85,
|
24 |
+
"learning_rate": 0.00026,
|
25 |
+
"loss": 1.4226,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.99,
|
30 |
+
"eval_loss": 2.0435445308685303,
|
31 |
+
"eval_runtime": 8.6793,
|
32 |
+
"eval_samples_per_second": 7.259,
|
33 |
+
"eval_steps_per_second": 1.843,
|
34 |
+
"eval_wer": 0.4154365246731337,
|
35 |
+
"step": 35
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 1.14,
|
39 |
+
"learning_rate": 0.00029948126801152735,
|
40 |
+
"loss": 1.0316,
|
41 |
+
"step": 40
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 1.43,
|
45 |
+
"learning_rate": 0.0002986167146974063,
|
46 |
+
"loss": 1.0479,
|
47 |
+
"step": 50
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 1.71,
|
51 |
+
"learning_rate": 0.0002977521613832853,
|
52 |
+
"loss": 1.0524,
|
53 |
+
"step": 60
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 1.99,
|
57 |
+
"learning_rate": 0.00029688760806916427,
|
58 |
+
"loss": 0.8744,
|
59 |
+
"step": 70
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.99,
|
63 |
+
"eval_loss": 1.71934175491333,
|
64 |
+
"eval_runtime": 9.0494,
|
65 |
+
"eval_samples_per_second": 6.962,
|
66 |
+
"eval_steps_per_second": 1.768,
|
67 |
+
"eval_wer": 0.43821172501054406,
|
68 |
+
"step": 70
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 2.28,
|
72 |
+
"learning_rate": 0.0002960230547550432,
|
73 |
+
"loss": 0.8641,
|
74 |
+
"step": 80
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 2.57,
|
78 |
+
"learning_rate": 0.00029515850144092216,
|
79 |
+
"loss": 0.8469,
|
80 |
+
"step": 90
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 2.85,
|
84 |
+
"learning_rate": 0.00029429394812680113,
|
85 |
+
"loss": 0.9474,
|
86 |
+
"step": 100
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 2.99,
|
90 |
+
"eval_loss": 1.7853245735168457,
|
91 |
+
"eval_runtime": 9.2275,
|
92 |
+
"eval_samples_per_second": 6.827,
|
93 |
+
"eval_steps_per_second": 1.734,
|
94 |
+
"eval_wer": 0.4373681990721215,
|
95 |
+
"step": 105
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 3.14,
|
99 |
+
"learning_rate": 0.00029342939481268005,
|
100 |
+
"loss": 0.9012,
|
101 |
+
"step": 110
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 3.43,
|
105 |
+
"learning_rate": 0.0002925648414985591,
|
106 |
+
"loss": 0.9915,
|
107 |
+
"step": 120
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 3.71,
|
111 |
+
"learning_rate": 0.000291700288184438,
|
112 |
+
"loss": 0.7743,
|
113 |
+
"step": 130
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 3.99,
|
117 |
+
"learning_rate": 0.00029083573487031697,
|
118 |
+
"loss": 0.8316,
|
119 |
+
"step": 140
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 3.99,
|
123 |
+
"eval_loss": 1.2827332019805908,
|
124 |
+
"eval_runtime": 9.447,
|
125 |
+
"eval_samples_per_second": 6.669,
|
126 |
+
"eval_steps_per_second": 1.694,
|
127 |
+
"eval_wer": 0.4306199915647406,
|
128 |
+
"step": 140
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 4.28,
|
132 |
+
"learning_rate": 0.00028997118155619595,
|
133 |
+
"loss": 0.7751,
|
134 |
+
"step": 150
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 4.57,
|
138 |
+
"learning_rate": 0.0002891066282420749,
|
139 |
+
"loss": 0.8104,
|
140 |
+
"step": 160
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 4.85,
|
144 |
+
"learning_rate": 0.00028824207492795384,
|
145 |
+
"loss": 0.8336,
|
146 |
+
"step": 170
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 4.99,
|
150 |
+
"eval_loss": 1.0676425695419312,
|
151 |
+
"eval_runtime": 8.8457,
|
152 |
+
"eval_samples_per_second": 7.122,
|
153 |
+
"eval_steps_per_second": 1.809,
|
154 |
+
"eval_wer": 0.40404892450442853,
|
155 |
+
"step": 175
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 5.14,
|
159 |
+
"learning_rate": 0.00028737752161383287,
|
160 |
+
"loss": 0.7711,
|
161 |
+
"step": 180
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 5.43,
|
165 |
+
"learning_rate": 0.0002865129682997118,
|
166 |
+
"loss": 0.6393,
|
167 |
+
"step": 190
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 5.71,
|
171 |
+
"learning_rate": 0.00028564841498559076,
|
172 |
+
"loss": 0.7828,
|
173 |
+
"step": 200
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 5.99,
|
177 |
+
"learning_rate": 0.00028478386167146973,
|
178 |
+
"loss": 0.7345,
|
179 |
+
"step": 210
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 5.99,
|
183 |
+
"eval_loss": 1.5363637208938599,
|
184 |
+
"eval_runtime": 8.9421,
|
185 |
+
"eval_samples_per_second": 7.045,
|
186 |
+
"eval_steps_per_second": 1.789,
|
187 |
+
"eval_wer": 0.4264023618726276,
|
188 |
+
"step": 210
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 6.28,
|
192 |
+
"learning_rate": 0.00028391930835734865,
|
193 |
+
"loss": 0.7525,
|
194 |
+
"step": 220
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 6.57,
|
198 |
+
"learning_rate": 0.0002830547550432276,
|
199 |
+
"loss": 0.6614,
|
200 |
+
"step": 230
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 6.85,
|
204 |
+
"learning_rate": 0.0002821902017291066,
|
205 |
+
"loss": 0.6666,
|
206 |
+
"step": 240
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 6.99,
|
210 |
+
"eval_loss": 1.428429365158081,
|
211 |
+
"eval_runtime": 8.8874,
|
212 |
+
"eval_samples_per_second": 7.089,
|
213 |
+
"eval_steps_per_second": 1.8,
|
214 |
+
"eval_wer": 0.4584563475326866,
|
215 |
+
"step": 245
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 7.14,
|
219 |
+
"learning_rate": 0.00028132564841498557,
|
220 |
+
"loss": 0.7812,
|
221 |
+
"step": 250
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 7.43,
|
225 |
+
"learning_rate": 0.00028046109510086454,
|
226 |
+
"loss": 0.6833,
|
227 |
+
"step": 260
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 7.71,
|
231 |
+
"learning_rate": 0.0002795965417867435,
|
232 |
+
"loss": 0.7328,
|
233 |
+
"step": 270
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 7.99,
|
237 |
+
"learning_rate": 0.00027873198847262244,
|
238 |
+
"loss": 0.6677,
|
239 |
+
"step": 280
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 7.99,
|
243 |
+
"eval_loss": 0.9474982023239136,
|
244 |
+
"eval_runtime": 8.8866,
|
245 |
+
"eval_samples_per_second": 7.089,
|
246 |
+
"eval_steps_per_second": 1.8,
|
247 |
+
"eval_wer": 0.40025305778152676,
|
248 |
+
"step": 280
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 8.28,
|
252 |
+
"learning_rate": 0.0002778674351585014,
|
253 |
+
"loss": 0.7165,
|
254 |
+
"step": 290
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 8.57,
|
258 |
+
"learning_rate": 0.0002770028818443804,
|
259 |
+
"loss": 0.667,
|
260 |
+
"step": 300
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 8.85,
|
264 |
+
"learning_rate": 0.00027613832853025936,
|
265 |
+
"loss": 0.6779,
|
266 |
+
"step": 310
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 8.99,
|
270 |
+
"eval_loss": 1.1171544790267944,
|
271 |
+
"eval_runtime": 9.1402,
|
272 |
+
"eval_samples_per_second": 6.893,
|
273 |
+
"eval_steps_per_second": 1.751,
|
274 |
+
"eval_wer": 0.42091944327288067,
|
275 |
+
"step": 315
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 9.14,
|
279 |
+
"learning_rate": 0.0002752737752161383,
|
280 |
+
"loss": 0.7132,
|
281 |
+
"step": 320
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 9.43,
|
285 |
+
"learning_rate": 0.00027440922190201725,
|
286 |
+
"loss": 0.7306,
|
287 |
+
"step": 330
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 9.71,
|
291 |
+
"learning_rate": 0.0002735446685878962,
|
292 |
+
"loss": 0.6461,
|
293 |
+
"step": 340
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 9.99,
|
297 |
+
"learning_rate": 0.0002726801152737752,
|
298 |
+
"loss": 0.6503,
|
299 |
+
"step": 350
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 9.99,
|
303 |
+
"eval_loss": 0.8998618721961975,
|
304 |
+
"eval_runtime": 8.8852,
|
305 |
+
"eval_samples_per_second": 7.09,
|
306 |
+
"eval_steps_per_second": 1.801,
|
307 |
+
"eval_wer": 0.3833825390130747,
|
308 |
+
"step": 350
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 10.28,
|
312 |
+
"learning_rate": 0.00027181556195965417,
|
313 |
+
"loss": 0.6801,
|
314 |
+
"step": 360
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 10.57,
|
318 |
+
"learning_rate": 0.0002709510086455331,
|
319 |
+
"loss": 0.6139,
|
320 |
+
"step": 370
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 10.85,
|
324 |
+
"learning_rate": 0.00027008645533141206,
|
325 |
+
"loss": 0.6159,
|
326 |
+
"step": 380
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 10.99,
|
330 |
+
"eval_loss": 1.150137186050415,
|
331 |
+
"eval_runtime": 8.854,
|
332 |
+
"eval_samples_per_second": 7.115,
|
333 |
+
"eval_steps_per_second": 1.807,
|
334 |
+
"eval_wer": 0.4386334879797554,
|
335 |
+
"step": 385
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 11.14,
|
339 |
+
"learning_rate": 0.00026922190201729104,
|
340 |
+
"loss": 0.8074,
|
341 |
+
"step": 390
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 11.43,
|
345 |
+
"learning_rate": 0.00026835734870317,
|
346 |
+
"loss": 0.6521,
|
347 |
+
"step": 400
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 11.71,
|
351 |
+
"learning_rate": 0.000267492795389049,
|
352 |
+
"loss": 0.6608,
|
353 |
+
"step": 410
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 11.99,
|
357 |
+
"learning_rate": 0.00026662824207492796,
|
358 |
+
"loss": 0.6831,
|
359 |
+
"step": 420
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 11.99,
|
363 |
+
"eval_loss": 1.0860137939453125,
|
364 |
+
"eval_runtime": 8.8923,
|
365 |
+
"eval_samples_per_second": 7.085,
|
366 |
+
"eval_steps_per_second": 1.799,
|
367 |
+
"eval_wer": 0.38253901307465205,
|
368 |
+
"step": 420
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 12.28,
|
372 |
+
"learning_rate": 0.0002657636887608069,
|
373 |
+
"loss": 0.6523,
|
374 |
+
"step": 430
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 12.57,
|
378 |
+
"learning_rate": 0.00026489913544668585,
|
379 |
+
"loss": 0.5839,
|
380 |
+
"step": 440
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 12.85,
|
384 |
+
"learning_rate": 0.0002640345821325648,
|
385 |
+
"loss": 0.5959,
|
386 |
+
"step": 450
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 12.99,
|
390 |
+
"eval_loss": 0.9410010576248169,
|
391 |
+
"eval_runtime": 8.8805,
|
392 |
+
"eval_samples_per_second": 7.094,
|
393 |
+
"eval_steps_per_second": 1.802,
|
394 |
+
"eval_wer": 0.4044706874736398,
|
395 |
+
"step": 455
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 13.14,
|
399 |
+
"learning_rate": 0.00026317002881844374,
|
400 |
+
"loss": 0.6781,
|
401 |
+
"step": 460
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 13.43,
|
405 |
+
"learning_rate": 0.00026230547550432277,
|
406 |
+
"loss": 0.6091,
|
407 |
+
"step": 470
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 13.71,
|
411 |
+
"learning_rate": 0.0002614409221902017,
|
412 |
+
"loss": 0.6359,
|
413 |
+
"step": 480
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 13.99,
|
417 |
+
"learning_rate": 0.00026057636887608066,
|
418 |
+
"loss": 0.7154,
|
419 |
+
"step": 490
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 13.99,
|
423 |
+
"eval_loss": 1.0462627410888672,
|
424 |
+
"eval_runtime": 9.045,
|
425 |
+
"eval_samples_per_second": 6.965,
|
426 |
+
"eval_steps_per_second": 1.769,
|
427 |
+
"eval_wer": 0.38211725010544073,
|
428 |
+
"step": 490
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 14.28,
|
432 |
+
"learning_rate": 0.00025971181556195964,
|
433 |
+
"loss": 0.6386,
|
434 |
+
"step": 500
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 14.57,
|
438 |
+
"learning_rate": 0.0002588472622478386,
|
439 |
+
"loss": 0.5766,
|
440 |
+
"step": 510
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 14.85,
|
444 |
+
"learning_rate": 0.00025798270893371753,
|
445 |
+
"loss": 0.6094,
|
446 |
+
"step": 520
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 14.99,
|
450 |
+
"eval_loss": 0.8598027229309082,
|
451 |
+
"eval_runtime": 8.8755,
|
452 |
+
"eval_samples_per_second": 7.098,
|
453 |
+
"eval_steps_per_second": 1.803,
|
454 |
+
"eval_wer": 0.39645719105862504,
|
455 |
+
"step": 525
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 15.14,
|
459 |
+
"learning_rate": 0.00025711815561959656,
|
460 |
+
"loss": 0.6392,
|
461 |
+
"step": 530
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 15.43,
|
465 |
+
"learning_rate": 0.0002562536023054755,
|
466 |
+
"loss": 0.6047,
|
467 |
+
"step": 540
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 15.71,
|
471 |
+
"learning_rate": 0.0002554755043227665,
|
472 |
+
"loss": 1.3821,
|
473 |
+
"step": 550
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 15.99,
|
477 |
+
"learning_rate": 0.00025461095100864553,
|
478 |
+
"loss": 0.6929,
|
479 |
+
"step": 560
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 15.99,
|
483 |
+
"eval_loss": 0.9493783116340637,
|
484 |
+
"eval_runtime": 8.8602,
|
485 |
+
"eval_samples_per_second": 7.11,
|
486 |
+
"eval_steps_per_second": 1.806,
|
487 |
+
"eval_wer": 0.39308308730493463,
|
488 |
+
"step": 560
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 16.28,
|
492 |
+
"learning_rate": 0.00025374639769452445,
|
493 |
+
"loss": 0.6521,
|
494 |
+
"step": 570
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 16.57,
|
498 |
+
"learning_rate": 0.00025288184438040343,
|
499 |
+
"loss": 0.6791,
|
500 |
+
"step": 580
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 16.85,
|
504 |
+
"learning_rate": 0.0002520172910662824,
|
505 |
+
"loss": 0.7627,
|
506 |
+
"step": 590
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 16.99,
|
510 |
+
"eval_loss": 0.8059821128845215,
|
511 |
+
"eval_runtime": 8.8362,
|
512 |
+
"eval_samples_per_second": 7.13,
|
513 |
+
"eval_steps_per_second": 1.811,
|
514 |
+
"eval_wer": 0.39477013918177983,
|
515 |
+
"step": 595
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 17.14,
|
519 |
+
"learning_rate": 0.0002511527377521614,
|
520 |
+
"loss": 0.6268,
|
521 |
+
"step": 600
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 17.43,
|
525 |
+
"learning_rate": 0.0002502881844380403,
|
526 |
+
"loss": 0.6496,
|
527 |
+
"step": 610
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 17.71,
|
531 |
+
"learning_rate": 0.0002494236311239193,
|
532 |
+
"loss": 0.6172,
|
533 |
+
"step": 620
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 17.99,
|
537 |
+
"learning_rate": 0.00024855907780979824,
|
538 |
+
"loss": 0.601,
|
539 |
+
"step": 630
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 17.99,
|
543 |
+
"eval_loss": 0.989000678062439,
|
544 |
+
"eval_runtime": 8.8942,
|
545 |
+
"eval_samples_per_second": 7.083,
|
546 |
+
"eval_steps_per_second": 1.799,
|
547 |
+
"eval_wer": 0.39645719105862504,
|
548 |
+
"step": 630
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 18.28,
|
552 |
+
"learning_rate": 0.0002476945244956772,
|
553 |
+
"loss": 0.6047,
|
554 |
+
"step": 640
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 18.57,
|
558 |
+
"learning_rate": 0.0002468299711815562,
|
559 |
+
"loss": 0.5297,
|
560 |
+
"step": 650
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 18.85,
|
564 |
+
"learning_rate": 0.0002459654178674351,
|
565 |
+
"loss": 0.546,
|
566 |
+
"step": 660
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 18.99,
|
570 |
+
"eval_loss": 0.8058700561523438,
|
571 |
+
"eval_runtime": 9.0371,
|
572 |
+
"eval_samples_per_second": 6.971,
|
573 |
+
"eval_steps_per_second": 1.77,
|
574 |
+
"eval_wer": 0.39898776887389287,
|
575 |
+
"step": 665
|
576 |
+
},
|
577 |
+
{
|
578 |
+
"epoch": 19.14,
|
579 |
+
"learning_rate": 0.0002451008645533141,
|
580 |
+
"loss": 0.5887,
|
581 |
+
"step": 670
|
582 |
+
},
|
583 |
+
{
|
584 |
+
"epoch": 19.43,
|
585 |
+
"learning_rate": 0.00024423631123919305,
|
586 |
+
"loss": 0.5546,
|
587 |
+
"step": 680
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 19.71,
|
591 |
+
"learning_rate": 0.00024337175792507203,
|
592 |
+
"loss": 0.6092,
|
593 |
+
"step": 690
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 19.99,
|
597 |
+
"learning_rate": 0.00024250720461095097,
|
598 |
+
"loss": 0.5222,
|
599 |
+
"step": 700
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 19.99,
|
603 |
+
"eval_loss": 0.6378720998764038,
|
604 |
+
"eval_runtime": 8.9217,
|
605 |
+
"eval_samples_per_second": 7.061,
|
606 |
+
"eval_steps_per_second": 1.793,
|
607 |
+
"eval_wer": 0.37916490932096164,
|
608 |
+
"step": 700
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 20.28,
|
612 |
+
"learning_rate": 0.00024164265129682995,
|
613 |
+
"loss": 0.6241,
|
614 |
+
"step": 710
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 20.57,
|
618 |
+
"learning_rate": 0.00024077809798270892,
|
619 |
+
"loss": 0.6277,
|
620 |
+
"step": 720
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 20.85,
|
624 |
+
"learning_rate": 0.0002399135446685879,
|
625 |
+
"loss": 0.5802,
|
626 |
+
"step": 730
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 20.99,
|
630 |
+
"eval_loss": 0.699506402015686,
|
631 |
+
"eval_runtime": 8.8339,
|
632 |
+
"eval_samples_per_second": 7.132,
|
633 |
+
"eval_steps_per_second": 1.811,
|
634 |
+
"eval_wer": 0.3660902572754112,
|
635 |
+
"step": 735
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 21.14,
|
639 |
+
"learning_rate": 0.00023904899135446684,
|
640 |
+
"loss": 0.6758,
|
641 |
+
"step": 740
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 21.43,
|
645 |
+
"learning_rate": 0.0002381844380403458,
|
646 |
+
"loss": 0.5938,
|
647 |
+
"step": 750
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 21.71,
|
651 |
+
"learning_rate": 0.00023731988472622476,
|
652 |
+
"loss": 0.6277,
|
653 |
+
"step": 760
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 21.99,
|
657 |
+
"learning_rate": 0.0002364553314121037,
|
658 |
+
"loss": 0.5731,
|
659 |
+
"step": 770
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 21.99,
|
663 |
+
"eval_loss": 0.8405329585075378,
|
664 |
+
"eval_runtime": 8.8637,
|
665 |
+
"eval_samples_per_second": 7.108,
|
666 |
+
"eval_steps_per_second": 1.805,
|
667 |
+
"eval_wer": 0.3606073386756643,
|
668 |
+
"step": 770
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 22.28,
|
672 |
+
"learning_rate": 0.00023559077809798268,
|
673 |
+
"loss": 0.5488,
|
674 |
+
"step": 780
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 22.57,
|
678 |
+
"learning_rate": 0.00023472622478386165,
|
679 |
+
"loss": 0.5718,
|
680 |
+
"step": 790
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 22.85,
|
684 |
+
"learning_rate": 0.00023386167146974063,
|
685 |
+
"loss": 0.5462,
|
686 |
+
"step": 800
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 22.99,
|
690 |
+
"eval_loss": 0.6666640639305115,
|
691 |
+
"eval_runtime": 8.9034,
|
692 |
+
"eval_samples_per_second": 7.076,
|
693 |
+
"eval_steps_per_second": 1.797,
|
694 |
+
"eval_wer": 0.39645719105862504,
|
695 |
+
"step": 805
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 23.14,
|
699 |
+
"learning_rate": 0.00023299711815561957,
|
700 |
+
"loss": 0.5111,
|
701 |
+
"step": 810
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 23.43,
|
705 |
+
"learning_rate": 0.00023213256484149854,
|
706 |
+
"loss": 0.5275,
|
707 |
+
"step": 820
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 23.71,
|
711 |
+
"learning_rate": 0.0002312680115273775,
|
712 |
+
"loss": 0.5692,
|
713 |
+
"step": 830
|
714 |
+
},
|
715 |
+
{
|
716 |
+
"epoch": 23.99,
|
717 |
+
"learning_rate": 0.00023040345821325646,
|
718 |
+
"loss": 0.6057,
|
719 |
+
"step": 840
|
720 |
+
},
|
721 |
+
{
|
722 |
+
"epoch": 23.99,
|
723 |
+
"eval_loss": 0.8396028280258179,
|
724 |
+
"eval_runtime": 9.0837,
|
725 |
+
"eval_samples_per_second": 6.935,
|
726 |
+
"eval_steps_per_second": 1.761,
|
727 |
+
"eval_wer": 0.3762125685364825,
|
728 |
+
"step": 840
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 24.28,
|
732 |
+
"learning_rate": 0.0002295389048991354,
|
733 |
+
"loss": 0.613,
|
734 |
+
"step": 850
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 24.57,
|
738 |
+
"learning_rate": 0.0002286743515850144,
|
739 |
+
"loss": 0.4814,
|
740 |
+
"step": 860
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 24.85,
|
744 |
+
"learning_rate": 0.00022780979827089336,
|
745 |
+
"loss": 0.5323,
|
746 |
+
"step": 870
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 24.99,
|
750 |
+
"eval_loss": 0.9054490327835083,
|
751 |
+
"eval_runtime": 8.8422,
|
752 |
+
"eval_samples_per_second": 7.125,
|
753 |
+
"eval_steps_per_second": 1.81,
|
754 |
+
"eval_wer": 0.39519190215099115,
|
755 |
+
"step": 875
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 25.14,
|
759 |
+
"learning_rate": 0.0002269452449567723,
|
760 |
+
"loss": 0.5585,
|
761 |
+
"step": 880
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 25.43,
|
765 |
+
"learning_rate": 0.00022608069164265128,
|
766 |
+
"loss": 0.5142,
|
767 |
+
"step": 890
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 25.71,
|
771 |
+
"learning_rate": 0.00022521613832853022,
|
772 |
+
"loss": 0.5325,
|
773 |
+
"step": 900
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 25.99,
|
777 |
+
"learning_rate": 0.0002243515850144092,
|
778 |
+
"loss": 0.683,
|
779 |
+
"step": 910
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 25.99,
|
783 |
+
"eval_loss": 0.689756453037262,
|
784 |
+
"eval_runtime": 8.8398,
|
785 |
+
"eval_samples_per_second": 7.127,
|
786 |
+
"eval_steps_per_second": 1.81,
|
787 |
+
"eval_wer": 0.40615773935048505,
|
788 |
+
"step": 910
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 26.28,
|
792 |
+
"learning_rate": 0.00022348703170028814,
|
793 |
+
"loss": 0.5484,
|
794 |
+
"step": 920
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 26.57,
|
798 |
+
"learning_rate": 0.00022262247838616714,
|
799 |
+
"loss": 0.4704,
|
800 |
+
"step": 930
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 26.85,
|
804 |
+
"learning_rate": 0.0002217579250720461,
|
805 |
+
"loss": 0.525,
|
806 |
+
"step": 940
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 26.99,
|
810 |
+
"eval_loss": 0.7244584560394287,
|
811 |
+
"eval_runtime": 8.8634,
|
812 |
+
"eval_samples_per_second": 7.108,
|
813 |
+
"eval_steps_per_second": 1.805,
|
814 |
+
"eval_wer": 0.3884436946436103,
|
815 |
+
"step": 945
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 27.14,
|
819 |
+
"learning_rate": 0.00022089337175792506,
|
820 |
+
"loss": 0.5174,
|
821 |
+
"step": 950
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 27.43,
|
825 |
+
"learning_rate": 0.000220028818443804,
|
826 |
+
"loss": 0.5248,
|
827 |
+
"step": 960
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 27.71,
|
831 |
+
"learning_rate": 0.00021916426512968298,
|
832 |
+
"loss": 0.5203,
|
833 |
+
"step": 970
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 27.99,
|
837 |
+
"learning_rate": 0.00021829971181556193,
|
838 |
+
"loss": 0.4885,
|
839 |
+
"step": 980
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 27.99,
|
843 |
+
"eval_loss": 0.807603120803833,
|
844 |
+
"eval_runtime": 8.8102,
|
845 |
+
"eval_samples_per_second": 7.151,
|
846 |
+
"eval_steps_per_second": 1.816,
|
847 |
+
"eval_wer": 0.4048924504428511,
|
848 |
+
"step": 980
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 28.28,
|
852 |
+
"learning_rate": 0.00021743515850144088,
|
853 |
+
"loss": 0.5582,
|
854 |
+
"step": 990
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 28.57,
|
858 |
+
"learning_rate": 0.00021657060518731988,
|
859 |
+
"loss": 0.4763,
|
860 |
+
"step": 1000
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 28.85,
|
864 |
+
"learning_rate": 0.00021570605187319882,
|
865 |
+
"loss": 0.4653,
|
866 |
+
"step": 1010
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 28.99,
|
870 |
+
"eval_loss": 0.8100489377975464,
|
871 |
+
"eval_runtime": 8.8831,
|
872 |
+
"eval_samples_per_second": 7.092,
|
873 |
+
"eval_steps_per_second": 1.801,
|
874 |
+
"eval_wer": 0.38380430198228593,
|
875 |
+
"step": 1015
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 29.14,
|
879 |
+
"learning_rate": 0.0002148414985590778,
|
880 |
+
"loss": 0.4895,
|
881 |
+
"step": 1020
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 29.43,
|
885 |
+
"learning_rate": 0.00021397694524495674,
|
886 |
+
"loss": 0.4331,
|
887 |
+
"step": 1030
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 29.71,
|
891 |
+
"learning_rate": 0.00021311239193083572,
|
892 |
+
"loss": 0.5435,
|
893 |
+
"step": 1040
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 29.99,
|
897 |
+
"learning_rate": 0.00021224783861671466,
|
898 |
+
"loss": 0.4827,
|
899 |
+
"step": 1050
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 29.99,
|
903 |
+
"eval_loss": 0.7246588468551636,
|
904 |
+
"eval_runtime": 8.886,
|
905 |
+
"eval_samples_per_second": 7.09,
|
906 |
+
"eval_steps_per_second": 1.801,
|
907 |
+
"eval_wer": 0.38633487979755377,
|
908 |
+
"step": 1050
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 30.28,
|
912 |
+
"learning_rate": 0.00021138328530259366,
|
913 |
+
"loss": 0.4931,
|
914 |
+
"step": 1060
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 30.57,
|
918 |
+
"learning_rate": 0.0002105187319884726,
|
919 |
+
"loss": 0.5082,
|
920 |
+
"step": 1070
|
921 |
+
},
|
922 |
+
{
|
923 |
+
"epoch": 30.85,
|
924 |
+
"learning_rate": 0.00020965417867435158,
|
925 |
+
"loss": 0.4839,
|
926 |
+
"step": 1080
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 30.99,
|
930 |
+
"eval_loss": 0.700883150100708,
|
931 |
+
"eval_runtime": 8.8179,
|
932 |
+
"eval_samples_per_second": 7.145,
|
933 |
+
"eval_steps_per_second": 1.814,
|
934 |
+
"eval_wer": 0.3816954871362294,
|
935 |
+
"step": 1085
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 31.14,
|
939 |
+
"learning_rate": 0.00020878962536023053,
|
940 |
+
"loss": 0.6065,
|
941 |
+
"step": 1090
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 31.43,
|
945 |
+
"learning_rate": 0.00020792507204610947,
|
946 |
+
"loss": 0.6116,
|
947 |
+
"step": 1100
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 31.71,
|
951 |
+
"learning_rate": 0.00020706051873198845,
|
952 |
+
"loss": 0.494,
|
953 |
+
"step": 1110
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 31.99,
|
957 |
+
"learning_rate": 0.0002061959654178674,
|
958 |
+
"loss": 0.4982,
|
959 |
+
"step": 1120
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 31.99,
|
963 |
+
"eval_loss": 0.7636587023735046,
|
964 |
+
"eval_runtime": 8.8569,
|
965 |
+
"eval_samples_per_second": 7.113,
|
966 |
+
"eval_steps_per_second": 1.806,
|
967 |
+
"eval_wer": 0.39139603542808943,
|
968 |
+
"step": 1120
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 32.28,
|
972 |
+
"learning_rate": 0.0002053314121037464,
|
973 |
+
"loss": 0.5046,
|
974 |
+
"step": 1130
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 32.57,
|
978 |
+
"learning_rate": 0.00020446685878962534,
|
979 |
+
"loss": 0.5496,
|
980 |
+
"step": 1140
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 32.85,
|
984 |
+
"learning_rate": 0.00020360230547550431,
|
985 |
+
"loss": 0.6105,
|
986 |
+
"step": 1150
|
987 |
+
},
|
988 |
+
{
|
989 |
+
"epoch": 32.99,
|
990 |
+
"eval_loss": 0.7342580556869507,
|
991 |
+
"eval_runtime": 8.8973,
|
992 |
+
"eval_samples_per_second": 7.081,
|
993 |
+
"eval_steps_per_second": 1.798,
|
994 |
+
"eval_wer": 0.39139603542808943,
|
995 |
+
"step": 1155
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 33.14,
|
999 |
+
"learning_rate": 0.00020273775216138326,
|
1000 |
+
"loss": 0.4872,
|
1001 |
+
"step": 1160
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 33.43,
|
1005 |
+
"learning_rate": 0.00020187319884726223,
|
1006 |
+
"loss": 0.4502,
|
1007 |
+
"step": 1170
|
1008 |
+
},
|
1009 |
+
{
|
1010 |
+
"epoch": 33.71,
|
1011 |
+
"learning_rate": 0.00020100864553314118,
|
1012 |
+
"loss": 0.7582,
|
1013 |
+
"step": 1180
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 33.99,
|
1017 |
+
"learning_rate": 0.00020014409221902018,
|
1018 |
+
"loss": 0.4936,
|
1019 |
+
"step": 1190
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 33.99,
|
1023 |
+
"eval_loss": 0.7389687299728394,
|
1024 |
+
"eval_runtime": 8.8292,
|
1025 |
+
"eval_samples_per_second": 7.135,
|
1026 |
+
"eval_steps_per_second": 1.812,
|
1027 |
+
"eval_wer": 0.3762125685364825,
|
1028 |
+
"step": 1190
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 34.28,
|
1032 |
+
"learning_rate": 0.00019927953890489913,
|
1033 |
+
"loss": 0.4426,
|
1034 |
+
"step": 1200
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 34.57,
|
1038 |
+
"learning_rate": 0.0001984149855907781,
|
1039 |
+
"loss": 0.5449,
|
1040 |
+
"step": 1210
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 34.85,
|
1044 |
+
"learning_rate": 0.00019755043227665705,
|
1045 |
+
"loss": 0.4674,
|
1046 |
+
"step": 1220
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 34.99,
|
1050 |
+
"eval_loss": 0.6723992228507996,
|
1051 |
+
"eval_runtime": 8.875,
|
1052 |
+
"eval_samples_per_second": 7.099,
|
1053 |
+
"eval_steps_per_second": 1.803,
|
1054 |
+
"eval_wer": 0.35807676086039647,
|
1055 |
+
"step": 1225
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 35.14,
|
1059 |
+
"learning_rate": 0.000196685878962536,
|
1060 |
+
"loss": 0.4659,
|
1061 |
+
"step": 1230
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 35.43,
|
1065 |
+
"learning_rate": 0.00019582132564841497,
|
1066 |
+
"loss": 0.5037,
|
1067 |
+
"step": 1240
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 35.71,
|
1071 |
+
"learning_rate": 0.0001949567723342939,
|
1072 |
+
"loss": 0.4473,
|
1073 |
+
"step": 1250
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 35.99,
|
1077 |
+
"learning_rate": 0.0001940922190201729,
|
1078 |
+
"loss": 0.4677,
|
1079 |
+
"step": 1260
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 35.99,
|
1083 |
+
"eval_loss": 0.6729607582092285,
|
1084 |
+
"eval_runtime": 9.0573,
|
1085 |
+
"eval_samples_per_second": 6.956,
|
1086 |
+
"eval_steps_per_second": 1.767,
|
1087 |
+
"eval_wer": 0.3487979755377478,
|
1088 |
+
"step": 1260
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 36.28,
|
1092 |
+
"learning_rate": 0.00019322766570605186,
|
1093 |
+
"loss": 0.4732,
|
1094 |
+
"step": 1270
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 36.57,
|
1098 |
+
"learning_rate": 0.00019236311239193083,
|
1099 |
+
"loss": 0.3948,
|
1100 |
+
"step": 1280
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 36.85,
|
1104 |
+
"learning_rate": 0.00019149855907780978,
|
1105 |
+
"loss": 0.516,
|
1106 |
+
"step": 1290
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 36.99,
|
1110 |
+
"eval_loss": 0.6956349611282349,
|
1111 |
+
"eval_runtime": 8.8175,
|
1112 |
+
"eval_samples_per_second": 7.145,
|
1113 |
+
"eval_steps_per_second": 1.815,
|
1114 |
+
"eval_wer": 0.3728384647827921,
|
1115 |
+
"step": 1295
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 37.14,
|
1119 |
+
"learning_rate": 0.00019063400576368875,
|
1120 |
+
"loss": 0.4974,
|
1121 |
+
"step": 1300
|
1122 |
+
},
|
1123 |
+
{
|
1124 |
+
"epoch": 37.43,
|
1125 |
+
"learning_rate": 0.0001897694524495677,
|
1126 |
+
"loss": 0.4605,
|
1127 |
+
"step": 1310
|
1128 |
+
},
|
1129 |
+
{
|
1130 |
+
"epoch": 37.71,
|
1131 |
+
"learning_rate": 0.00018890489913544667,
|
1132 |
+
"loss": 0.5494,
|
1133 |
+
"step": 1320
|
1134 |
+
},
|
1135 |
+
{
|
1136 |
+
"epoch": 37.99,
|
1137 |
+
"learning_rate": 0.00018804034582132565,
|
1138 |
+
"loss": 0.4507,
|
1139 |
+
"step": 1330
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 37.99,
|
1143 |
+
"eval_loss": 0.6482787132263184,
|
1144 |
+
"eval_runtime": 8.8046,
|
1145 |
+
"eval_samples_per_second": 7.155,
|
1146 |
+
"eval_steps_per_second": 1.817,
|
1147 |
+
"eval_wer": 0.3614508646140869,
|
1148 |
+
"step": 1330
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 38.28,
|
1152 |
+
"learning_rate": 0.0001871757925072046,
|
1153 |
+
"loss": 0.4868,
|
1154 |
+
"step": 1340
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 38.57,
|
1158 |
+
"learning_rate": 0.00018631123919308357,
|
1159 |
+
"loss": 0.3934,
|
1160 |
+
"step": 1350
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 38.85,
|
1164 |
+
"learning_rate": 0.0001854466858789625,
|
1165 |
+
"loss": 0.4207,
|
1166 |
+
"step": 1360
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 38.99,
|
1170 |
+
"eval_loss": 0.7718012928962708,
|
1171 |
+
"eval_runtime": 8.854,
|
1172 |
+
"eval_samples_per_second": 7.115,
|
1173 |
+
"eval_steps_per_second": 1.807,
|
1174 |
+
"eval_wer": 0.34837621256853646,
|
1175 |
+
"step": 1365
|
1176 |
+
},
|
1177 |
+
{
|
1178 |
+
"epoch": 39.14,
|
1179 |
+
"learning_rate": 0.00018458213256484148,
|
1180 |
+
"loss": 0.4837,
|
1181 |
+
"step": 1370
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 39.43,
|
1185 |
+
"learning_rate": 0.00018371757925072043,
|
1186 |
+
"loss": 0.4663,
|
1187 |
+
"step": 1380
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 39.71,
|
1191 |
+
"learning_rate": 0.0001828530259365994,
|
1192 |
+
"loss": 0.4572,
|
1193 |
+
"step": 1390
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 39.99,
|
1197 |
+
"learning_rate": 0.00018198847262247838,
|
1198 |
+
"loss": 0.4803,
|
1199 |
+
"step": 1400
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 39.99,
|
1203 |
+
"eval_loss": 0.831644594669342,
|
1204 |
+
"eval_runtime": 8.7883,
|
1205 |
+
"eval_samples_per_second": 7.169,
|
1206 |
+
"eval_steps_per_second": 1.821,
|
1207 |
+
"eval_wer": 0.3774778574441164,
|
1208 |
+
"step": 1400
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 40.28,
|
1212 |
+
"learning_rate": 0.00018112391930835735,
|
1213 |
+
"loss": 0.5059,
|
1214 |
+
"step": 1410
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 40.57,
|
1218 |
+
"learning_rate": 0.0001802593659942363,
|
1219 |
+
"loss": 0.4758,
|
1220 |
+
"step": 1420
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 40.85,
|
1224 |
+
"learning_rate": 0.00017939481268011527,
|
1225 |
+
"loss": 0.3946,
|
1226 |
+
"step": 1430
|
1227 |
+
},
|
1228 |
+
{
|
1229 |
+
"epoch": 40.99,
|
1230 |
+
"eval_loss": 0.8322010636329651,
|
1231 |
+
"eval_runtime": 8.8486,
|
1232 |
+
"eval_samples_per_second": 7.12,
|
1233 |
+
"eval_steps_per_second": 1.808,
|
1234 |
+
"eval_wer": 0.3568114719527625,
|
1235 |
+
"step": 1435
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 41.14,
|
1239 |
+
"learning_rate": 0.00017853025936599422,
|
1240 |
+
"loss": 0.4533,
|
1241 |
+
"step": 1440
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 41.43,
|
1245 |
+
"learning_rate": 0.00017766570605187316,
|
1246 |
+
"loss": 0.4331,
|
1247 |
+
"step": 1450
|
1248 |
+
},
|
1249 |
+
{
|
1250 |
+
"epoch": 41.71,
|
1251 |
+
"learning_rate": 0.00017680115273775214,
|
1252 |
+
"loss": 0.4563,
|
1253 |
+
"step": 1460
|
1254 |
+
},
|
1255 |
+
{
|
1256 |
+
"epoch": 41.99,
|
1257 |
+
"learning_rate": 0.0001759365994236311,
|
1258 |
+
"loss": 0.411,
|
1259 |
+
"step": 1470
|
1260 |
+
},
|
1261 |
+
{
|
1262 |
+
"epoch": 41.99,
|
1263 |
+
"eval_loss": 0.9933066368103027,
|
1264 |
+
"eval_runtime": 8.8766,
|
1265 |
+
"eval_samples_per_second": 7.097,
|
1266 |
+
"eval_steps_per_second": 1.802,
|
1267 |
+
"eval_wer": 0.3707296499367356,
|
1268 |
+
"step": 1470
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 42.28,
|
1272 |
+
"learning_rate": 0.00017507204610951008,
|
1273 |
+
"loss": 0.4233,
|
1274 |
+
"step": 1480
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 42.57,
|
1278 |
+
"learning_rate": 0.00017420749279538903,
|
1279 |
+
"loss": 0.4351,
|
1280 |
+
"step": 1490
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 42.85,
|
1284 |
+
"learning_rate": 0.000173342939481268,
|
1285 |
+
"loss": 0.4405,
|
1286 |
+
"step": 1500
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 42.99,
|
1290 |
+
"eval_loss": 0.8789033889770508,
|
1291 |
+
"eval_runtime": 8.9051,
|
1292 |
+
"eval_samples_per_second": 7.075,
|
1293 |
+
"eval_steps_per_second": 1.797,
|
1294 |
+
"eval_wer": 0.3943483762125685,
|
1295 |
+
"step": 1505
|
1296 |
+
},
|
1297 |
+
{
|
1298 |
+
"epoch": 43.14,
|
1299 |
+
"learning_rate": 0.00017247838616714695,
|
1300 |
+
"loss": 0.4044,
|
1301 |
+
"step": 1510
|
1302 |
+
},
|
1303 |
+
{
|
1304 |
+
"epoch": 43.43,
|
1305 |
+
"learning_rate": 0.00017161383285302592,
|
1306 |
+
"loss": 0.4411,
|
1307 |
+
"step": 1520
|
1308 |
+
},
|
1309 |
+
{
|
1310 |
+
"epoch": 43.71,
|
1311 |
+
"learning_rate": 0.00017074927953890487,
|
1312 |
+
"loss": 0.4061,
|
1313 |
+
"step": 1530
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 43.99,
|
1317 |
+
"learning_rate": 0.00016988472622478387,
|
1318 |
+
"loss": 0.5124,
|
1319 |
+
"step": 1540
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 43.99,
|
1323 |
+
"eval_loss": 0.9030116200447083,
|
1324 |
+
"eval_runtime": 8.8171,
|
1325 |
+
"eval_samples_per_second": 7.145,
|
1326 |
+
"eval_steps_per_second": 1.815,
|
1327 |
+
"eval_wer": 0.3707296499367356,
|
1328 |
+
"step": 1540
|
1329 |
+
},
|
1330 |
+
{
|
1331 |
+
"epoch": 44.28,
|
1332 |
+
"learning_rate": 0.00016902017291066282,
|
1333 |
+
"loss": 0.4745,
|
1334 |
+
"step": 1550
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 44.57,
|
1338 |
+
"learning_rate": 0.00016815561959654176,
|
1339 |
+
"loss": 0.4145,
|
1340 |
+
"step": 1560
|
1341 |
+
},
|
1342 |
+
{
|
1343 |
+
"epoch": 44.85,
|
1344 |
+
"learning_rate": 0.00016729106628242074,
|
1345 |
+
"loss": 0.5959,
|
1346 |
+
"step": 1570
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 44.99,
|
1350 |
+
"eval_loss": 0.7809452414512634,
|
1351 |
+
"eval_runtime": 8.9655,
|
1352 |
+
"eval_samples_per_second": 7.027,
|
1353 |
+
"eval_steps_per_second": 1.785,
|
1354 |
+
"eval_wer": 0.39477013918177983,
|
1355 |
+
"step": 1575
|
1356 |
+
},
|
1357 |
+
{
|
1358 |
+
"epoch": 45.14,
|
1359 |
+
"learning_rate": 0.00016642651296829968,
|
1360 |
+
"loss": 0.4527,
|
1361 |
+
"step": 1580
|
1362 |
+
},
|
1363 |
+
{
|
1364 |
+
"epoch": 45.43,
|
1365 |
+
"learning_rate": 0.00016556195965417866,
|
1366 |
+
"loss": 0.3988,
|
1367 |
+
"step": 1590
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 45.71,
|
1371 |
+
"learning_rate": 0.0001646974063400576,
|
1372 |
+
"loss": 0.4175,
|
1373 |
+
"step": 1600
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 45.99,
|
1377 |
+
"learning_rate": 0.0001638328530259366,
|
1378 |
+
"loss": 0.3841,
|
1379 |
+
"step": 1610
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 45.99,
|
1383 |
+
"eval_loss": 0.7716331481933594,
|
1384 |
+
"eval_runtime": 8.8671,
|
1385 |
+
"eval_samples_per_second": 7.105,
|
1386 |
+
"eval_steps_per_second": 1.804,
|
1387 |
+
"eval_wer": 0.39645719105862504,
|
1388 |
+
"step": 1610
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 46.28,
|
1392 |
+
"learning_rate": 0.00016296829971181555,
|
1393 |
+
"loss": 0.4015,
|
1394 |
+
"step": 1620
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 46.57,
|
1398 |
+
"learning_rate": 0.00016210374639769452,
|
1399 |
+
"loss": 0.453,
|
1400 |
+
"step": 1630
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 46.85,
|
1404 |
+
"learning_rate": 0.00016123919308357347,
|
1405 |
+
"loss": 0.3975,
|
1406 |
+
"step": 1640
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 46.99,
|
1410 |
+
"eval_loss": 0.7064468860626221,
|
1411 |
+
"eval_runtime": 8.9124,
|
1412 |
+
"eval_samples_per_second": 7.069,
|
1413 |
+
"eval_steps_per_second": 1.795,
|
1414 |
+
"eval_wer": 0.39308308730493463,
|
1415 |
+
"step": 1645
|
1416 |
+
},
|
1417 |
+
{
|
1418 |
+
"epoch": 47.14,
|
1419 |
+
"learning_rate": 0.00016037463976945244,
|
1420 |
+
"loss": 0.5041,
|
1421 |
+
"step": 1650
|
1422 |
+
},
|
1423 |
+
{
|
1424 |
+
"epoch": 47.43,
|
1425 |
+
"learning_rate": 0.0001595100864553314,
|
1426 |
+
"loss": 0.4389,
|
1427 |
+
"step": 1660
|
1428 |
+
},
|
1429 |
+
{
|
1430 |
+
"epoch": 47.71,
|
1431 |
+
"learning_rate": 0.00015873198847262245,
|
1432 |
+
"loss": 1.7618,
|
1433 |
+
"step": 1670
|
1434 |
+
},
|
1435 |
+
{
|
1436 |
+
"epoch": 47.99,
|
1437 |
+
"learning_rate": 0.00015786743515850142,
|
1438 |
+
"loss": 1.4983,
|
1439 |
+
"step": 1680
|
1440 |
+
},
|
1441 |
+
{
|
1442 |
+
"epoch": 47.99,
|
1443 |
+
"eval_loss": 3.286560297012329,
|
1444 |
+
"eval_runtime": 8.8248,
|
1445 |
+
"eval_samples_per_second": 7.139,
|
1446 |
+
"eval_steps_per_second": 1.813,
|
1447 |
+
"eval_wer": 0.3627161535217208,
|
1448 |
+
"step": 1680
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 48.28,
|
1452 |
+
"learning_rate": 0.00015700288184438037,
|
1453 |
+
"loss": 0.6489,
|
1454 |
+
"step": 1690
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 48.57,
|
1458 |
+
"learning_rate": 0.00015613832853025937,
|
1459 |
+
"loss": 0.4141,
|
1460 |
+
"step": 1700
|
1461 |
+
},
|
1462 |
+
{
|
1463 |
+
"epoch": 48.85,
|
1464 |
+
"learning_rate": 0.00015527377521613831,
|
1465 |
+
"loss": 0.3962,
|
1466 |
+
"step": 1710
|
1467 |
+
},
|
1468 |
+
{
|
1469 |
+
"epoch": 48.99,
|
1470 |
+
"eval_loss": 0.6486034989356995,
|
1471 |
+
"eval_runtime": 8.9267,
|
1472 |
+
"eval_samples_per_second": 7.057,
|
1473 |
+
"eval_steps_per_second": 1.792,
|
1474 |
+
"eval_wer": 0.36482496836777734,
|
1475 |
+
"step": 1715
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 49.14,
|
1479 |
+
"learning_rate": 0.0001544092219020173,
|
1480 |
+
"loss": 0.4294,
|
1481 |
+
"step": 1720
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 49.43,
|
1485 |
+
"learning_rate": 0.00015354466858789623,
|
1486 |
+
"loss": 0.3977,
|
1487 |
+
"step": 1730
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 49.71,
|
1491 |
+
"learning_rate": 0.0001526801152737752,
|
1492 |
+
"loss": 0.398,
|
1493 |
+
"step": 1740
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 49.99,
|
1497 |
+
"learning_rate": 0.00015181556195965415,
|
1498 |
+
"loss": 0.4422,
|
1499 |
+
"step": 1750
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 49.99,
|
1503 |
+
"eval_loss": 0.8449748158454895,
|
1504 |
+
"eval_runtime": 8.8924,
|
1505 |
+
"eval_samples_per_second": 7.085,
|
1506 |
+
"eval_steps_per_second": 1.799,
|
1507 |
+
"eval_wer": 0.3778996204133277,
|
1508 |
+
"step": 1750
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 50.28,
|
1512 |
+
"learning_rate": 0.0001509510086455331,
|
1513 |
+
"loss": 0.4246,
|
1514 |
+
"step": 1760
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 50.57,
|
1518 |
+
"learning_rate": 0.0001500864553314121,
|
1519 |
+
"loss": 0.4109,
|
1520 |
+
"step": 1770
|
1521 |
+
},
|
1522 |
+
{
|
1523 |
+
"epoch": 50.85,
|
1524 |
+
"learning_rate": 0.00014922190201729105,
|
1525 |
+
"loss": 0.4198,
|
1526 |
+
"step": 1780
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 50.99,
|
1530 |
+
"eval_loss": 0.7628262639045715,
|
1531 |
+
"eval_runtime": 8.8995,
|
1532 |
+
"eval_samples_per_second": 7.079,
|
1533 |
+
"eval_steps_per_second": 1.798,
|
1534 |
+
"eval_wer": 0.35638970898355127,
|
1535 |
+
"step": 1785
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 51.14,
|
1539 |
+
"learning_rate": 0.00014835734870317002,
|
1540 |
+
"loss": 0.5414,
|
1541 |
+
"step": 1790
|
1542 |
+
},
|
1543 |
+
{
|
1544 |
+
"epoch": 51.43,
|
1545 |
+
"learning_rate": 0.000147492795389049,
|
1546 |
+
"loss": 0.4775,
|
1547 |
+
"step": 1800
|
1548 |
+
},
|
1549 |
+
{
|
1550 |
+
"epoch": 51.71,
|
1551 |
+
"learning_rate": 0.00014662824207492794,
|
1552 |
+
"loss": 0.3867,
|
1553 |
+
"step": 1810
|
1554 |
+
},
|
1555 |
+
{
|
1556 |
+
"epoch": 51.99,
|
1557 |
+
"learning_rate": 0.00014576368876080689,
|
1558 |
+
"loss": 0.3577,
|
1559 |
+
"step": 1820
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 51.99,
|
1563 |
+
"eval_loss": 0.7552944421768188,
|
1564 |
+
"eval_runtime": 8.8794,
|
1565 |
+
"eval_samples_per_second": 7.095,
|
1566 |
+
"eval_steps_per_second": 1.802,
|
1567 |
+
"eval_wer": 0.3677773091522564,
|
1568 |
+
"step": 1820
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 52.28,
|
1572 |
+
"learning_rate": 0.00014489913544668586,
|
1573 |
+
"loss": 0.4055,
|
1574 |
+
"step": 1830
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 52.57,
|
1578 |
+
"learning_rate": 0.00014403458213256483,
|
1579 |
+
"loss": 0.3304,
|
1580 |
+
"step": 1840
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 52.85,
|
1584 |
+
"learning_rate": 0.00014317002881844378,
|
1585 |
+
"loss": 0.4425,
|
1586 |
+
"step": 1850
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 52.99,
|
1590 |
+
"eval_loss": 0.7565546631813049,
|
1591 |
+
"eval_runtime": 8.899,
|
1592 |
+
"eval_samples_per_second": 7.079,
|
1593 |
+
"eval_steps_per_second": 1.798,
|
1594 |
+
"eval_wer": 0.37157317587515815,
|
1595 |
+
"step": 1855
|
1596 |
+
},
|
1597 |
+
{
|
1598 |
+
"epoch": 53.14,
|
1599 |
+
"learning_rate": 0.00014230547550432275,
|
1600 |
+
"loss": 0.3659,
|
1601 |
+
"step": 1860
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 53.43,
|
1605 |
+
"learning_rate": 0.00014144092219020173,
|
1606 |
+
"loss": 0.3904,
|
1607 |
+
"step": 1870
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 53.71,
|
1611 |
+
"learning_rate": 0.00014057636887608067,
|
1612 |
+
"loss": 0.3905,
|
1613 |
+
"step": 1880
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 53.99,
|
1617 |
+
"learning_rate": 0.00013971181556195965,
|
1618 |
+
"loss": 0.3492,
|
1619 |
+
"step": 1890
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 53.99,
|
1623 |
+
"eval_loss": 0.7709900736808777,
|
1624 |
+
"eval_runtime": 8.8285,
|
1625 |
+
"eval_samples_per_second": 7.136,
|
1626 |
+
"eval_steps_per_second": 1.812,
|
1627 |
+
"eval_wer": 0.3631379164909321,
|
1628 |
+
"step": 1890
|
1629 |
+
},
|
1630 |
+
{
|
1631 |
+
"epoch": 54.28,
|
1632 |
+
"learning_rate": 0.00013884726224783862,
|
1633 |
+
"loss": 0.4355,
|
1634 |
+
"step": 1900
|
1635 |
+
},
|
1636 |
+
{
|
1637 |
+
"epoch": 54.57,
|
1638 |
+
"learning_rate": 0.00013798270893371757,
|
1639 |
+
"loss": 0.3472,
|
1640 |
+
"step": 1910
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 54.85,
|
1644 |
+
"learning_rate": 0.00013711815561959654,
|
1645 |
+
"loss": 0.3731,
|
1646 |
+
"step": 1920
|
1647 |
+
},
|
1648 |
+
{
|
1649 |
+
"epoch": 54.99,
|
1650 |
+
"eval_loss": 0.7737426161766052,
|
1651 |
+
"eval_runtime": 8.9423,
|
1652 |
+
"eval_samples_per_second": 7.045,
|
1653 |
+
"eval_steps_per_second": 1.789,
|
1654 |
+
"eval_wer": 0.3627161535217208,
|
1655 |
+
"step": 1925
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 55.14,
|
1659 |
+
"learning_rate": 0.00013625360230547548,
|
1660 |
+
"loss": 0.3994,
|
1661 |
+
"step": 1930
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 55.43,
|
1665 |
+
"learning_rate": 0.00013538904899135446,
|
1666 |
+
"loss": 0.41,
|
1667 |
+
"step": 1940
|
1668 |
+
},
|
1669 |
+
{
|
1670 |
+
"epoch": 55.71,
|
1671 |
+
"learning_rate": 0.0001345244956772334,
|
1672 |
+
"loss": 0.3445,
|
1673 |
+
"step": 1950
|
1674 |
+
},
|
1675 |
+
{
|
1676 |
+
"epoch": 55.99,
|
1677 |
+
"learning_rate": 0.00013365994236311238,
|
1678 |
+
"loss": 0.3868,
|
1679 |
+
"step": 1960
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 55.99,
|
1683 |
+
"eval_loss": 0.7020638585090637,
|
1684 |
+
"eval_runtime": 8.7931,
|
1685 |
+
"eval_samples_per_second": 7.165,
|
1686 |
+
"eval_steps_per_second": 1.82,
|
1687 |
+
"eval_wer": 0.35723323492197384,
|
1688 |
+
"step": 1960
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 56.28,
|
1692 |
+
"learning_rate": 0.00013279538904899135,
|
1693 |
+
"loss": 0.5803,
|
1694 |
+
"step": 1970
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 56.57,
|
1698 |
+
"learning_rate": 0.0001319308357348703,
|
1699 |
+
"loss": 0.4467,
|
1700 |
+
"step": 1980
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 56.85,
|
1704 |
+
"learning_rate": 0.00013106628242074927,
|
1705 |
+
"loss": 0.3311,
|
1706 |
+
"step": 1990
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 56.99,
|
1710 |
+
"eval_loss": 0.6602898240089417,
|
1711 |
+
"eval_runtime": 8.8334,
|
1712 |
+
"eval_samples_per_second": 7.132,
|
1713 |
+
"eval_steps_per_second": 1.811,
|
1714 |
+
"eval_wer": 0.3517503163222269,
|
1715 |
+
"step": 1995
|
1716 |
+
},
|
1717 |
+
{
|
1718 |
+
"epoch": 57.14,
|
1719 |
+
"learning_rate": 0.00013020172910662824,
|
1720 |
+
"loss": 0.3973,
|
1721 |
+
"step": 2000
|
1722 |
+
},
|
1723 |
+
{
|
1724 |
+
"epoch": 57.43,
|
1725 |
+
"learning_rate": 0.0001293371757925072,
|
1726 |
+
"loss": 0.3832,
|
1727 |
+
"step": 2010
|
1728 |
+
},
|
1729 |
+
{
|
1730 |
+
"epoch": 57.71,
|
1731 |
+
"learning_rate": 0.00012847262247838616,
|
1732 |
+
"loss": 0.3418,
|
1733 |
+
"step": 2020
|
1734 |
+
},
|
1735 |
+
{
|
1736 |
+
"epoch": 57.99,
|
1737 |
+
"learning_rate": 0.00012760806916426514,
|
1738 |
+
"loss": 0.3993,
|
1739 |
+
"step": 2030
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 57.99,
|
1743 |
+
"eval_loss": 0.666448712348938,
|
1744 |
+
"eval_runtime": 8.8405,
|
1745 |
+
"eval_samples_per_second": 7.126,
|
1746 |
+
"eval_steps_per_second": 1.81,
|
1747 |
+
"eval_wer": 0.35807676086039647,
|
1748 |
+
"step": 2030
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 58.28,
|
1752 |
+
"learning_rate": 0.00012674351585014408,
|
1753 |
+
"loss": 0.3879,
|
1754 |
+
"step": 2040
|
1755 |
+
},
|
1756 |
+
{
|
1757 |
+
"epoch": 58.57,
|
1758 |
+
"learning_rate": 0.00012587896253602303,
|
1759 |
+
"loss": 0.3436,
|
1760 |
+
"step": 2050
|
1761 |
+
},
|
1762 |
+
{
|
1763 |
+
"epoch": 58.85,
|
1764 |
+
"learning_rate": 0.000125014409221902,
|
1765 |
+
"loss": 0.4226,
|
1766 |
+
"step": 2060
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 58.99,
|
1770 |
+
"eval_loss": 0.6813275814056396,
|
1771 |
+
"eval_runtime": 8.8439,
|
1772 |
+
"eval_samples_per_second": 7.124,
|
1773 |
+
"eval_steps_per_second": 1.809,
|
1774 |
+
"eval_wer": 0.3551244200759173,
|
1775 |
+
"step": 2065
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 59.14,
|
1779 |
+
"learning_rate": 0.00012414985590778098,
|
1780 |
+
"loss": 0.4158,
|
1781 |
+
"step": 2070
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 59.43,
|
1785 |
+
"learning_rate": 0.00012328530259365992,
|
1786 |
+
"loss": 0.3934,
|
1787 |
+
"step": 2080
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 59.71,
|
1791 |
+
"learning_rate": 0.0001224207492795389,
|
1792 |
+
"loss": 0.3911,
|
1793 |
+
"step": 2090
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 59.99,
|
1797 |
+
"learning_rate": 0.00012155619596541786,
|
1798 |
+
"loss": 0.4143,
|
1799 |
+
"step": 2100
|
1800 |
+
},
|
1801 |
+
{
|
1802 |
+
"epoch": 59.99,
|
1803 |
+
"eval_loss": 0.6566696763038635,
|
1804 |
+
"eval_runtime": 8.8077,
|
1805 |
+
"eval_samples_per_second": 7.153,
|
1806 |
+
"eval_steps_per_second": 1.817,
|
1807 |
+
"eval_wer": 0.3568114719527625,
|
1808 |
+
"step": 2100
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 60.28,
|
1812 |
+
"learning_rate": 0.00012069164265129682,
|
1813 |
+
"loss": 0.3702,
|
1814 |
+
"step": 2110
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 60.57,
|
1818 |
+
"learning_rate": 0.00011982708933717579,
|
1819 |
+
"loss": 0.3573,
|
1820 |
+
"step": 2120
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 60.85,
|
1824 |
+
"learning_rate": 0.00011896253602305475,
|
1825 |
+
"loss": 0.3623,
|
1826 |
+
"step": 2130
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 60.99,
|
1830 |
+
"eval_loss": 0.6568236947059631,
|
1831 |
+
"eval_runtime": 8.8843,
|
1832 |
+
"eval_samples_per_second": 7.091,
|
1833 |
+
"eval_steps_per_second": 1.801,
|
1834 |
+
"eval_wer": 0.34542387178405737,
|
1835 |
+
"step": 2135
|
1836 |
+
},
|
1837 |
+
{
|
1838 |
+
"epoch": 61.14,
|
1839 |
+
"learning_rate": 0.00011809798270893371,
|
1840 |
+
"loss": 0.3849,
|
1841 |
+
"step": 2140
|
1842 |
+
},
|
1843 |
+
{
|
1844 |
+
"epoch": 61.43,
|
1845 |
+
"learning_rate": 0.00011723342939481268,
|
1846 |
+
"loss": 0.4037,
|
1847 |
+
"step": 2150
|
1848 |
+
},
|
1849 |
+
{
|
1850 |
+
"epoch": 61.71,
|
1851 |
+
"learning_rate": 0.00011636887608069163,
|
1852 |
+
"loss": 0.405,
|
1853 |
+
"step": 2160
|
1854 |
+
},
|
1855 |
+
{
|
1856 |
+
"epoch": 61.99,
|
1857 |
+
"learning_rate": 0.00011550432276657059,
|
1858 |
+
"loss": 0.3228,
|
1859 |
+
"step": 2170
|
1860 |
+
},
|
1861 |
+
{
|
1862 |
+
"epoch": 61.99,
|
1863 |
+
"eval_loss": 0.7325943112373352,
|
1864 |
+
"eval_runtime": 8.898,
|
1865 |
+
"eval_samples_per_second": 7.08,
|
1866 |
+
"eval_steps_per_second": 1.798,
|
1867 |
+
"eval_wer": 0.3568114719527625,
|
1868 |
+
"step": 2170
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 62.28,
|
1872 |
+
"learning_rate": 0.00011463976945244955,
|
1873 |
+
"loss": 0.363,
|
1874 |
+
"step": 2180
|
1875 |
+
},
|
1876 |
+
{
|
1877 |
+
"epoch": 62.57,
|
1878 |
+
"learning_rate": 0.00011377521613832852,
|
1879 |
+
"loss": 0.3833,
|
1880 |
+
"step": 2190
|
1881 |
+
},
|
1882 |
+
{
|
1883 |
+
"epoch": 62.85,
|
1884 |
+
"learning_rate": 0.00011291066282420748,
|
1885 |
+
"loss": 0.3204,
|
1886 |
+
"step": 2200
|
1887 |
+
},
|
1888 |
+
{
|
1889 |
+
"epoch": 62.99,
|
1890 |
+
"eval_loss": 0.7277215123176575,
|
1891 |
+
"eval_runtime": 8.9735,
|
1892 |
+
"eval_samples_per_second": 7.021,
|
1893 |
+
"eval_steps_per_second": 1.783,
|
1894 |
+
"eval_wer": 0.3639814424293547,
|
1895 |
+
"step": 2205
|
1896 |
+
},
|
1897 |
+
{
|
1898 |
+
"epoch": 63.14,
|
1899 |
+
"learning_rate": 0.00011204610951008644,
|
1900 |
+
"loss": 0.3488,
|
1901 |
+
"step": 2210
|
1902 |
+
},
|
1903 |
+
{
|
1904 |
+
"epoch": 63.43,
|
1905 |
+
"learning_rate": 0.00011118155619596542,
|
1906 |
+
"loss": 0.3561,
|
1907 |
+
"step": 2220
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 63.71,
|
1911 |
+
"learning_rate": 0.00011031700288184437,
|
1912 |
+
"loss": 0.376,
|
1913 |
+
"step": 2230
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 63.99,
|
1917 |
+
"learning_rate": 0.00010945244956772333,
|
1918 |
+
"loss": 0.377,
|
1919 |
+
"step": 2240
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 63.99,
|
1923 |
+
"eval_loss": 0.7144931554794312,
|
1924 |
+
"eval_runtime": 8.9061,
|
1925 |
+
"eval_samples_per_second": 7.074,
|
1926 |
+
"eval_steps_per_second": 1.797,
|
1927 |
+
"eval_wer": 0.3584985238296078,
|
1928 |
+
"step": 2240
|
1929 |
+
},
|
1930 |
+
{
|
1931 |
+
"epoch": 64.28,
|
1932 |
+
"learning_rate": 0.00010858789625360231,
|
1933 |
+
"loss": 0.3267,
|
1934 |
+
"step": 2250
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 64.57,
|
1938 |
+
"learning_rate": 0.00010772334293948127,
|
1939 |
+
"loss": 0.4145,
|
1940 |
+
"step": 2260
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 64.85,
|
1944 |
+
"learning_rate": 0.00010685878962536021,
|
1945 |
+
"loss": 0.3487,
|
1946 |
+
"step": 2270
|
1947 |
+
},
|
1948 |
+
{
|
1949 |
+
"epoch": 64.99,
|
1950 |
+
"eval_loss": 0.6943061351776123,
|
1951 |
+
"eval_runtime": 8.8303,
|
1952 |
+
"eval_samples_per_second": 7.134,
|
1953 |
+
"eval_steps_per_second": 1.812,
|
1954 |
+
"eval_wer": 0.350485027414593,
|
1955 |
+
"step": 2275
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 65.14,
|
1959 |
+
"learning_rate": 0.00010599423631123917,
|
1960 |
+
"loss": 0.3827,
|
1961 |
+
"step": 2280
|
1962 |
+
},
|
1963 |
+
{
|
1964 |
+
"epoch": 65.43,
|
1965 |
+
"learning_rate": 0.00010512968299711815,
|
1966 |
+
"loss": 0.3328,
|
1967 |
+
"step": 2290
|
1968 |
+
},
|
1969 |
+
{
|
1970 |
+
"epoch": 65.71,
|
1971 |
+
"learning_rate": 0.00010426512968299711,
|
1972 |
+
"loss": 0.3904,
|
1973 |
+
"step": 2300
|
1974 |
+
},
|
1975 |
+
{
|
1976 |
+
"epoch": 65.99,
|
1977 |
+
"learning_rate": 0.00010340057636887607,
|
1978 |
+
"loss": 0.343,
|
1979 |
+
"step": 2310
|
1980 |
+
},
|
1981 |
+
{
|
1982 |
+
"epoch": 65.99,
|
1983 |
+
"eval_loss": 0.74614018201828,
|
1984 |
+
"eval_runtime": 8.7684,
|
1985 |
+
"eval_samples_per_second": 7.185,
|
1986 |
+
"eval_steps_per_second": 1.825,
|
1987 |
+
"eval_wer": 0.33951919021509913,
|
1988 |
+
"step": 2310
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 66.28,
|
1992 |
+
"learning_rate": 0.00010253602305475504,
|
1993 |
+
"loss": 0.336,
|
1994 |
+
"step": 2320
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 66.57,
|
1998 |
+
"learning_rate": 0.000101671469740634,
|
1999 |
+
"loss": 0.3663,
|
2000 |
+
"step": 2330
|
2001 |
+
},
|
2002 |
+
{
|
2003 |
+
"epoch": 66.85,
|
2004 |
+
"learning_rate": 0.00010080691642651296,
|
2005 |
+
"loss": 0.3251,
|
2006 |
+
"step": 2340
|
2007 |
+
},
|
2008 |
+
{
|
2009 |
+
"epoch": 66.99,
|
2010 |
+
"eval_loss": 0.7441524863243103,
|
2011 |
+
"eval_runtime": 8.8232,
|
2012 |
+
"eval_samples_per_second": 7.14,
|
2013 |
+
"eval_steps_per_second": 1.813,
|
2014 |
+
"eval_wer": 0.35638970898355127,
|
2015 |
+
"step": 2345
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 67.14,
|
2019 |
+
"learning_rate": 9.994236311239192e-05,
|
2020 |
+
"loss": 0.3964,
|
2021 |
+
"step": 2350
|
2022 |
+
},
|
2023 |
+
{
|
2024 |
+
"epoch": 67.43,
|
2025 |
+
"learning_rate": 9.90778097982709e-05,
|
2026 |
+
"loss": 0.3658,
|
2027 |
+
"step": 2360
|
2028 |
+
},
|
2029 |
+
{
|
2030 |
+
"epoch": 67.71,
|
2031 |
+
"learning_rate": 9.821325648414985e-05,
|
2032 |
+
"loss": 0.3515,
|
2033 |
+
"step": 2370
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 67.99,
|
2037 |
+
"learning_rate": 9.73487031700288e-05,
|
2038 |
+
"loss": 0.3135,
|
2039 |
+
"step": 2380
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 67.99,
|
2043 |
+
"eval_loss": 0.7330945134162903,
|
2044 |
+
"eval_runtime": 8.8098,
|
2045 |
+
"eval_samples_per_second": 7.151,
|
2046 |
+
"eval_steps_per_second": 1.816,
|
2047 |
+
"eval_wer": 0.3530156052298608,
|
2048 |
+
"step": 2380
|
2049 |
+
},
|
2050 |
+
{
|
2051 |
+
"epoch": 68.28,
|
2052 |
+
"learning_rate": 9.648414985590777e-05,
|
2053 |
+
"loss": 0.345,
|
2054 |
+
"step": 2390
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 68.57,
|
2058 |
+
"learning_rate": 9.561959654178673e-05,
|
2059 |
+
"loss": 0.3349,
|
2060 |
+
"step": 2400
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 68.85,
|
2064 |
+
"learning_rate": 9.475504322766569e-05,
|
2065 |
+
"loss": 0.381,
|
2066 |
+
"step": 2410
|
2067 |
+
},
|
2068 |
+
{
|
2069 |
+
"epoch": 68.99,
|
2070 |
+
"eval_loss": 0.7306443452835083,
|
2071 |
+
"eval_runtime": 8.8012,
|
2072 |
+
"eval_samples_per_second": 7.158,
|
2073 |
+
"eval_steps_per_second": 1.818,
|
2074 |
+
"eval_wer": 0.3513285533530156,
|
2075 |
+
"step": 2415
|
2076 |
+
},
|
2077 |
+
{
|
2078 |
+
"epoch": 69.14,
|
2079 |
+
"learning_rate": 9.389048991354465e-05,
|
2080 |
+
"loss": 0.3782,
|
2081 |
+
"step": 2420
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 69.43,
|
2085 |
+
"learning_rate": 9.302593659942363e-05,
|
2086 |
+
"loss": 0.3198,
|
2087 |
+
"step": 2430
|
2088 |
+
},
|
2089 |
+
{
|
2090 |
+
"epoch": 69.71,
|
2091 |
+
"learning_rate": 9.216138328530259e-05,
|
2092 |
+
"loss": 0.3742,
|
2093 |
+
"step": 2440
|
2094 |
+
},
|
2095 |
+
{
|
2096 |
+
"epoch": 69.99,
|
2097 |
+
"learning_rate": 9.129682997118155e-05,
|
2098 |
+
"loss": 0.3319,
|
2099 |
+
"step": 2450
|
2100 |
+
},
|
2101 |
+
{
|
2102 |
+
"epoch": 69.99,
|
2103 |
+
"eval_loss": 0.8494887948036194,
|
2104 |
+
"eval_runtime": 8.8477,
|
2105 |
+
"eval_samples_per_second": 7.12,
|
2106 |
+
"eval_steps_per_second": 1.808,
|
2107 |
+
"eval_wer": 0.34837621256853646,
|
2108 |
+
"step": 2450
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 70.28,
|
2112 |
+
"learning_rate": 9.043227665706052e-05,
|
2113 |
+
"loss": 0.3632,
|
2114 |
+
"step": 2460
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 70.57,
|
2118 |
+
"learning_rate": 8.956772334293948e-05,
|
2119 |
+
"loss": 0.2967,
|
2120 |
+
"step": 2470
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 70.85,
|
2124 |
+
"learning_rate": 8.870317002881844e-05,
|
2125 |
+
"loss": 0.3552,
|
2126 |
+
"step": 2480
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 70.99,
|
2130 |
+
"eval_loss": 0.7546125054359436,
|
2131 |
+
"eval_runtime": 8.821,
|
2132 |
+
"eval_samples_per_second": 7.142,
|
2133 |
+
"eval_steps_per_second": 1.814,
|
2134 |
+
"eval_wer": 0.3551244200759173,
|
2135 |
+
"step": 2485
|
2136 |
+
},
|
2137 |
+
{
|
2138 |
+
"epoch": 71.14,
|
2139 |
+
"learning_rate": 8.783861671469738e-05,
|
2140 |
+
"loss": 0.3304,
|
2141 |
+
"step": 2490
|
2142 |
+
},
|
2143 |
+
{
|
2144 |
+
"epoch": 71.43,
|
2145 |
+
"learning_rate": 8.697406340057636e-05,
|
2146 |
+
"loss": 0.3718,
|
2147 |
+
"step": 2500
|
2148 |
+
},
|
2149 |
+
{
|
2150 |
+
"epoch": 71.71,
|
2151 |
+
"learning_rate": 8.610951008645532e-05,
|
2152 |
+
"loss": 0.3271,
|
2153 |
+
"step": 2510
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 71.99,
|
2157 |
+
"learning_rate": 8.524495677233428e-05,
|
2158 |
+
"loss": 0.3292,
|
2159 |
+
"step": 2520
|
2160 |
+
},
|
2161 |
+
{
|
2162 |
+
"epoch": 71.99,
|
2163 |
+
"eval_loss": 0.7483251094818115,
|
2164 |
+
"eval_runtime": 8.83,
|
2165 |
+
"eval_samples_per_second": 7.135,
|
2166 |
+
"eval_steps_per_second": 1.812,
|
2167 |
+
"eval_wer": 0.34500210881484605,
|
2168 |
+
"step": 2520
|
2169 |
+
},
|
2170 |
+
{
|
2171 |
+
"epoch": 72.28,
|
2172 |
+
"learning_rate": 8.438040345821325e-05,
|
2173 |
+
"loss": 0.4341,
|
2174 |
+
"step": 2530
|
2175 |
+
},
|
2176 |
+
{
|
2177 |
+
"epoch": 72.57,
|
2178 |
+
"learning_rate": 8.351585014409221e-05,
|
2179 |
+
"loss": 0.3173,
|
2180 |
+
"step": 2540
|
2181 |
+
},
|
2182 |
+
{
|
2183 |
+
"epoch": 72.85,
|
2184 |
+
"learning_rate": 8.265129682997117e-05,
|
2185 |
+
"loss": 0.3041,
|
2186 |
+
"step": 2550
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 72.99,
|
2190 |
+
"eval_loss": 0.7304584383964539,
|
2191 |
+
"eval_runtime": 8.8882,
|
2192 |
+
"eval_samples_per_second": 7.088,
|
2193 |
+
"eval_steps_per_second": 1.8,
|
2194 |
+
"eval_wer": 0.35217207929143823,
|
2195 |
+
"step": 2555
|
2196 |
+
},
|
2197 |
+
{
|
2198 |
+
"epoch": 73.14,
|
2199 |
+
"learning_rate": 8.178674351585014e-05,
|
2200 |
+
"loss": 0.345,
|
2201 |
+
"step": 2560
|
2202 |
+
},
|
2203 |
+
{
|
2204 |
+
"epoch": 73.43,
|
2205 |
+
"learning_rate": 8.09221902017291e-05,
|
2206 |
+
"loss": 0.3132,
|
2207 |
+
"step": 2570
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 73.71,
|
2211 |
+
"learning_rate": 8.005763688760806e-05,
|
2212 |
+
"loss": 0.3838,
|
2213 |
+
"step": 2580
|
2214 |
+
},
|
2215 |
+
{
|
2216 |
+
"epoch": 73.99,
|
2217 |
+
"learning_rate": 7.919308357348704e-05,
|
2218 |
+
"loss": 0.3606,
|
2219 |
+
"step": 2590
|
2220 |
+
},
|
2221 |
+
{
|
2222 |
+
"epoch": 73.99,
|
2223 |
+
"eval_loss": 0.7358484864234924,
|
2224 |
+
"eval_runtime": 8.8826,
|
2225 |
+
"eval_samples_per_second": 7.093,
|
2226 |
+
"eval_steps_per_second": 1.801,
|
2227 |
+
"eval_wer": 0.34837621256853646,
|
2228 |
+
"step": 2590
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 74.28,
|
2232 |
+
"learning_rate": 7.832853025936598e-05,
|
2233 |
+
"loss": 0.3354,
|
2234 |
+
"step": 2600
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 74.57,
|
2238 |
+
"learning_rate": 7.746397694524494e-05,
|
2239 |
+
"loss": 0.319,
|
2240 |
+
"step": 2610
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 74.85,
|
2244 |
+
"learning_rate": 7.65994236311239e-05,
|
2245 |
+
"loss": 0.3629,
|
2246 |
+
"step": 2620
|
2247 |
+
},
|
2248 |
+
{
|
2249 |
+
"epoch": 74.99,
|
2250 |
+
"eval_loss": 0.7708744406700134,
|
2251 |
+
"eval_runtime": 8.8268,
|
2252 |
+
"eval_samples_per_second": 7.137,
|
2253 |
+
"eval_steps_per_second": 1.813,
|
2254 |
+
"eval_wer": 0.34458034584563474,
|
2255 |
+
"step": 2625
|
2256 |
+
},
|
2257 |
+
{
|
2258 |
+
"epoch": 75.14,
|
2259 |
+
"learning_rate": 7.573487031700288e-05,
|
2260 |
+
"loss": 0.3404,
|
2261 |
+
"step": 2630
|
2262 |
+
},
|
2263 |
+
{
|
2264 |
+
"epoch": 75.43,
|
2265 |
+
"learning_rate": 7.487031700288184e-05,
|
2266 |
+
"loss": 0.3678,
|
2267 |
+
"step": 2640
|
2268 |
+
},
|
2269 |
+
{
|
2270 |
+
"epoch": 75.71,
|
2271 |
+
"learning_rate": 7.40057636887608e-05,
|
2272 |
+
"loss": 0.2948,
|
2273 |
+
"step": 2650
|
2274 |
+
},
|
2275 |
+
{
|
2276 |
+
"epoch": 75.99,
|
2277 |
+
"learning_rate": 7.314121037463977e-05,
|
2278 |
+
"loss": 0.3409,
|
2279 |
+
"step": 2660
|
2280 |
+
},
|
2281 |
+
{
|
2282 |
+
"epoch": 75.99,
|
2283 |
+
"eval_loss": 0.756772518157959,
|
2284 |
+
"eval_runtime": 8.8295,
|
2285 |
+
"eval_samples_per_second": 7.135,
|
2286 |
+
"eval_steps_per_second": 1.812,
|
2287 |
+
"eval_wer": 0.3584985238296078,
|
2288 |
+
"step": 2660
|
2289 |
+
},
|
2290 |
+
{
|
2291 |
+
"epoch": 76.28,
|
2292 |
+
"learning_rate": 7.227665706051873e-05,
|
2293 |
+
"loss": 0.3575,
|
2294 |
+
"step": 2670
|
2295 |
+
},
|
2296 |
+
{
|
2297 |
+
"epoch": 76.57,
|
2298 |
+
"learning_rate": 7.141210374639769e-05,
|
2299 |
+
"loss": 0.3195,
|
2300 |
+
"step": 2680
|
2301 |
+
},
|
2302 |
+
{
|
2303 |
+
"epoch": 76.85,
|
2304 |
+
"learning_rate": 7.054755043227665e-05,
|
2305 |
+
"loss": 0.3315,
|
2306 |
+
"step": 2690
|
2307 |
+
},
|
2308 |
+
{
|
2309 |
+
"epoch": 76.99,
|
2310 |
+
"eval_loss": 0.7466398477554321,
|
2311 |
+
"eval_runtime": 8.8563,
|
2312 |
+
"eval_samples_per_second": 7.114,
|
2313 |
+
"eval_steps_per_second": 1.807,
|
2314 |
+
"eval_wer": 0.3475326866301139,
|
2315 |
+
"step": 2695
|
2316 |
+
},
|
2317 |
+
{
|
2318 |
+
"epoch": 77.14,
|
2319 |
+
"learning_rate": 6.968299711815561e-05,
|
2320 |
+
"loss": 0.334,
|
2321 |
+
"step": 2700
|
2322 |
+
},
|
2323 |
+
{
|
2324 |
+
"epoch": 77.43,
|
2325 |
+
"learning_rate": 6.881844380403457e-05,
|
2326 |
+
"loss": 0.2933,
|
2327 |
+
"step": 2710
|
2328 |
+
},
|
2329 |
+
{
|
2330 |
+
"epoch": 77.71,
|
2331 |
+
"learning_rate": 6.795389048991354e-05,
|
2332 |
+
"loss": 0.399,
|
2333 |
+
"step": 2720
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 77.99,
|
2337 |
+
"learning_rate": 6.70893371757925e-05,
|
2338 |
+
"loss": 0.2934,
|
2339 |
+
"step": 2730
|
2340 |
+
},
|
2341 |
+
{
|
2342 |
+
"epoch": 77.99,
|
2343 |
+
"eval_loss": 0.7351065874099731,
|
2344 |
+
"eval_runtime": 8.8959,
|
2345 |
+
"eval_samples_per_second": 7.082,
|
2346 |
+
"eval_steps_per_second": 1.799,
|
2347 |
+
"eval_wer": 0.3496415014761704,
|
2348 |
+
"step": 2730
|
2349 |
+
},
|
2350 |
+
{
|
2351 |
+
"epoch": 78.28,
|
2352 |
+
"learning_rate": 6.622478386167146e-05,
|
2353 |
+
"loss": 0.3346,
|
2354 |
+
"step": 2740
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 78.57,
|
2358 |
+
"learning_rate": 6.536023054755042e-05,
|
2359 |
+
"loss": 0.3117,
|
2360 |
+
"step": 2750
|
2361 |
+
},
|
2362 |
+
{
|
2363 |
+
"epoch": 78.85,
|
2364 |
+
"learning_rate": 6.449567723342938e-05,
|
2365 |
+
"loss": 0.3366,
|
2366 |
+
"step": 2760
|
2367 |
+
},
|
2368 |
+
{
|
2369 |
+
"epoch": 78.99,
|
2370 |
+
"eval_loss": 0.8014491200447083,
|
2371 |
+
"eval_runtime": 8.8741,
|
2372 |
+
"eval_samples_per_second": 7.099,
|
2373 |
+
"eval_steps_per_second": 1.803,
|
2374 |
+
"eval_wer": 0.34837621256853646,
|
2375 |
+
"step": 2765
|
2376 |
+
},
|
2377 |
+
{
|
2378 |
+
"epoch": 79.14,
|
2379 |
+
"learning_rate": 6.363112391930836e-05,
|
2380 |
+
"loss": 0.343,
|
2381 |
+
"step": 2770
|
2382 |
+
},
|
2383 |
+
{
|
2384 |
+
"epoch": 79.43,
|
2385 |
+
"learning_rate": 6.276657060518732e-05,
|
2386 |
+
"loss": 0.3168,
|
2387 |
+
"step": 2780
|
2388 |
+
},
|
2389 |
+
{
|
2390 |
+
"epoch": 79.71,
|
2391 |
+
"learning_rate": 6.190201729106627e-05,
|
2392 |
+
"loss": 0.341,
|
2393 |
+
"step": 2790
|
2394 |
+
},
|
2395 |
+
{
|
2396 |
+
"epoch": 79.99,
|
2397 |
+
"learning_rate": 6.103746397694523e-05,
|
2398 |
+
"loss": 0.3176,
|
2399 |
+
"step": 2800
|
2400 |
+
},
|
2401 |
+
{
|
2402 |
+
"epoch": 79.99,
|
2403 |
+
"eval_loss": 0.8014456629753113,
|
2404 |
+
"eval_runtime": 8.9378,
|
2405 |
+
"eval_samples_per_second": 7.049,
|
2406 |
+
"eval_steps_per_second": 1.79,
|
2407 |
+
"eval_wer": 0.3420497680303669,
|
2408 |
+
"step": 2800
|
2409 |
+
},
|
2410 |
+
{
|
2411 |
+
"epoch": 80.28,
|
2412 |
+
"learning_rate": 6.01729106628242e-05,
|
2413 |
+
"loss": 0.3027,
|
2414 |
+
"step": 2810
|
2415 |
+
},
|
2416 |
+
{
|
2417 |
+
"epoch": 80.57,
|
2418 |
+
"learning_rate": 5.930835734870316e-05,
|
2419 |
+
"loss": 0.3063,
|
2420 |
+
"step": 2820
|
2421 |
+
},
|
2422 |
+
{
|
2423 |
+
"epoch": 80.85,
|
2424 |
+
"learning_rate": 5.844380403458213e-05,
|
2425 |
+
"loss": 0.3319,
|
2426 |
+
"step": 2830
|
2427 |
+
},
|
2428 |
+
{
|
2429 |
+
"epoch": 80.99,
|
2430 |
+
"eval_loss": 0.799625039100647,
|
2431 |
+
"eval_runtime": 8.8579,
|
2432 |
+
"eval_samples_per_second": 7.112,
|
2433 |
+
"eval_steps_per_second": 1.806,
|
2434 |
+
"eval_wer": 0.34373681990721217,
|
2435 |
+
"step": 2835
|
2436 |
+
},
|
2437 |
+
{
|
2438 |
+
"epoch": 81.14,
|
2439 |
+
"learning_rate": 5.7579250720461094e-05,
|
2440 |
+
"loss": 0.3286,
|
2441 |
+
"step": 2840
|
2442 |
+
},
|
2443 |
+
{
|
2444 |
+
"epoch": 81.43,
|
2445 |
+
"learning_rate": 5.6714697406340054e-05,
|
2446 |
+
"loss": 0.3039,
|
2447 |
+
"step": 2850
|
2448 |
+
},
|
2449 |
+
{
|
2450 |
+
"epoch": 81.71,
|
2451 |
+
"learning_rate": 5.5850144092219014e-05,
|
2452 |
+
"loss": 0.3028,
|
2453 |
+
"step": 2860
|
2454 |
+
},
|
2455 |
+
{
|
2456 |
+
"epoch": 81.99,
|
2457 |
+
"learning_rate": 5.4985590778097974e-05,
|
2458 |
+
"loss": 0.2967,
|
2459 |
+
"step": 2870
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 81.99,
|
2463 |
+
"eval_loss": 0.8156365156173706,
|
2464 |
+
"eval_runtime": 8.901,
|
2465 |
+
"eval_samples_per_second": 7.078,
|
2466 |
+
"eval_steps_per_second": 1.798,
|
2467 |
+
"eval_wer": 0.34120624209194433,
|
2468 |
+
"step": 2870
|
2469 |
+
},
|
2470 |
+
{
|
2471 |
+
"epoch": 82.28,
|
2472 |
+
"learning_rate": 5.412103746397694e-05,
|
2473 |
+
"loss": 0.327,
|
2474 |
+
"step": 2880
|
2475 |
+
},
|
2476 |
+
{
|
2477 |
+
"epoch": 82.57,
|
2478 |
+
"learning_rate": 5.325648414985591e-05,
|
2479 |
+
"loss": 0.2834,
|
2480 |
+
"step": 2890
|
2481 |
+
},
|
2482 |
+
{
|
2483 |
+
"epoch": 82.85,
|
2484 |
+
"learning_rate": 5.239193083573487e-05,
|
2485 |
+
"loss": 0.3137,
|
2486 |
+
"step": 2900
|
2487 |
+
},
|
2488 |
+
{
|
2489 |
+
"epoch": 82.99,
|
2490 |
+
"eval_loss": 0.8024904131889343,
|
2491 |
+
"eval_runtime": 8.8882,
|
2492 |
+
"eval_samples_per_second": 7.088,
|
2493 |
+
"eval_steps_per_second": 1.8,
|
2494 |
+
"eval_wer": 0.33614508646140867,
|
2495 |
+
"step": 2905
|
2496 |
+
},
|
2497 |
+
{
|
2498 |
+
"epoch": 83.14,
|
2499 |
+
"learning_rate": 5.152737752161383e-05,
|
2500 |
+
"loss": 0.2981,
|
2501 |
+
"step": 2910
|
2502 |
+
},
|
2503 |
+
{
|
2504 |
+
"epoch": 83.43,
|
2505 |
+
"learning_rate": 5.066282420749279e-05,
|
2506 |
+
"loss": 0.2669,
|
2507 |
+
"step": 2920
|
2508 |
+
},
|
2509 |
+
{
|
2510 |
+
"epoch": 83.71,
|
2511 |
+
"learning_rate": 4.979827089337175e-05,
|
2512 |
+
"loss": 0.3344,
|
2513 |
+
"step": 2930
|
2514 |
+
},
|
2515 |
+
{
|
2516 |
+
"epoch": 83.99,
|
2517 |
+
"learning_rate": 4.893371757925072e-05,
|
2518 |
+
"loss": 0.3133,
|
2519 |
+
"step": 2940
|
2520 |
+
},
|
2521 |
+
{
|
2522 |
+
"epoch": 83.99,
|
2523 |
+
"eval_loss": 0.778435230255127,
|
2524 |
+
"eval_runtime": 8.8669,
|
2525 |
+
"eval_samples_per_second": 7.105,
|
2526 |
+
"eval_steps_per_second": 1.804,
|
2527 |
+
"eval_wer": 0.34162800506115565,
|
2528 |
+
"step": 2940
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 84.28,
|
2532 |
+
"learning_rate": 4.806916426512968e-05,
|
2533 |
+
"loss": 0.2994,
|
2534 |
+
"step": 2950
|
2535 |
+
},
|
2536 |
+
{
|
2537 |
+
"epoch": 84.57,
|
2538 |
+
"learning_rate": 4.7204610951008646e-05,
|
2539 |
+
"loss": 0.2719,
|
2540 |
+
"step": 2960
|
2541 |
+
},
|
2542 |
+
{
|
2543 |
+
"epoch": 84.85,
|
2544 |
+
"learning_rate": 4.63400576368876e-05,
|
2545 |
+
"loss": 0.3134,
|
2546 |
+
"step": 2970
|
2547 |
+
},
|
2548 |
+
{
|
2549 |
+
"epoch": 84.99,
|
2550 |
+
"eval_loss": 0.7894186973571777,
|
2551 |
+
"eval_runtime": 8.8795,
|
2552 |
+
"eval_samples_per_second": 7.095,
|
2553 |
+
"eval_steps_per_second": 1.802,
|
2554 |
+
"eval_wer": 0.3336145086461409,
|
2555 |
+
"step": 2975
|
2556 |
+
},
|
2557 |
+
{
|
2558 |
+
"epoch": 85.14,
|
2559 |
+
"learning_rate": 4.5475504322766566e-05,
|
2560 |
+
"loss": 0.2941,
|
2561 |
+
"step": 2980
|
2562 |
+
},
|
2563 |
+
{
|
2564 |
+
"epoch": 85.43,
|
2565 |
+
"learning_rate": 4.4610951008645526e-05,
|
2566 |
+
"loss": 0.2841,
|
2567 |
+
"step": 2990
|
2568 |
+
},
|
2569 |
+
{
|
2570 |
+
"epoch": 85.71,
|
2571 |
+
"learning_rate": 4.374639769452449e-05,
|
2572 |
+
"loss": 0.3184,
|
2573 |
+
"step": 3000
|
2574 |
+
},
|
2575 |
+
{
|
2576 |
+
"epoch": 85.99,
|
2577 |
+
"learning_rate": 4.288184438040346e-05,
|
2578 |
+
"loss": 0.3216,
|
2579 |
+
"step": 3010
|
2580 |
+
},
|
2581 |
+
{
|
2582 |
+
"epoch": 85.99,
|
2583 |
+
"eval_loss": 0.8330649137496948,
|
2584 |
+
"eval_runtime": 8.8914,
|
2585 |
+
"eval_samples_per_second": 7.086,
|
2586 |
+
"eval_steps_per_second": 1.799,
|
2587 |
+
"eval_wer": 0.33951919021509913,
|
2588 |
+
"step": 3010
|
2589 |
+
},
|
2590 |
+
{
|
2591 |
+
"epoch": 86.28,
|
2592 |
+
"learning_rate": 4.201729106628241e-05,
|
2593 |
+
"loss": 0.3181,
|
2594 |
+
"step": 3020
|
2595 |
+
},
|
2596 |
+
{
|
2597 |
+
"epoch": 86.57,
|
2598 |
+
"learning_rate": 4.115273775216138e-05,
|
2599 |
+
"loss": 0.289,
|
2600 |
+
"step": 3030
|
2601 |
+
},
|
2602 |
+
{
|
2603 |
+
"epoch": 86.85,
|
2604 |
+
"learning_rate": 4.028818443804034e-05,
|
2605 |
+
"loss": 0.365,
|
2606 |
+
"step": 3040
|
2607 |
+
},
|
2608 |
+
{
|
2609 |
+
"epoch": 86.99,
|
2610 |
+
"eval_loss": 0.7980260252952576,
|
2611 |
+
"eval_runtime": 8.8324,
|
2612 |
+
"eval_samples_per_second": 7.133,
|
2613 |
+
"eval_steps_per_second": 1.812,
|
2614 |
+
"eval_wer": 0.3353015605229861,
|
2615 |
+
"step": 3045
|
2616 |
+
},
|
2617 |
+
{
|
2618 |
+
"epoch": 87.14,
|
2619 |
+
"learning_rate": 3.9423631123919305e-05,
|
2620 |
+
"loss": 0.3463,
|
2621 |
+
"step": 3050
|
2622 |
+
},
|
2623 |
+
{
|
2624 |
+
"epoch": 87.43,
|
2625 |
+
"learning_rate": 3.855907780979827e-05,
|
2626 |
+
"loss": 0.3385,
|
2627 |
+
"step": 3060
|
2628 |
+
},
|
2629 |
+
{
|
2630 |
+
"epoch": 87.71,
|
2631 |
+
"learning_rate": 3.769452449567723e-05,
|
2632 |
+
"loss": 0.2897,
|
2633 |
+
"step": 3070
|
2634 |
+
},
|
2635 |
+
{
|
2636 |
+
"epoch": 87.99,
|
2637 |
+
"learning_rate": 3.682997118155619e-05,
|
2638 |
+
"loss": 0.2962,
|
2639 |
+
"step": 3080
|
2640 |
+
},
|
2641 |
+
{
|
2642 |
+
"epoch": 87.99,
|
2643 |
+
"eval_loss": 0.7964635491371155,
|
2644 |
+
"eval_runtime": 8.8283,
|
2645 |
+
"eval_samples_per_second": 7.136,
|
2646 |
+
"eval_steps_per_second": 1.812,
|
2647 |
+
"eval_wer": 0.3403627161535217,
|
2648 |
+
"step": 3080
|
2649 |
+
},
|
2650 |
+
{
|
2651 |
+
"epoch": 88.28,
|
2652 |
+
"learning_rate": 3.596541786743515e-05,
|
2653 |
+
"loss": 0.3322,
|
2654 |
+
"step": 3090
|
2655 |
+
},
|
2656 |
+
{
|
2657 |
+
"epoch": 88.57,
|
2658 |
+
"learning_rate": 3.510086455331412e-05,
|
2659 |
+
"loss": 0.2878,
|
2660 |
+
"step": 3100
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 88.85,
|
2664 |
+
"learning_rate": 3.4236311239193085e-05,
|
2665 |
+
"loss": 0.3126,
|
2666 |
+
"step": 3110
|
2667 |
+
},
|
2668 |
+
{
|
2669 |
+
"epoch": 88.99,
|
2670 |
+
"eval_loss": 0.7469554543495178,
|
2671 |
+
"eval_runtime": 8.8328,
|
2672 |
+
"eval_samples_per_second": 7.133,
|
2673 |
+
"eval_steps_per_second": 1.811,
|
2674 |
+
"eval_wer": 0.3420497680303669,
|
2675 |
+
"step": 3115
|
2676 |
+
},
|
2677 |
+
{
|
2678 |
+
"epoch": 89.14,
|
2679 |
+
"learning_rate": 3.3371757925072045e-05,
|
2680 |
+
"loss": 0.3497,
|
2681 |
+
"step": 3120
|
2682 |
+
},
|
2683 |
+
{
|
2684 |
+
"epoch": 89.43,
|
2685 |
+
"learning_rate": 3.2507204610951004e-05,
|
2686 |
+
"loss": 0.3006,
|
2687 |
+
"step": 3130
|
2688 |
+
},
|
2689 |
+
{
|
2690 |
+
"epoch": 89.71,
|
2691 |
+
"learning_rate": 3.164265129682997e-05,
|
2692 |
+
"loss": 0.3237,
|
2693 |
+
"step": 3140
|
2694 |
+
},
|
2695 |
+
{
|
2696 |
+
"epoch": 89.99,
|
2697 |
+
"learning_rate": 3.077809798270893e-05,
|
2698 |
+
"loss": 0.2843,
|
2699 |
+
"step": 3150
|
2700 |
+
},
|
2701 |
+
{
|
2702 |
+
"epoch": 89.99,
|
2703 |
+
"eval_loss": 0.7787944078445435,
|
2704 |
+
"eval_runtime": 8.8054,
|
2705 |
+
"eval_samples_per_second": 7.155,
|
2706 |
+
"eval_steps_per_second": 1.817,
|
2707 |
+
"eval_wer": 0.3403627161535217,
|
2708 |
+
"step": 3150
|
2709 |
+
},
|
2710 |
+
{
|
2711 |
+
"epoch": 90.28,
|
2712 |
+
"learning_rate": 2.9913544668587894e-05,
|
2713 |
+
"loss": 0.3021,
|
2714 |
+
"step": 3160
|
2715 |
+
},
|
2716 |
+
{
|
2717 |
+
"epoch": 90.57,
|
2718 |
+
"learning_rate": 2.9048991354466857e-05,
|
2719 |
+
"loss": 0.2765,
|
2720 |
+
"step": 3170
|
2721 |
+
},
|
2722 |
+
{
|
2723 |
+
"epoch": 90.85,
|
2724 |
+
"learning_rate": 2.8184438040345817e-05,
|
2725 |
+
"loss": 0.2967,
|
2726 |
+
"step": 3180
|
2727 |
+
},
|
2728 |
+
{
|
2729 |
+
"epoch": 90.99,
|
2730 |
+
"eval_loss": 0.790235698223114,
|
2731 |
+
"eval_runtime": 8.8387,
|
2732 |
+
"eval_samples_per_second": 7.128,
|
2733 |
+
"eval_steps_per_second": 1.81,
|
2734 |
+
"eval_wer": 0.3374103753690426,
|
2735 |
+
"step": 3185
|
2736 |
+
},
|
2737 |
+
{
|
2738 |
+
"epoch": 91.14,
|
2739 |
+
"learning_rate": 2.731988472622478e-05,
|
2740 |
+
"loss": 0.2985,
|
2741 |
+
"step": 3190
|
2742 |
+
},
|
2743 |
+
{
|
2744 |
+
"epoch": 91.43,
|
2745 |
+
"learning_rate": 2.6455331412103747e-05,
|
2746 |
+
"loss": 0.2949,
|
2747 |
+
"step": 3200
|
2748 |
+
},
|
2749 |
+
{
|
2750 |
+
"epoch": 91.71,
|
2751 |
+
"learning_rate": 2.5590778097982707e-05,
|
2752 |
+
"loss": 0.2924,
|
2753 |
+
"step": 3210
|
2754 |
+
},
|
2755 |
+
{
|
2756 |
+
"epoch": 91.99,
|
2757 |
+
"learning_rate": 2.472622478386167e-05,
|
2758 |
+
"loss": 0.3171,
|
2759 |
+
"step": 3220
|
2760 |
+
},
|
2761 |
+
{
|
2762 |
+
"epoch": 91.99,
|
2763 |
+
"eval_loss": 0.8021612167358398,
|
2764 |
+
"eval_runtime": 8.9042,
|
2765 |
+
"eval_samples_per_second": 7.075,
|
2766 |
+
"eval_steps_per_second": 1.797,
|
2767 |
+
"eval_wer": 0.3403627161535217,
|
2768 |
+
"step": 3220
|
2769 |
+
},
|
2770 |
+
{
|
2771 |
+
"epoch": 92.28,
|
2772 |
+
"learning_rate": 2.386167146974063e-05,
|
2773 |
+
"loss": 0.2985,
|
2774 |
+
"step": 3230
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 92.57,
|
2778 |
+
"learning_rate": 2.2997118155619593e-05,
|
2779 |
+
"loss": 0.2964,
|
2780 |
+
"step": 3240
|
2781 |
+
},
|
2782 |
+
{
|
2783 |
+
"epoch": 92.85,
|
2784 |
+
"learning_rate": 2.213256484149856e-05,
|
2785 |
+
"loss": 0.3069,
|
2786 |
+
"step": 3250
|
2787 |
+
},
|
2788 |
+
{
|
2789 |
+
"epoch": 92.99,
|
2790 |
+
"eval_loss": 0.7999234795570374,
|
2791 |
+
"eval_runtime": 8.9491,
|
2792 |
+
"eval_samples_per_second": 7.04,
|
2793 |
+
"eval_steps_per_second": 1.788,
|
2794 |
+
"eval_wer": 0.33445803458456347,
|
2795 |
+
"step": 3255
|
2796 |
+
},
|
2797 |
+
{
|
2798 |
+
"epoch": 93.14,
|
2799 |
+
"learning_rate": 2.126801152737752e-05,
|
2800 |
+
"loss": 0.3008,
|
2801 |
+
"step": 3260
|
2802 |
+
},
|
2803 |
+
{
|
2804 |
+
"epoch": 93.43,
|
2805 |
+
"learning_rate": 2.0403458213256483e-05,
|
2806 |
+
"loss": 0.3183,
|
2807 |
+
"step": 3270
|
2808 |
+
},
|
2809 |
+
{
|
2810 |
+
"epoch": 93.71,
|
2811 |
+
"learning_rate": 1.9538904899135446e-05,
|
2812 |
+
"loss": 0.2784,
|
2813 |
+
"step": 3280
|
2814 |
+
},
|
2815 |
+
{
|
2816 |
+
"epoch": 93.99,
|
2817 |
+
"learning_rate": 1.867435158501441e-05,
|
2818 |
+
"loss": 0.3571,
|
2819 |
+
"step": 3290
|
2820 |
+
},
|
2821 |
+
{
|
2822 |
+
"epoch": 93.99,
|
2823 |
+
"eval_loss": 0.7896485328674316,
|
2824 |
+
"eval_runtime": 8.8589,
|
2825 |
+
"eval_samples_per_second": 7.111,
|
2826 |
+
"eval_steps_per_second": 1.806,
|
2827 |
+
"eval_wer": 0.3403627161535217,
|
2828 |
+
"step": 3290
|
2829 |
+
},
|
2830 |
+
{
|
2831 |
+
"epoch": 94.28,
|
2832 |
+
"learning_rate": 1.780979827089337e-05,
|
2833 |
+
"loss": 0.2793,
|
2834 |
+
"step": 3300
|
2835 |
+
},
|
2836 |
+
{
|
2837 |
+
"epoch": 94.57,
|
2838 |
+
"learning_rate": 1.6945244956772332e-05,
|
2839 |
+
"loss": 0.3032,
|
2840 |
+
"step": 3310
|
2841 |
+
},
|
2842 |
+
{
|
2843 |
+
"epoch": 94.85,
|
2844 |
+
"learning_rate": 1.6080691642651295e-05,
|
2845 |
+
"loss": 0.2805,
|
2846 |
+
"step": 3320
|
2847 |
+
},
|
2848 |
+
{
|
2849 |
+
"epoch": 94.99,
|
2850 |
+
"eval_loss": 0.783069908618927,
|
2851 |
+
"eval_runtime": 8.8967,
|
2852 |
+
"eval_samples_per_second": 7.081,
|
2853 |
+
"eval_steps_per_second": 1.798,
|
2854 |
+
"eval_wer": 0.3390974272458878,
|
2855 |
+
"step": 3325
|
2856 |
+
},
|
2857 |
+
{
|
2858 |
+
"epoch": 95.14,
|
2859 |
+
"learning_rate": 1.5216138328530259e-05,
|
2860 |
+
"loss": 0.2734,
|
2861 |
+
"step": 3330
|
2862 |
+
},
|
2863 |
+
{
|
2864 |
+
"epoch": 95.43,
|
2865 |
+
"learning_rate": 1.435158501440922e-05,
|
2866 |
+
"loss": 0.2819,
|
2867 |
+
"step": 3340
|
2868 |
+
},
|
2869 |
+
{
|
2870 |
+
"epoch": 95.71,
|
2871 |
+
"learning_rate": 1.3487031700288182e-05,
|
2872 |
+
"loss": 0.3385,
|
2873 |
+
"step": 3350
|
2874 |
+
},
|
2875 |
+
{
|
2876 |
+
"epoch": 95.99,
|
2877 |
+
"learning_rate": 1.2622478386167147e-05,
|
2878 |
+
"loss": 0.3099,
|
2879 |
+
"step": 3360
|
2880 |
+
},
|
2881 |
+
{
|
2882 |
+
"epoch": 95.99,
|
2883 |
+
"eval_loss": 0.790939450263977,
|
2884 |
+
"eval_runtime": 8.9444,
|
2885 |
+
"eval_samples_per_second": 7.043,
|
2886 |
+
"eval_steps_per_second": 1.789,
|
2887 |
+
"eval_wer": 0.33656684943062,
|
2888 |
+
"step": 3360
|
2889 |
+
},
|
2890 |
+
{
|
2891 |
+
"epoch": 96.28,
|
2892 |
+
"learning_rate": 1.1757925072046108e-05,
|
2893 |
+
"loss": 0.3293,
|
2894 |
+
"step": 3370
|
2895 |
+
},
|
2896 |
+
{
|
2897 |
+
"epoch": 96.57,
|
2898 |
+
"learning_rate": 1.0893371757925071e-05,
|
2899 |
+
"loss": 0.2569,
|
2900 |
+
"step": 3380
|
2901 |
+
},
|
2902 |
+
{
|
2903 |
+
"epoch": 96.85,
|
2904 |
+
"learning_rate": 1.0028818443804033e-05,
|
2905 |
+
"loss": 0.2868,
|
2906 |
+
"step": 3390
|
2907 |
+
},
|
2908 |
+
{
|
2909 |
+
"epoch": 96.99,
|
2910 |
+
"eval_loss": 0.7918463945388794,
|
2911 |
+
"eval_runtime": 9.0538,
|
2912 |
+
"eval_samples_per_second": 6.958,
|
2913 |
+
"eval_steps_per_second": 1.767,
|
2914 |
+
"eval_wer": 0.33951919021509913,
|
2915 |
+
"step": 3395
|
2916 |
+
},
|
2917 |
+
{
|
2918 |
+
"epoch": 97.14,
|
2919 |
+
"learning_rate": 9.164265129682996e-06,
|
2920 |
+
"loss": 0.299,
|
2921 |
+
"step": 3400
|
2922 |
+
},
|
2923 |
+
{
|
2924 |
+
"epoch": 97.43,
|
2925 |
+
"learning_rate": 8.29971181556196e-06,
|
2926 |
+
"loss": 0.2942,
|
2927 |
+
"step": 3410
|
2928 |
+
},
|
2929 |
+
{
|
2930 |
+
"epoch": 97.71,
|
2931 |
+
"learning_rate": 7.435158501440922e-06,
|
2932 |
+
"loss": 0.2931,
|
2933 |
+
"step": 3420
|
2934 |
+
},
|
2935 |
+
{
|
2936 |
+
"epoch": 97.99,
|
2937 |
+
"learning_rate": 6.570605187319884e-06,
|
2938 |
+
"loss": 0.2626,
|
2939 |
+
"step": 3430
|
2940 |
+
},
|
2941 |
+
{
|
2942 |
+
"epoch": 97.99,
|
2943 |
+
"eval_loss": 0.776630699634552,
|
2944 |
+
"eval_runtime": 8.8695,
|
2945 |
+
"eval_samples_per_second": 7.103,
|
2946 |
+
"eval_steps_per_second": 1.804,
|
2947 |
+
"eval_wer": 0.34289329396878954,
|
2948 |
+
"step": 3430
|
2949 |
+
},
|
2950 |
+
{
|
2951 |
+
"epoch": 98.28,
|
2952 |
+
"learning_rate": 5.706051873198847e-06,
|
2953 |
+
"loss": 0.282,
|
2954 |
+
"step": 3440
|
2955 |
+
},
|
2956 |
+
{
|
2957 |
+
"epoch": 98.57,
|
2958 |
+
"learning_rate": 4.841498559077809e-06,
|
2959 |
+
"loss": 0.339,
|
2960 |
+
"step": 3450
|
2961 |
+
},
|
2962 |
+
{
|
2963 |
+
"epoch": 98.85,
|
2964 |
+
"learning_rate": 3.976945244956772e-06,
|
2965 |
+
"loss": 0.2634,
|
2966 |
+
"step": 3460
|
2967 |
+
},
|
2968 |
+
{
|
2969 |
+
"epoch": 98.99,
|
2970 |
+
"eval_loss": 0.777043879032135,
|
2971 |
+
"eval_runtime": 8.8512,
|
2972 |
+
"eval_samples_per_second": 7.118,
|
2973 |
+
"eval_steps_per_second": 1.808,
|
2974 |
+
"eval_wer": 0.34373681990721217,
|
2975 |
+
"step": 3465
|
2976 |
+
},
|
2977 |
+
{
|
2978 |
+
"epoch": 99.14,
|
2979 |
+
"learning_rate": 3.1123919308357346e-06,
|
2980 |
+
"loss": 0.3132,
|
2981 |
+
"step": 3470
|
2982 |
+
},
|
2983 |
+
{
|
2984 |
+
"epoch": 99.43,
|
2985 |
+
"learning_rate": 2.247838616714697e-06,
|
2986 |
+
"loss": 0.2755,
|
2987 |
+
"step": 3480
|
2988 |
+
},
|
2989 |
+
{
|
2990 |
+
"epoch": 99.71,
|
2991 |
+
"learning_rate": 1.38328530259366e-06,
|
2992 |
+
"loss": 0.3658,
|
2993 |
+
"step": 3490
|
2994 |
+
},
|
2995 |
+
{
|
2996 |
+
"epoch": 99.99,
|
2997 |
+
"learning_rate": 5.187319884726225e-07,
|
2998 |
+
"loss": 0.288,
|
2999 |
+
"step": 3500
|
3000 |
+
},
|
3001 |
+
{
|
3002 |
+
"epoch": 99.99,
|
3003 |
+
"eval_loss": 0.7762196063995361,
|
3004 |
+
"eval_runtime": 8.8123,
|
3005 |
+
"eval_samples_per_second": 7.149,
|
3006 |
+
"eval_steps_per_second": 1.816,
|
3007 |
+
"eval_wer": 0.34289329396878954,
|
3008 |
+
"step": 3500
|
3009 |
+
},
|
3010 |
+
{
|
3011 |
+
"epoch": 99.99,
|
3012 |
+
"step": 3500,
|
3013 |
+
"total_flos": 5.739635222597828e+18,
|
3014 |
+
"train_loss": 0.4800486422266279,
|
3015 |
+
"train_runtime": 12025.3803,
|
3016 |
+
"train_samples_per_second": 4.682,
|
3017 |
+
"train_steps_per_second": 0.291
|
3018 |
+
}
|
3019 |
+
],
|
3020 |
+
"max_steps": 3500,
|
3021 |
+
"num_train_epochs": 100,
|
3022 |
+
"total_flos": 5.739635222597828e+18,
|
3023 |
+
"trial_name": null,
|
3024 |
+
"trial_params": null
|
3025 |
+
}
|