End of training
Browse files- README.md +4 -4
- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +1510 -0
README.md
CHANGED
|
@@ -23,7 +23,7 @@ model-index:
|
|
| 23 |
metrics:
|
| 24 |
- name: Wer
|
| 25 |
type: wer
|
| 26 |
-
value: 0.
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -33,9 +33,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [mouseyy/result_data_2-3](https://huggingface.co/mouseyy/result_data_2-3) on the common_voice_17_0 dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
-
- Wer: 0.
|
| 38 |
-
- Cer: 0.
|
| 39 |
|
| 40 |
## Model description
|
| 41 |
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Wer
|
| 25 |
type: wer
|
| 26 |
+
value: 0.2984287348943652
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [mouseyy/result_data_2-3](https://huggingface.co/mouseyy/result_data_2-3) on the common_voice_17_0 dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.2132
|
| 37 |
+
- Wer: 0.2984
|
| 38 |
+
- Cer: 0.1512
|
| 39 |
|
| 40 |
## Model description
|
| 41 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_cer": 0.1512119333778769,
|
| 4 |
+
"eval_loss": 0.21320246160030365,
|
| 5 |
+
"eval_runtime": 130.2967,
|
| 6 |
+
"eval_samples": 5000,
|
| 7 |
+
"eval_samples_per_second": 38.374,
|
| 8 |
+
"eval_steps_per_second": 1.205,
|
| 9 |
+
"eval_wer": 0.2984287348943652,
|
| 10 |
+
"total_flos": 1.21495045308783e+20,
|
| 11 |
+
"train_loss": 0.13136799893019088,
|
| 12 |
+
"train_runtime": 24963.2457,
|
| 13 |
+
"train_samples": 35144,
|
| 14 |
+
"train_samples_per_second": 21.117,
|
| 15 |
+
"train_steps_per_second": 0.66
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_cer": 0.1512119333778769,
|
| 4 |
+
"eval_loss": 0.21320246160030365,
|
| 5 |
+
"eval_runtime": 130.2967,
|
| 6 |
+
"eval_samples": 5000,
|
| 7 |
+
"eval_samples_per_second": 38.374,
|
| 8 |
+
"eval_steps_per_second": 1.205,
|
| 9 |
+
"eval_wer": 0.2984287348943652
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"total_flos": 1.21495045308783e+20,
|
| 4 |
+
"train_loss": 0.13136799893019088,
|
| 5 |
+
"train_runtime": 24963.2457,
|
| 6 |
+
"train_samples": 35144,
|
| 7 |
+
"train_samples_per_second": 21.117,
|
| 8 |
+
"train_steps_per_second": 0.66
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1510 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.29856827886467024,
|
| 3 |
+
"best_model_checkpoint": "/home/senyk/result_wav2vec/best_model_2_copy/checkpoint-15500",
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 16485,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.09099181073703366,
|
| 13 |
+
"grad_norm": 1.8533661365509033,
|
| 14 |
+
"learning_rate": 3.9757355171367915e-05,
|
| 15 |
+
"loss": 0.1805,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.18198362147406733,
|
| 20 |
+
"grad_norm": 1.234511137008667,
|
| 21 |
+
"learning_rate": 3.951471034273582e-05,
|
| 22 |
+
"loss": 0.1797,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.272975432211101,
|
| 27 |
+
"grad_norm": 2.1243155002593994,
|
| 28 |
+
"learning_rate": 3.927206551410373e-05,
|
| 29 |
+
"loss": 0.185,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.36396724294813465,
|
| 34 |
+
"grad_norm": 1.542136549949646,
|
| 35 |
+
"learning_rate": 3.902942068547165e-05,
|
| 36 |
+
"loss": 0.1808,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4549590536851683,
|
| 41 |
+
"grad_norm": 1.6958433389663696,
|
| 42 |
+
"learning_rate": 3.8786775856839554e-05,
|
| 43 |
+
"loss": 0.1936,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.4549590536851683,
|
| 48 |
+
"eval_cer": 0.15896814599550807,
|
| 49 |
+
"eval_loss": 0.2081621140241623,
|
| 50 |
+
"eval_runtime": 132.4545,
|
| 51 |
+
"eval_samples_per_second": 37.749,
|
| 52 |
+
"eval_steps_per_second": 1.185,
|
| 53 |
+
"eval_wer": 0.3304680304764031,
|
| 54 |
+
"step": 500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.545950864422202,
|
| 58 |
+
"grad_norm": 1.8750900030136108,
|
| 59 |
+
"learning_rate": 3.8544131028207465e-05,
|
| 60 |
+
"loss": 0.1848,
|
| 61 |
+
"step": 600
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.6369426751592356,
|
| 65 |
+
"grad_norm": 0.7346371412277222,
|
| 66 |
+
"learning_rate": 3.8301486199575376e-05,
|
| 67 |
+
"loss": 0.1899,
|
| 68 |
+
"step": 700
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.7279344858962693,
|
| 72 |
+
"grad_norm": 0.8798062205314636,
|
| 73 |
+
"learning_rate": 3.805884137094329e-05,
|
| 74 |
+
"loss": 0.1761,
|
| 75 |
+
"step": 800
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.818926296633303,
|
| 79 |
+
"grad_norm": 3.140528917312622,
|
| 80 |
+
"learning_rate": 3.7818622990597514e-05,
|
| 81 |
+
"loss": 0.1766,
|
| 82 |
+
"step": 900
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.9099181073703366,
|
| 86 |
+
"grad_norm": 1.5800111293792725,
|
| 87 |
+
"learning_rate": 3.7575978161965426e-05,
|
| 88 |
+
"loss": 0.174,
|
| 89 |
+
"step": 1000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.9099181073703366,
|
| 93 |
+
"eval_cer": 0.15860186921906572,
|
| 94 |
+
"eval_loss": 0.20823511481285095,
|
| 95 |
+
"eval_runtime": 132.6784,
|
| 96 |
+
"eval_samples_per_second": 37.685,
|
| 97 |
+
"eval_steps_per_second": 1.183,
|
| 98 |
+
"eval_wer": 0.3283748709218275,
|
| 99 |
+
"step": 1000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.0009099181073704,
|
| 103 |
+
"grad_norm": 0.9137107133865356,
|
| 104 |
+
"learning_rate": 3.733333333333334e-05,
|
| 105 |
+
"loss": 0.1746,
|
| 106 |
+
"step": 1100
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.091901728844404,
|
| 110 |
+
"grad_norm": 1.5409084558486938,
|
| 111 |
+
"learning_rate": 3.709068850470125e-05,
|
| 112 |
+
"loss": 0.175,
|
| 113 |
+
"step": 1200
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.1828935395814377,
|
| 117 |
+
"grad_norm": 1.2571290731430054,
|
| 118 |
+
"learning_rate": 3.684804367606915e-05,
|
| 119 |
+
"loss": 0.161,
|
| 120 |
+
"step": 1300
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.2738853503184713,
|
| 124 |
+
"grad_norm": 0.7620792388916016,
|
| 125 |
+
"learning_rate": 3.6605398847437065e-05,
|
| 126 |
+
"loss": 0.1704,
|
| 127 |
+
"step": 1400
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 1.364877161055505,
|
| 131 |
+
"grad_norm": 1.167091727256775,
|
| 132 |
+
"learning_rate": 3.6362754018804976e-05,
|
| 133 |
+
"loss": 0.1855,
|
| 134 |
+
"step": 1500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.364877161055505,
|
| 138 |
+
"eval_cer": 0.15846501855533998,
|
| 139 |
+
"eval_loss": 0.19808036088943481,
|
| 140 |
+
"eval_runtime": 131.4602,
|
| 141 |
+
"eval_samples_per_second": 38.034,
|
| 142 |
+
"eval_steps_per_second": 1.194,
|
| 143 |
+
"eval_wer": 0.3291563171555357,
|
| 144 |
+
"step": 1500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.4558689717925386,
|
| 148 |
+
"grad_norm": 1.6935396194458008,
|
| 149 |
+
"learning_rate": 3.612010919017289e-05,
|
| 150 |
+
"loss": 0.1698,
|
| 151 |
+
"step": 1600
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.5468607825295724,
|
| 155 |
+
"grad_norm": 1.1173765659332275,
|
| 156 |
+
"learning_rate": 3.58774643615408e-05,
|
| 157 |
+
"loss": 0.1705,
|
| 158 |
+
"step": 1700
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.6378525932666061,
|
| 162 |
+
"grad_norm": 1.6780610084533691,
|
| 163 |
+
"learning_rate": 3.563481953290871e-05,
|
| 164 |
+
"loss": 0.177,
|
| 165 |
+
"step": 1800
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.7288444040036397,
|
| 169 |
+
"grad_norm": 0.9590256214141846,
|
| 170 |
+
"learning_rate": 3.539217470427662e-05,
|
| 171 |
+
"loss": 0.1778,
|
| 172 |
+
"step": 1900
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.8198362147406733,
|
| 176 |
+
"grad_norm": 0.6319223642349243,
|
| 177 |
+
"learning_rate": 3.5149529875644526e-05,
|
| 178 |
+
"loss": 0.1724,
|
| 179 |
+
"step": 2000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.8198362147406733,
|
| 183 |
+
"eval_cer": 0.15803031644703477,
|
| 184 |
+
"eval_loss": 0.19977422058582306,
|
| 185 |
+
"eval_runtime": 131.3497,
|
| 186 |
+
"eval_samples_per_second": 38.066,
|
| 187 |
+
"eval_steps_per_second": 1.195,
|
| 188 |
+
"eval_wer": 0.3265049817197399,
|
| 189 |
+
"step": 2000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.910828025477707,
|
| 193 |
+
"grad_norm": 1.1627233028411865,
|
| 194 |
+
"learning_rate": 3.490688504701244e-05,
|
| 195 |
+
"loss": 0.1745,
|
| 196 |
+
"step": 2100
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 2.001819836214741,
|
| 200 |
+
"grad_norm": 0.90104740858078,
|
| 201 |
+
"learning_rate": 3.466424021838035e-05,
|
| 202 |
+
"loss": 0.1669,
|
| 203 |
+
"step": 2200
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.092811646951774,
|
| 207 |
+
"grad_norm": 0.7599895000457764,
|
| 208 |
+
"learning_rate": 3.4421595389748254e-05,
|
| 209 |
+
"loss": 0.1523,
|
| 210 |
+
"step": 2300
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.183803457688808,
|
| 214 |
+
"grad_norm": 0.6392490267753601,
|
| 215 |
+
"learning_rate": 3.417895056111617e-05,
|
| 216 |
+
"loss": 0.1596,
|
| 217 |
+
"step": 2400
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 2.2747952684258417,
|
| 221 |
+
"grad_norm": 0.7351034283638,
|
| 222 |
+
"learning_rate": 3.3936305732484083e-05,
|
| 223 |
+
"loss": 0.1667,
|
| 224 |
+
"step": 2500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.2747952684258417,
|
| 228 |
+
"eval_cer": 0.15830804279400756,
|
| 229 |
+
"eval_loss": 0.19935080409049988,
|
| 230 |
+
"eval_runtime": 131.4244,
|
| 231 |
+
"eval_samples_per_second": 38.045,
|
| 232 |
+
"eval_steps_per_second": 1.195,
|
| 233 |
+
"eval_wer": 0.3288214116268036,
|
| 234 |
+
"step": 2500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.3657870791628755,
|
| 238 |
+
"grad_norm": 1.4110583066940308,
|
| 239 |
+
"learning_rate": 3.369366090385199e-05,
|
| 240 |
+
"loss": 0.1592,
|
| 241 |
+
"step": 2600
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 2.4567788898999092,
|
| 245 |
+
"grad_norm": 1.1413319110870361,
|
| 246 |
+
"learning_rate": 3.34510160752199e-05,
|
| 247 |
+
"loss": 0.1557,
|
| 248 |
+
"step": 2700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.5477707006369426,
|
| 252 |
+
"grad_norm": 1.064664363861084,
|
| 253 |
+
"learning_rate": 3.320837124658781e-05,
|
| 254 |
+
"loss": 0.1551,
|
| 255 |
+
"step": 2800
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.6387625113739763,
|
| 259 |
+
"grad_norm": 0.9104379415512085,
|
| 260 |
+
"learning_rate": 3.296572641795572e-05,
|
| 261 |
+
"loss": 0.1465,
|
| 262 |
+
"step": 2900
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 2.72975432211101,
|
| 266 |
+
"grad_norm": 0.9136043787002563,
|
| 267 |
+
"learning_rate": 3.272308158932363e-05,
|
| 268 |
+
"loss": 0.1635,
|
| 269 |
+
"step": 3000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 2.72975432211101,
|
| 273 |
+
"eval_cer": 0.1568509857272808,
|
| 274 |
+
"eval_loss": 0.20498071610927582,
|
| 275 |
+
"eval_runtime": 131.4378,
|
| 276 |
+
"eval_samples_per_second": 38.041,
|
| 277 |
+
"eval_steps_per_second": 1.194,
|
| 278 |
+
"eval_wer": 0.3222907538165276,
|
| 279 |
+
"step": 3000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.8207461328480434,
|
| 283 |
+
"grad_norm": 1.207688570022583,
|
| 284 |
+
"learning_rate": 3.248043676069154e-05,
|
| 285 |
+
"loss": 0.1662,
|
| 286 |
+
"step": 3100
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 2.911737943585077,
|
| 290 |
+
"grad_norm": 2.2350857257843018,
|
| 291 |
+
"learning_rate": 3.223779193205945e-05,
|
| 292 |
+
"loss": 0.1602,
|
| 293 |
+
"step": 3200
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.002729754322111,
|
| 297 |
+
"grad_norm": 1.0255531072616577,
|
| 298 |
+
"learning_rate": 3.199514710342736e-05,
|
| 299 |
+
"loss": 0.1485,
|
| 300 |
+
"step": 3300
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 3.0937215650591448,
|
| 304 |
+
"grad_norm": 1.2408232688903809,
|
| 305 |
+
"learning_rate": 3.175250227479527e-05,
|
| 306 |
+
"loss": 0.1528,
|
| 307 |
+
"step": 3400
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 3.1847133757961785,
|
| 311 |
+
"grad_norm": 0.6412222981452942,
|
| 312 |
+
"learning_rate": 3.1509857446163184e-05,
|
| 313 |
+
"loss": 0.14,
|
| 314 |
+
"step": 3500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.1847133757961785,
|
| 318 |
+
"eval_cer": 0.1566980349854697,
|
| 319 |
+
"eval_loss": 0.2068762630224228,
|
| 320 |
+
"eval_runtime": 131.207,
|
| 321 |
+
"eval_samples_per_second": 38.108,
|
| 322 |
+
"eval_steps_per_second": 1.197,
|
| 323 |
+
"eval_wer": 0.32053249979068404,
|
| 324 |
+
"step": 3500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 3.275705186533212,
|
| 328 |
+
"grad_norm": 1.1700862646102905,
|
| 329 |
+
"learning_rate": 3.126721261753109e-05,
|
| 330 |
+
"loss": 0.1522,
|
| 331 |
+
"step": 3600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 3.3666969972702456,
|
| 335 |
+
"grad_norm": 2.105367660522461,
|
| 336 |
+
"learning_rate": 3.1024567788899e-05,
|
| 337 |
+
"loss": 0.1501,
|
| 338 |
+
"step": 3700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.4576888080072794,
|
| 342 |
+
"grad_norm": 0.5450145602226257,
|
| 343 |
+
"learning_rate": 3.0784349408553233e-05,
|
| 344 |
+
"loss": 0.1446,
|
| 345 |
+
"step": 3800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.548680618744313,
|
| 349 |
+
"grad_norm": 0.7369564771652222,
|
| 350 |
+
"learning_rate": 3.0541704579921145e-05,
|
| 351 |
+
"loss": 0.1509,
|
| 352 |
+
"step": 3900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 3.6396724294813465,
|
| 356 |
+
"grad_norm": 1.5404269695281982,
|
| 357 |
+
"learning_rate": 3.0299059751289053e-05,
|
| 358 |
+
"loss": 0.1573,
|
| 359 |
+
"step": 4000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 3.6396724294813465,
|
| 363 |
+
"eval_cer": 0.156923436078665,
|
| 364 |
+
"eval_loss": 0.2066684067249298,
|
| 365 |
+
"eval_runtime": 132.0661,
|
| 366 |
+
"eval_samples_per_second": 37.86,
|
| 367 |
+
"eval_steps_per_second": 1.189,
|
| 368 |
+
"eval_wer": 0.32067204376098907,
|
| 369 |
+
"step": 4000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.7306642402183803,
|
| 373 |
+
"grad_norm": 1.83568274974823,
|
| 374 |
+
"learning_rate": 3.005641492265696e-05,
|
| 375 |
+
"loss": 0.1489,
|
| 376 |
+
"step": 4100
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 3.821656050955414,
|
| 380 |
+
"grad_norm": 0.9000252485275269,
|
| 381 |
+
"learning_rate": 2.9813770094024872e-05,
|
| 382 |
+
"loss": 0.1446,
|
| 383 |
+
"step": 4200
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.912647861692448,
|
| 387 |
+
"grad_norm": 1.9459755420684814,
|
| 388 |
+
"learning_rate": 2.957112526539278e-05,
|
| 389 |
+
"loss": 0.1541,
|
| 390 |
+
"step": 4300
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 4.003639672429482,
|
| 394 |
+
"grad_norm": 1.2044388055801392,
|
| 395 |
+
"learning_rate": 2.9328480436760695e-05,
|
| 396 |
+
"loss": 0.1555,
|
| 397 |
+
"step": 4400
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 4.094631483166515,
|
| 401 |
+
"grad_norm": 0.984316885471344,
|
| 402 |
+
"learning_rate": 2.9085835608128607e-05,
|
| 403 |
+
"loss": 0.1487,
|
| 404 |
+
"step": 4500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.094631483166515,
|
| 408 |
+
"eval_cer": 0.15719713740611643,
|
| 409 |
+
"eval_loss": 0.21014198660850525,
|
| 410 |
+
"eval_runtime": 131.975,
|
| 411 |
+
"eval_samples_per_second": 37.886,
|
| 412 |
+
"eval_steps_per_second": 1.19,
|
| 413 |
+
"eval_wer": 0.3224023889927716,
|
| 414 |
+
"step": 4500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.185623293903548,
|
| 418 |
+
"grad_norm": 0.879389762878418,
|
| 419 |
+
"learning_rate": 2.8843190779496515e-05,
|
| 420 |
+
"loss": 0.1384,
|
| 421 |
+
"step": 4600
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 4.276615104640582,
|
| 425 |
+
"grad_norm": 1.0293374061584473,
|
| 426 |
+
"learning_rate": 2.8600545950864426e-05,
|
| 427 |
+
"loss": 0.1419,
|
| 428 |
+
"step": 4700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.367606915377616,
|
| 432 |
+
"grad_norm": 0.7723912596702576,
|
| 433 |
+
"learning_rate": 2.8357901122232334e-05,
|
| 434 |
+
"loss": 0.142,
|
| 435 |
+
"step": 4800
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.45859872611465,
|
| 439 |
+
"grad_norm": 1.056503176689148,
|
| 440 |
+
"learning_rate": 2.8115256293600245e-05,
|
| 441 |
+
"loss": 0.1492,
|
| 442 |
+
"step": 4900
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 4.549590536851683,
|
| 446 |
+
"grad_norm": 1.1004029512405396,
|
| 447 |
+
"learning_rate": 2.7872611464968153e-05,
|
| 448 |
+
"loss": 0.1501,
|
| 449 |
+
"step": 5000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 4.549590536851683,
|
| 453 |
+
"eval_cer": 0.1563116331114206,
|
| 454 |
+
"eval_loss": 0.2110494077205658,
|
| 455 |
+
"eval_runtime": 131.8894,
|
| 456 |
+
"eval_samples_per_second": 37.911,
|
| 457 |
+
"eval_steps_per_second": 1.19,
|
| 458 |
+
"eval_wer": 0.3176020764142781,
|
| 459 |
+
"step": 5000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 4.640582347588717,
|
| 463 |
+
"grad_norm": 1.6329824924468994,
|
| 464 |
+
"learning_rate": 2.7629966636336068e-05,
|
| 465 |
+
"loss": 0.1398,
|
| 466 |
+
"step": 5100
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 4.731574158325751,
|
| 470 |
+
"grad_norm": 1.094870924949646,
|
| 471 |
+
"learning_rate": 2.7387321807703976e-05,
|
| 472 |
+
"loss": 0.1382,
|
| 473 |
+
"step": 5200
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 4.822565969062785,
|
| 477 |
+
"grad_norm": 1.094946026802063,
|
| 478 |
+
"learning_rate": 2.7144676979071888e-05,
|
| 479 |
+
"loss": 0.1471,
|
| 480 |
+
"step": 5300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 4.9135577797998184,
|
| 484 |
+
"grad_norm": 1.1323785781860352,
|
| 485 |
+
"learning_rate": 2.6902032150439796e-05,
|
| 486 |
+
"loss": 0.1487,
|
| 487 |
+
"step": 5400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 5.004549590536851,
|
| 491 |
+
"grad_norm": 1.0509577989578247,
|
| 492 |
+
"learning_rate": 2.6659387321807707e-05,
|
| 493 |
+
"loss": 0.1486,
|
| 494 |
+
"step": 5500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 5.004549590536851,
|
| 498 |
+
"eval_cer": 0.15565957994896276,
|
| 499 |
+
"eval_loss": 0.20399489998817444,
|
| 500 |
+
"eval_runtime": 131.3537,
|
| 501 |
+
"eval_samples_per_second": 38.065,
|
| 502 |
+
"eval_steps_per_second": 1.195,
|
| 503 |
+
"eval_wer": 0.3159275487706176,
|
| 504 |
+
"step": 5500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 5.095541401273885,
|
| 508 |
+
"grad_norm": 1.0751926898956299,
|
| 509 |
+
"learning_rate": 2.6416742493175615e-05,
|
| 510 |
+
"loss": 0.1311,
|
| 511 |
+
"step": 5600
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 5.186533212010919,
|
| 515 |
+
"grad_norm": 1.638108730316162,
|
| 516 |
+
"learning_rate": 2.6174097664543526e-05,
|
| 517 |
+
"loss": 0.1437,
|
| 518 |
+
"step": 5700
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 5.277525022747953,
|
| 522 |
+
"grad_norm": 0.8366700410842896,
|
| 523 |
+
"learning_rate": 2.5931452835911434e-05,
|
| 524 |
+
"loss": 0.1334,
|
| 525 |
+
"step": 5800
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 5.368516833484986,
|
| 529 |
+
"grad_norm": 0.9168310761451721,
|
| 530 |
+
"learning_rate": 2.568880800727935e-05,
|
| 531 |
+
"loss": 0.1291,
|
| 532 |
+
"step": 5900
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 5.45950864422202,
|
| 536 |
+
"grad_norm": 0.8272154331207275,
|
| 537 |
+
"learning_rate": 2.5446163178647257e-05,
|
| 538 |
+
"loss": 0.1342,
|
| 539 |
+
"step": 6000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 5.45950864422202,
|
| 543 |
+
"eval_cer": 0.15511217729405988,
|
| 544 |
+
"eval_loss": 0.20407769083976746,
|
| 545 |
+
"eval_runtime": 130.9722,
|
| 546 |
+
"eval_samples_per_second": 38.176,
|
| 547 |
+
"eval_steps_per_second": 1.199,
|
| 548 |
+
"eval_wer": 0.31436465630320115,
|
| 549 |
+
"step": 6000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 5.550500454959054,
|
| 553 |
+
"grad_norm": 0.7047191262245178,
|
| 554 |
+
"learning_rate": 2.520351835001517e-05,
|
| 555 |
+
"loss": 0.131,
|
| 556 |
+
"step": 6100
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 5.641492265696087,
|
| 560 |
+
"grad_norm": 1.3190622329711914,
|
| 561 |
+
"learning_rate": 2.4960873521383077e-05,
|
| 562 |
+
"loss": 0.1335,
|
| 563 |
+
"step": 6200
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 5.732484076433121,
|
| 567 |
+
"grad_norm": 1.6135904788970947,
|
| 568 |
+
"learning_rate": 2.472065514103731e-05,
|
| 569 |
+
"loss": 0.1352,
|
| 570 |
+
"step": 6300
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 5.823475887170154,
|
| 574 |
+
"grad_norm": 1.2571635246276855,
|
| 575 |
+
"learning_rate": 2.447801031240522e-05,
|
| 576 |
+
"loss": 0.1405,
|
| 577 |
+
"step": 6400
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 5.914467697907188,
|
| 581 |
+
"grad_norm": 1.103220820426941,
|
| 582 |
+
"learning_rate": 2.423536548377313e-05,
|
| 583 |
+
"loss": 0.1396,
|
| 584 |
+
"step": 6500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 5.914467697907188,
|
| 588 |
+
"eval_cer": 0.15522085282113618,
|
| 589 |
+
"eval_loss": 0.20565420389175415,
|
| 590 |
+
"eval_runtime": 131.5948,
|
| 591 |
+
"eval_samples_per_second": 37.995,
|
| 592 |
+
"eval_steps_per_second": 1.193,
|
| 593 |
+
"eval_wer": 0.31433674750914015,
|
| 594 |
+
"step": 6500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 6.005459508644222,
|
| 598 |
+
"grad_norm": 1.2131693363189697,
|
| 599 |
+
"learning_rate": 2.399272065514104e-05,
|
| 600 |
+
"loss": 0.1379,
|
| 601 |
+
"step": 6600
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 6.096451319381256,
|
| 605 |
+
"grad_norm": 0.8314226865768433,
|
| 606 |
+
"learning_rate": 2.375007582650895e-05,
|
| 607 |
+
"loss": 0.1289,
|
| 608 |
+
"step": 6700
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 6.1874431301182895,
|
| 612 |
+
"grad_norm": 0.929862380027771,
|
| 613 |
+
"learning_rate": 2.350743099787686e-05,
|
| 614 |
+
"loss": 0.1291,
|
| 615 |
+
"step": 6800
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 6.278434940855323,
|
| 619 |
+
"grad_norm": 1.2670739889144897,
|
| 620 |
+
"learning_rate": 2.326478616924477e-05,
|
| 621 |
+
"loss": 0.1349,
|
| 622 |
+
"step": 6900
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 6.369426751592357,
|
| 626 |
+
"grad_norm": 0.979325532913208,
|
| 627 |
+
"learning_rate": 2.3022141340612677e-05,
|
| 628 |
+
"loss": 0.136,
|
| 629 |
+
"step": 7000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 6.369426751592357,
|
| 633 |
+
"eval_cer": 0.15454464954155028,
|
| 634 |
+
"eval_loss": 0.20977585017681122,
|
| 635 |
+
"eval_runtime": 132.5521,
|
| 636 |
+
"eval_samples_per_second": 37.721,
|
| 637 |
+
"eval_steps_per_second": 1.184,
|
| 638 |
+
"eval_wer": 0.3130529429823337,
|
| 639 |
+
"step": 7000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 6.460418562329391,
|
| 643 |
+
"grad_norm": 1.1290801763534546,
|
| 644 |
+
"learning_rate": 2.277949651198059e-05,
|
| 645 |
+
"loss": 0.1419,
|
| 646 |
+
"step": 7100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 6.551410373066424,
|
| 650 |
+
"grad_norm": 0.8063333034515381,
|
| 651 |
+
"learning_rate": 2.2536851683348503e-05,
|
| 652 |
+
"loss": 0.1251,
|
| 653 |
+
"step": 7200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 6.6424021838034575,
|
| 657 |
+
"grad_norm": 0.7427828907966614,
|
| 658 |
+
"learning_rate": 2.229420685471641e-05,
|
| 659 |
+
"loss": 0.137,
|
| 660 |
+
"step": 7300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 6.733393994540491,
|
| 664 |
+
"grad_norm": 0.6900395154953003,
|
| 665 |
+
"learning_rate": 2.2051562026084322e-05,
|
| 666 |
+
"loss": 0.1359,
|
| 667 |
+
"step": 7400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 6.824385805277525,
|
| 671 |
+
"grad_norm": 1.7373096942901611,
|
| 672 |
+
"learning_rate": 2.180891719745223e-05,
|
| 673 |
+
"loss": 0.1266,
|
| 674 |
+
"step": 7500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 6.824385805277525,
|
| 678 |
+
"eval_cer": 0.15423472303840674,
|
| 679 |
+
"eval_loss": 0.2094658762216568,
|
| 680 |
+
"eval_runtime": 132.3336,
|
| 681 |
+
"eval_samples_per_second": 37.783,
|
| 682 |
+
"eval_steps_per_second": 1.186,
|
| 683 |
+
"eval_wer": 0.310596969104965,
|
| 684 |
+
"step": 7500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 6.915377616014559,
|
| 688 |
+
"grad_norm": 0.7071816921234131,
|
| 689 |
+
"learning_rate": 2.156627236882014e-05,
|
| 690 |
+
"loss": 0.1267,
|
| 691 |
+
"step": 7600
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 7.006369426751593,
|
| 695 |
+
"grad_norm": 1.2711293697357178,
|
| 696 |
+
"learning_rate": 2.132362754018805e-05,
|
| 697 |
+
"loss": 0.1284,
|
| 698 |
+
"step": 7700
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 7.097361237488626,
|
| 702 |
+
"grad_norm": 0.9852485656738281,
|
| 703 |
+
"learning_rate": 2.108098271155596e-05,
|
| 704 |
+
"loss": 0.1223,
|
| 705 |
+
"step": 7800
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 7.188353048225659,
|
| 709 |
+
"grad_norm": 1.4875195026397705,
|
| 710 |
+
"learning_rate": 2.0838337882923872e-05,
|
| 711 |
+
"loss": 0.1273,
|
| 712 |
+
"step": 7900
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 7.279344858962693,
|
| 716 |
+
"grad_norm": 0.8052563071250916,
|
| 717 |
+
"learning_rate": 2.0595693054291784e-05,
|
| 718 |
+
"loss": 0.1283,
|
| 719 |
+
"step": 8000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 7.279344858962693,
|
| 723 |
+
"eval_cer": 0.1537718457934521,
|
| 724 |
+
"eval_loss": 0.2159704566001892,
|
| 725 |
+
"eval_runtime": 132.3477,
|
| 726 |
+
"eval_samples_per_second": 37.779,
|
| 727 |
+
"eval_steps_per_second": 1.186,
|
| 728 |
+
"eval_wer": 0.3085317183444503,
|
| 729 |
+
"step": 8000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 7.370336669699727,
|
| 733 |
+
"grad_norm": 2.028435230255127,
|
| 734 |
+
"learning_rate": 2.0353048225659692e-05,
|
| 735 |
+
"loss": 0.1346,
|
| 736 |
+
"step": 8100
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 7.461328480436761,
|
| 740 |
+
"grad_norm": 0.8297127485275269,
|
| 741 |
+
"learning_rate": 2.0110403397027603e-05,
|
| 742 |
+
"loss": 0.1235,
|
| 743 |
+
"step": 8200
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 7.552320291173794,
|
| 747 |
+
"grad_norm": 1.299826741218567,
|
| 748 |
+
"learning_rate": 1.986775856839551e-05,
|
| 749 |
+
"loss": 0.1243,
|
| 750 |
+
"step": 8300
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 7.643312101910828,
|
| 754 |
+
"grad_norm": 2.9055380821228027,
|
| 755 |
+
"learning_rate": 1.9625113739763423e-05,
|
| 756 |
+
"loss": 0.1199,
|
| 757 |
+
"step": 8400
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 7.734303912647862,
|
| 761 |
+
"grad_norm": 1.5928077697753906,
|
| 762 |
+
"learning_rate": 1.9382468911131334e-05,
|
| 763 |
+
"loss": 0.1229,
|
| 764 |
+
"step": 8500
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 7.734303912647862,
|
| 768 |
+
"eval_cer": 0.1537919708910588,
|
| 769 |
+
"eval_loss": 0.21750004589557648,
|
| 770 |
+
"eval_runtime": 131.9305,
|
| 771 |
+
"eval_samples_per_second": 37.899,
|
| 772 |
+
"eval_steps_per_second": 1.19,
|
| 773 |
+
"eval_wer": 0.30755491055231504,
|
| 774 |
+
"step": 8500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 7.825295723384896,
|
| 778 |
+
"grad_norm": 1.1604926586151123,
|
| 779 |
+
"learning_rate": 1.9139824082499242e-05,
|
| 780 |
+
"loss": 0.1237,
|
| 781 |
+
"step": 8600
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 7.916287534121929,
|
| 785 |
+
"grad_norm": 2.2635440826416016,
|
| 786 |
+
"learning_rate": 1.8897179253867153e-05,
|
| 787 |
+
"loss": 0.1248,
|
| 788 |
+
"step": 8700
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 8.007279344858963,
|
| 792 |
+
"grad_norm": 0.6642023324966431,
|
| 793 |
+
"learning_rate": 1.8654534425235065e-05,
|
| 794 |
+
"loss": 0.1276,
|
| 795 |
+
"step": 8800
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 8.098271155595997,
|
| 799 |
+
"grad_norm": 2.0704445838928223,
|
| 800 |
+
"learning_rate": 1.8411889596602973e-05,
|
| 801 |
+
"loss": 0.1124,
|
| 802 |
+
"step": 8900
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 8.18926296633303,
|
| 806 |
+
"grad_norm": 0.8484503030776978,
|
| 807 |
+
"learning_rate": 1.8169244767970884e-05,
|
| 808 |
+
"loss": 0.1267,
|
| 809 |
+
"step": 9000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 8.18926296633303,
|
| 813 |
+
"eval_cer": 0.15306746737721677,
|
| 814 |
+
"eval_loss": 0.21141663193702698,
|
| 815 |
+
"eval_runtime": 132.5307,
|
| 816 |
+
"eval_samples_per_second": 37.727,
|
| 817 |
+
"eval_steps_per_second": 1.185,
|
| 818 |
+
"eval_wer": 0.30568502135022746,
|
| 819 |
+
"step": 9000
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 8.280254777070065,
|
| 823 |
+
"grad_norm": 0.8810114860534668,
|
| 824 |
+
"learning_rate": 1.7926599939338792e-05,
|
| 825 |
+
"loss": 0.1191,
|
| 826 |
+
"step": 9100
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 8.371246587807097,
|
| 830 |
+
"grad_norm": 0.343791663646698,
|
| 831 |
+
"learning_rate": 1.7686381558993026e-05,
|
| 832 |
+
"loss": 0.1135,
|
| 833 |
+
"step": 9200
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 8.46223839854413,
|
| 837 |
+
"grad_norm": 1.2659159898757935,
|
| 838 |
+
"learning_rate": 1.7443736730360937e-05,
|
| 839 |
+
"loss": 0.1215,
|
| 840 |
+
"step": 9300
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 8.553230209281164,
|
| 844 |
+
"grad_norm": 2.19396710395813,
|
| 845 |
+
"learning_rate": 1.7201091901728845e-05,
|
| 846 |
+
"loss": 0.1228,
|
| 847 |
+
"step": 9400
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 8.644222020018198,
|
| 851 |
+
"grad_norm": 0.6617141366004944,
|
| 852 |
+
"learning_rate": 1.6958447073096757e-05,
|
| 853 |
+
"loss": 0.1127,
|
| 854 |
+
"step": 9500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 8.644222020018198,
|
| 858 |
+
"eval_cer": 0.1528179161668934,
|
| 859 |
+
"eval_loss": 0.20628662407398224,
|
| 860 |
+
"eval_runtime": 131.7465,
|
| 861 |
+
"eval_samples_per_second": 37.952,
|
| 862 |
+
"eval_steps_per_second": 1.192,
|
| 863 |
+
"eval_wer": 0.3068571907007898,
|
| 864 |
+
"step": 9500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 8.735213830755232,
|
| 868 |
+
"grad_norm": 1.5183671712875366,
|
| 869 |
+
"learning_rate": 1.6715802244464668e-05,
|
| 870 |
+
"loss": 0.1253,
|
| 871 |
+
"step": 9600
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 8.826205641492265,
|
| 875 |
+
"grad_norm": 0.6270197629928589,
|
| 876 |
+
"learning_rate": 1.6473157415832576e-05,
|
| 877 |
+
"loss": 0.1245,
|
| 878 |
+
"step": 9700
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 8.9171974522293,
|
| 882 |
+
"grad_norm": 0.7786601185798645,
|
| 883 |
+
"learning_rate": 1.6230512587200488e-05,
|
| 884 |
+
"loss": 0.1268,
|
| 885 |
+
"step": 9800
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 9.008189262966333,
|
| 889 |
+
"grad_norm": 1.2779630422592163,
|
| 890 |
+
"learning_rate": 1.5987867758568396e-05,
|
| 891 |
+
"loss": 0.1285,
|
| 892 |
+
"step": 9900
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 9.099181073703367,
|
| 896 |
+
"grad_norm": 0.8640280365943909,
|
| 897 |
+
"learning_rate": 1.5745222929936307e-05,
|
| 898 |
+
"loss": 0.1165,
|
| 899 |
+
"step": 10000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 9.099181073703367,
|
| 903 |
+
"eval_cer": 0.15322041811902787,
|
| 904 |
+
"eval_loss": 0.20939494669437408,
|
| 905 |
+
"eval_runtime": 130.9926,
|
| 906 |
+
"eval_samples_per_second": 38.17,
|
| 907 |
+
"eval_steps_per_second": 1.199,
|
| 908 |
+
"eval_wer": 0.3048477575283972,
|
| 909 |
+
"step": 10000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 9.1901728844404,
|
| 913 |
+
"grad_norm": 0.9638440012931824,
|
| 914 |
+
"learning_rate": 1.550257810130422e-05,
|
| 915 |
+
"loss": 0.1076,
|
| 916 |
+
"step": 10100
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 9.281164695177434,
|
| 920 |
+
"grad_norm": 0.8095070719718933,
|
| 921 |
+
"learning_rate": 1.526235972095845e-05,
|
| 922 |
+
"loss": 0.1268,
|
| 923 |
+
"step": 10200
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 9.372156505914468,
|
| 927 |
+
"grad_norm": 0.8781161308288574,
|
| 928 |
+
"learning_rate": 1.5019714892326358e-05,
|
| 929 |
+
"loss": 0.1207,
|
| 930 |
+
"step": 10300
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 9.463148316651502,
|
| 934 |
+
"grad_norm": 1.6324824094772339,
|
| 935 |
+
"learning_rate": 1.4777070063694268e-05,
|
| 936 |
+
"loss": 0.1256,
|
| 937 |
+
"step": 10400
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 9.554140127388536,
|
| 941 |
+
"grad_norm": 0.8956096172332764,
|
| 942 |
+
"learning_rate": 1.453442523506218e-05,
|
| 943 |
+
"loss": 0.1222,
|
| 944 |
+
"step": 10500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 9.554140127388536,
|
| 948 |
+
"eval_cer": 0.1532043180409425,
|
| 949 |
+
"eval_loss": 0.20789633691310883,
|
| 950 |
+
"eval_runtime": 131.5404,
|
| 951 |
+
"eval_samples_per_second": 38.011,
|
| 952 |
+
"eval_steps_per_second": 1.194,
|
| 953 |
+
"eval_wer": 0.3066897379364238,
|
| 954 |
+
"step": 10500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 9.64513193812557,
|
| 958 |
+
"grad_norm": 1.19681978225708,
|
| 959 |
+
"learning_rate": 1.4291780406430089e-05,
|
| 960 |
+
"loss": 0.1197,
|
| 961 |
+
"step": 10600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 9.736123748862603,
|
| 965 |
+
"grad_norm": 5.549036026000977,
|
| 966 |
+
"learning_rate": 1.4049135577797999e-05,
|
| 967 |
+
"loss": 0.1128,
|
| 968 |
+
"step": 10700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 9.827115559599637,
|
| 972 |
+
"grad_norm": 0.7990231513977051,
|
| 973 |
+
"learning_rate": 1.380649074916591e-05,
|
| 974 |
+
"loss": 0.1164,
|
| 975 |
+
"step": 10800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 9.918107370336669,
|
| 979 |
+
"grad_norm": 0.9332329630851746,
|
| 980 |
+
"learning_rate": 1.356384592053382e-05,
|
| 981 |
+
"loss": 0.1236,
|
| 982 |
+
"step": 10900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 10.009099181073703,
|
| 986 |
+
"grad_norm": 1.4885659217834473,
|
| 987 |
+
"learning_rate": 1.332120109190173e-05,
|
| 988 |
+
"loss": 0.1127,
|
| 989 |
+
"step": 11000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 10.009099181073703,
|
| 993 |
+
"eval_cer": 0.15307551741625947,
|
| 994 |
+
"eval_loss": 0.20891791582107544,
|
| 995 |
+
"eval_runtime": 131.2946,
|
| 996 |
+
"eval_samples_per_second": 38.082,
|
| 997 |
+
"eval_steps_per_second": 1.196,
|
| 998 |
+
"eval_wer": 0.30557338617398344,
|
| 999 |
+
"step": 11000
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 10.100090991810736,
|
| 1003 |
+
"grad_norm": 0.5533010959625244,
|
| 1004 |
+
"learning_rate": 1.307855626326964e-05,
|
| 1005 |
+
"loss": 0.1158,
|
| 1006 |
+
"step": 11100
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 10.19108280254777,
|
| 1010 |
+
"grad_norm": 2.3450381755828857,
|
| 1011 |
+
"learning_rate": 1.283591143463755e-05,
|
| 1012 |
+
"loss": 0.1136,
|
| 1013 |
+
"step": 11200
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 10.282074613284804,
|
| 1017 |
+
"grad_norm": 0.6852346062660217,
|
| 1018 |
+
"learning_rate": 1.259326660600546e-05,
|
| 1019 |
+
"loss": 0.1296,
|
| 1020 |
+
"step": 11300
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 10.373066424021838,
|
| 1024 |
+
"grad_norm": 0.872776985168457,
|
| 1025 |
+
"learning_rate": 1.235062177737337e-05,
|
| 1026 |
+
"loss": 0.1069,
|
| 1027 |
+
"step": 11400
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 10.464058234758872,
|
| 1031 |
+
"grad_norm": 0.547275722026825,
|
| 1032 |
+
"learning_rate": 1.210797694874128e-05,
|
| 1033 |
+
"loss": 0.1084,
|
| 1034 |
+
"step": 11500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 10.464058234758872,
|
| 1038 |
+
"eval_cer": 0.15258044001513407,
|
| 1039 |
+
"eval_loss": 0.2116706520318985,
|
| 1040 |
+
"eval_runtime": 131.7531,
|
| 1041 |
+
"eval_samples_per_second": 37.95,
|
| 1042 |
+
"eval_steps_per_second": 1.192,
|
| 1043 |
+
"eval_wer": 0.30317322988473666,
|
| 1044 |
+
"step": 11500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 10.555050045495905,
|
| 1048 |
+
"grad_norm": 1.4432127475738525,
|
| 1049 |
+
"learning_rate": 1.1865332120109191e-05,
|
| 1050 |
+
"loss": 0.1237,
|
| 1051 |
+
"step": 11600
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 10.646041856232939,
|
| 1055 |
+
"grad_norm": 0.5424668192863464,
|
| 1056 |
+
"learning_rate": 1.1622687291477101e-05,
|
| 1057 |
+
"loss": 0.116,
|
| 1058 |
+
"step": 11700
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 10.737033666969973,
|
| 1062 |
+
"grad_norm": 0.6486382484436035,
|
| 1063 |
+
"learning_rate": 1.138004246284501e-05,
|
| 1064 |
+
"loss": 0.1097,
|
| 1065 |
+
"step": 11800
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 10.828025477707007,
|
| 1069 |
+
"grad_norm": 1.37655770778656,
|
| 1070 |
+
"learning_rate": 1.113739763421292e-05,
|
| 1071 |
+
"loss": 0.1026,
|
| 1072 |
+
"step": 11900
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 10.91901728844404,
|
| 1076 |
+
"grad_norm": 0.7191163897514343,
|
| 1077 |
+
"learning_rate": 1.0894752805580833e-05,
|
| 1078 |
+
"loss": 0.1155,
|
| 1079 |
+
"step": 12000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 10.91901728844404,
|
| 1083 |
+
"eval_cer": 0.15271326565933846,
|
| 1084 |
+
"eval_loss": 0.20751111209392548,
|
| 1085 |
+
"eval_runtime": 131.6857,
|
| 1086 |
+
"eval_samples_per_second": 37.969,
|
| 1087 |
+
"eval_steps_per_second": 1.192,
|
| 1088 |
+
"eval_wer": 0.3045407607937261,
|
| 1089 |
+
"step": 12000
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 11.010009099181074,
|
| 1093 |
+
"grad_norm": 1.8931193351745605,
|
| 1094 |
+
"learning_rate": 1.0652107976948743e-05,
|
| 1095 |
+
"loss": 0.1154,
|
| 1096 |
+
"step": 12100
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 11.101000909918108,
|
| 1100 |
+
"grad_norm": 0.43597936630249023,
|
| 1101 |
+
"learning_rate": 1.0409463148316651e-05,
|
| 1102 |
+
"loss": 0.0999,
|
| 1103 |
+
"step": 12200
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 11.191992720655142,
|
| 1107 |
+
"grad_norm": 1.1339422464370728,
|
| 1108 |
+
"learning_rate": 1.0166818319684561e-05,
|
| 1109 |
+
"loss": 0.1107,
|
| 1110 |
+
"step": 12300
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 11.282984531392175,
|
| 1114 |
+
"grad_norm": 0.9059270620346069,
|
| 1115 |
+
"learning_rate": 9.924173491052472e-06,
|
| 1116 |
+
"loss": 0.1045,
|
| 1117 |
+
"step": 12400
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 11.373976342129207,
|
| 1121 |
+
"grad_norm": 0.8777015209197998,
|
| 1122 |
+
"learning_rate": 9.681528662420384e-06,
|
| 1123 |
+
"loss": 0.0955,
|
| 1124 |
+
"step": 12500
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 11.373976342129207,
|
| 1128 |
+
"eval_cer": 0.15231076370720398,
|
| 1129 |
+
"eval_loss": 0.21829599142074585,
|
| 1130 |
+
"eval_runtime": 131.8689,
|
| 1131 |
+
"eval_samples_per_second": 37.916,
|
| 1132 |
+
"eval_steps_per_second": 1.191,
|
| 1133 |
+
"eval_wer": 0.3025871452094555,
|
| 1134 |
+
"step": 12500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 11.464968152866241,
|
| 1138 |
+
"grad_norm": 0.7121880650520325,
|
| 1139 |
+
"learning_rate": 9.438883833788293e-06,
|
| 1140 |
+
"loss": 0.1068,
|
| 1141 |
+
"step": 12600
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 11.555959963603275,
|
| 1145 |
+
"grad_norm": 1.0068190097808838,
|
| 1146 |
+
"learning_rate": 9.196239005156203e-06,
|
| 1147 |
+
"loss": 0.1047,
|
| 1148 |
+
"step": 12700
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 11.646951774340309,
|
| 1152 |
+
"grad_norm": 1.2295094728469849,
|
| 1153 |
+
"learning_rate": 8.953594176524115e-06,
|
| 1154 |
+
"loss": 0.1041,
|
| 1155 |
+
"step": 12800
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 11.737943585077343,
|
| 1159 |
+
"grad_norm": 1.1067237854003906,
|
| 1160 |
+
"learning_rate": 8.710949347892024e-06,
|
| 1161 |
+
"loss": 0.1071,
|
| 1162 |
+
"step": 12900
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 11.828935395814376,
|
| 1166 |
+
"grad_norm": 0.9286106824874878,
|
| 1167 |
+
"learning_rate": 8.468304519259934e-06,
|
| 1168 |
+
"loss": 0.1146,
|
| 1169 |
+
"step": 13000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 11.828935395814376,
|
| 1173 |
+
"eval_cer": 0.15205316245783793,
|
| 1174 |
+
"eval_loss": 0.21160683035850525,
|
| 1175 |
+
"eval_runtime": 131.9777,
|
| 1176 |
+
"eval_samples_per_second": 37.885,
|
| 1177 |
+
"eval_steps_per_second": 1.19,
|
| 1178 |
+
"eval_wer": 0.30147079344701516,
|
| 1179 |
+
"step": 13000
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 11.91992720655141,
|
| 1183 |
+
"grad_norm": 1.0843188762664795,
|
| 1184 |
+
"learning_rate": 8.225659690627844e-06,
|
| 1185 |
+
"loss": 0.1099,
|
| 1186 |
+
"step": 13100
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 12.010919017288444,
|
| 1190 |
+
"grad_norm": 0.8679298162460327,
|
| 1191 |
+
"learning_rate": 7.983014861995755e-06,
|
| 1192 |
+
"loss": 0.1076,
|
| 1193 |
+
"step": 13200
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 12.101910828025478,
|
| 1197 |
+
"grad_norm": 1.5552619695663452,
|
| 1198 |
+
"learning_rate": 7.742796481649985e-06,
|
| 1199 |
+
"loss": 0.1167,
|
| 1200 |
+
"step": 13300
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 12.192902638762511,
|
| 1204 |
+
"grad_norm": 1.5181180238723755,
|
| 1205 |
+
"learning_rate": 7.500151653017895e-06,
|
| 1206 |
+
"loss": 0.1096,
|
| 1207 |
+
"step": 13400
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 12.283894449499545,
|
| 1211 |
+
"grad_norm": 0.6448826789855957,
|
| 1212 |
+
"learning_rate": 7.257506824385806e-06,
|
| 1213 |
+
"loss": 0.1094,
|
| 1214 |
+
"step": 13500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 12.283894449499545,
|
| 1218 |
+
"eval_cer": 0.15156613509575523,
|
| 1219 |
+
"eval_loss": 0.2089649885892868,
|
| 1220 |
+
"eval_runtime": 131.4101,
|
| 1221 |
+
"eval_samples_per_second": 38.049,
|
| 1222 |
+
"eval_steps_per_second": 1.195,
|
| 1223 |
+
"eval_wer": 0.2992939075102565,
|
| 1224 |
+
"step": 13500
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 12.374886260236579,
|
| 1228 |
+
"grad_norm": 0.68089759349823,
|
| 1229 |
+
"learning_rate": 7.014861995753716e-06,
|
| 1230 |
+
"loss": 0.1042,
|
| 1231 |
+
"step": 13600
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 12.465878070973613,
|
| 1235 |
+
"grad_norm": 0.5364871025085449,
|
| 1236 |
+
"learning_rate": 6.7722171671216266e-06,
|
| 1237 |
+
"loss": 0.1042,
|
| 1238 |
+
"step": 13700
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 12.556869881710647,
|
| 1242 |
+
"grad_norm": 0.9213688969612122,
|
| 1243 |
+
"learning_rate": 6.529572338489536e-06,
|
| 1244 |
+
"loss": 0.1075,
|
| 1245 |
+
"step": 13800
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 12.64786169244768,
|
| 1249 |
+
"grad_norm": 0.8983300924301147,
|
| 1250 |
+
"learning_rate": 6.286927509857447e-06,
|
| 1251 |
+
"loss": 0.1085,
|
| 1252 |
+
"step": 13900
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 12.738853503184714,
|
| 1256 |
+
"grad_norm": 1.987417459487915,
|
| 1257 |
+
"learning_rate": 6.0442826812253566e-06,
|
| 1258 |
+
"loss": 0.1072,
|
| 1259 |
+
"step": 14000
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 12.738853503184714,
|
| 1263 |
+
"eval_cer": 0.151727135876609,
|
| 1264 |
+
"eval_loss": 0.2124236822128296,
|
| 1265 |
+
"eval_runtime": 131.355,
|
| 1266 |
+
"eval_samples_per_second": 38.065,
|
| 1267 |
+
"eval_steps_per_second": 1.195,
|
| 1268 |
+
"eval_wer": 0.3002148977142698,
|
| 1269 |
+
"step": 14000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 12.829845313921748,
|
| 1273 |
+
"grad_norm": 1.3773497343063354,
|
| 1274 |
+
"learning_rate": 5.801637852593267e-06,
|
| 1275 |
+
"loss": 0.1077,
|
| 1276 |
+
"step": 14100
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 12.920837124658782,
|
| 1280 |
+
"grad_norm": 1.0006029605865479,
|
| 1281 |
+
"learning_rate": 5.558993023961178e-06,
|
| 1282 |
+
"loss": 0.0996,
|
| 1283 |
+
"step": 14200
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 13.011828935395814,
|
| 1287 |
+
"grad_norm": 1.134149193763733,
|
| 1288 |
+
"learning_rate": 5.316348195329087e-06,
|
| 1289 |
+
"loss": 0.1089,
|
| 1290 |
+
"step": 14300
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 13.102820746132847,
|
| 1294 |
+
"grad_norm": 1.6549540758132935,
|
| 1295 |
+
"learning_rate": 5.073703366696998e-06,
|
| 1296 |
+
"loss": 0.1143,
|
| 1297 |
+
"step": 14400
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 13.193812556869881,
|
| 1301 |
+
"grad_norm": 0.8590063452720642,
|
| 1302 |
+
"learning_rate": 4.831058538064908e-06,
|
| 1303 |
+
"loss": 0.1125,
|
| 1304 |
+
"step": 14500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 13.193812556869881,
|
| 1308 |
+
"eval_cer": 0.15173116089613034,
|
| 1309 |
+
"eval_loss": 0.2130936086177826,
|
| 1310 |
+
"eval_runtime": 131.6874,
|
| 1311 |
+
"eval_samples_per_second": 37.969,
|
| 1312 |
+
"eval_steps_per_second": 1.192,
|
| 1313 |
+
"eval_wer": 0.3000195361558427,
|
| 1314 |
+
"step": 14500
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 13.284804367606915,
|
| 1318 |
+
"grad_norm": 0.8164013028144836,
|
| 1319 |
+
"learning_rate": 4.588413709432818e-06,
|
| 1320 |
+
"loss": 0.0982,
|
| 1321 |
+
"step": 14600
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 13.375796178343949,
|
| 1325 |
+
"grad_norm": 0.8457829356193542,
|
| 1326 |
+
"learning_rate": 4.345768880800728e-06,
|
| 1327 |
+
"loss": 0.0988,
|
| 1328 |
+
"step": 14700
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 13.466787989080983,
|
| 1332 |
+
"grad_norm": 0.952691912651062,
|
| 1333 |
+
"learning_rate": 4.1031240521686385e-06,
|
| 1334 |
+
"loss": 0.1007,
|
| 1335 |
+
"step": 14800
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 13.557779799818016,
|
| 1339 |
+
"grad_norm": 0.4639749825000763,
|
| 1340 |
+
"learning_rate": 3.860479223536548e-06,
|
| 1341 |
+
"loss": 0.1088,
|
| 1342 |
+
"step": 14900
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 13.64877161055505,
|
| 1346 |
+
"grad_norm": 1.1107044219970703,
|
| 1347 |
+
"learning_rate": 3.6178343949044588e-06,
|
| 1348 |
+
"loss": 0.1058,
|
| 1349 |
+
"step": 15000
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 13.64877161055505,
|
| 1353 |
+
"eval_cer": 0.1514735596467643,
|
| 1354 |
+
"eval_loss": 0.217019721865654,
|
| 1355 |
+
"eval_runtime": 132.0061,
|
| 1356 |
+
"eval_samples_per_second": 37.877,
|
| 1357 |
+
"eval_steps_per_second": 1.189,
|
| 1358 |
+
"eval_wer": 0.29923808992213446,
|
| 1359 |
+
"step": 15000
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 13.739763421292084,
|
| 1363 |
+
"grad_norm": 1.299850583076477,
|
| 1364 |
+
"learning_rate": 3.375189566272369e-06,
|
| 1365 |
+
"loss": 0.1036,
|
| 1366 |
+
"step": 15100
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 13.830755232029118,
|
| 1370 |
+
"grad_norm": 1.5436201095581055,
|
| 1371 |
+
"learning_rate": 3.1325447376402795e-06,
|
| 1372 |
+
"loss": 0.0973,
|
| 1373 |
+
"step": 15200
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 13.921747042766151,
|
| 1377 |
+
"grad_norm": 1.0776885747909546,
|
| 1378 |
+
"learning_rate": 2.8898999090081896e-06,
|
| 1379 |
+
"loss": 0.0963,
|
| 1380 |
+
"step": 15300
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 14.012738853503185,
|
| 1384 |
+
"grad_norm": 0.9970951080322266,
|
| 1385 |
+
"learning_rate": 2.6472550803760997e-06,
|
| 1386 |
+
"loss": 0.1104,
|
| 1387 |
+
"step": 15400
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 14.103730664240219,
|
| 1391 |
+
"grad_norm": 0.5252935886383057,
|
| 1392 |
+
"learning_rate": 2.40461025174401e-06,
|
| 1393 |
+
"loss": 0.0951,
|
| 1394 |
+
"step": 15500
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 14.103730664240219,
|
| 1398 |
+
"eval_cer": 0.15132463392447454,
|
| 1399 |
+
"eval_loss": 0.21604645252227783,
|
| 1400 |
+
"eval_runtime": 132.1387,
|
| 1401 |
+
"eval_samples_per_second": 37.839,
|
| 1402 |
+
"eval_steps_per_second": 1.188,
|
| 1403 |
+
"eval_wer": 0.29856827886467024,
|
| 1404 |
+
"step": 15500
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 14.194722474977253,
|
| 1408 |
+
"grad_norm": 1.1974434852600098,
|
| 1409 |
+
"learning_rate": 2.16196542311192e-06,
|
| 1410 |
+
"loss": 0.1019,
|
| 1411 |
+
"step": 15600
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 14.285714285714286,
|
| 1415 |
+
"grad_norm": 0.5327410697937012,
|
| 1416 |
+
"learning_rate": 1.9193205944798306e-06,
|
| 1417 |
+
"loss": 0.0968,
|
| 1418 |
+
"step": 15700
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 14.376706096451318,
|
| 1422 |
+
"grad_norm": 0.8405203819274902,
|
| 1423 |
+
"learning_rate": 1.6766757658477407e-06,
|
| 1424 |
+
"loss": 0.0941,
|
| 1425 |
+
"step": 15800
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 14.467697907188352,
|
| 1429 |
+
"grad_norm": 0.38401368260383606,
|
| 1430 |
+
"learning_rate": 1.4340309372156508e-06,
|
| 1431 |
+
"loss": 0.1036,
|
| 1432 |
+
"step": 15900
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 14.558689717925386,
|
| 1436 |
+
"grad_norm": 1.1276684999465942,
|
| 1437 |
+
"learning_rate": 1.191386108583561e-06,
|
| 1438 |
+
"loss": 0.1035,
|
| 1439 |
+
"step": 16000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 14.558689717925386,
|
| 1443 |
+
"eval_cer": 0.15134475902208125,
|
| 1444 |
+
"eval_loss": 0.21339672803878784,
|
| 1445 |
+
"eval_runtime": 132.1196,
|
| 1446 |
+
"eval_samples_per_second": 37.844,
|
| 1447 |
+
"eval_steps_per_second": 1.188,
|
| 1448 |
+
"eval_wer": 0.29856827886467024,
|
| 1449 |
+
"step": 16000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 14.64968152866242,
|
| 1453 |
+
"grad_norm": 0.5266655683517456,
|
| 1454 |
+
"learning_rate": 9.487412799514711e-07,
|
| 1455 |
+
"loss": 0.098,
|
| 1456 |
+
"step": 16100
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 14.740673339399454,
|
| 1460 |
+
"grad_norm": 0.8043445348739624,
|
| 1461 |
+
"learning_rate": 7.060964513193813e-07,
|
| 1462 |
+
"loss": 0.1005,
|
| 1463 |
+
"step": 16200
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 14.831665150136487,
|
| 1467 |
+
"grad_norm": 1.0907572507858276,
|
| 1468 |
+
"learning_rate": 4.6345162268729147e-07,
|
| 1469 |
+
"loss": 0.1023,
|
| 1470 |
+
"step": 16300
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 14.922656960873521,
|
| 1474 |
+
"grad_norm": 2.0936965942382812,
|
| 1475 |
+
"learning_rate": 2.2080679405520171e-07,
|
| 1476 |
+
"loss": 0.0963,
|
| 1477 |
+
"step": 16400
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 15.0,
|
| 1481 |
+
"step": 16485,
|
| 1482 |
+
"total_flos": 1.21495045308783e+20,
|
| 1483 |
+
"train_loss": 0.13136799893019088,
|
| 1484 |
+
"train_runtime": 24963.2457,
|
| 1485 |
+
"train_samples_per_second": 21.117,
|
| 1486 |
+
"train_steps_per_second": 0.66
|
| 1487 |
+
}
|
| 1488 |
+
],
|
| 1489 |
+
"logging_steps": 100,
|
| 1490 |
+
"max_steps": 16485,
|
| 1491 |
+
"num_input_tokens_seen": 0,
|
| 1492 |
+
"num_train_epochs": 15,
|
| 1493 |
+
"save_steps": 500,
|
| 1494 |
+
"stateful_callbacks": {
|
| 1495 |
+
"TrainerControl": {
|
| 1496 |
+
"args": {
|
| 1497 |
+
"should_epoch_stop": false,
|
| 1498 |
+
"should_evaluate": false,
|
| 1499 |
+
"should_log": false,
|
| 1500 |
+
"should_save": true,
|
| 1501 |
+
"should_training_stop": true
|
| 1502 |
+
},
|
| 1503 |
+
"attributes": {}
|
| 1504 |
+
}
|
| 1505 |
+
},
|
| 1506 |
+
"total_flos": 1.21495045308783e+20,
|
| 1507 |
+
"train_batch_size": 16,
|
| 1508 |
+
"trial_name": null,
|
| 1509 |
+
"trial_params": null
|
| 1510 |
+
}
|