End of training
Browse files- README.md +7 -3
- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +1316 -0
README.md
CHANGED
|
@@ -1,8 +1,12 @@
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
|
|
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: rinna/japanese-hubert-base
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- wer
|
|
@@ -16,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# Hubert-noisy_common_voice_debug
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [rinna/japanese-hubert-base](https://huggingface.co/rinna/japanese-hubert-base) on the
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
- Wer: 1.0
|
| 23 |
-
- Cer: 0.
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
+
language:
|
| 4 |
+
- ja
|
| 5 |
license: apache-2.0
|
| 6 |
base_model: rinna/japanese-hubert-base
|
| 7 |
tags:
|
| 8 |
+
- automatic-speech-recognition
|
| 9 |
+
- original_noisy_common_voice
|
| 10 |
- generated_from_trainer
|
| 11 |
metrics:
|
| 12 |
- wer
|
|
|
|
| 20 |
|
| 21 |
# Hubert-noisy_common_voice_debug
|
| 22 |
|
| 23 |
+
This model is a fine-tuned version of [rinna/japanese-hubert-base](https://huggingface.co/rinna/japanese-hubert-base) on the ORIGINAL_NOISY_COMMON_VOICE - JA dataset.
|
| 24 |
It achieves the following results on the evaluation set:
|
| 25 |
+
- Loss: 0.9787
|
| 26 |
- Wer: 1.0
|
| 27 |
+
- Cer: 0.3145
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_cer": 0.31446325281076837,
|
| 4 |
+
"eval_loss": 0.9786815643310547,
|
| 5 |
+
"eval_runtime": 151.6459,
|
| 6 |
+
"eval_samples": 4961,
|
| 7 |
+
"eval_samples_per_second": 32.714,
|
| 8 |
+
"eval_steps_per_second": 4.095,
|
| 9 |
+
"eval_wer": 1.0,
|
| 10 |
+
"total_flos": 1.1567283350118728e+19,
|
| 11 |
+
"train_loss": 1.009270377192937,
|
| 12 |
+
"train_runtime": 42299.7955,
|
| 13 |
+
"train_samples": 12032,
|
| 14 |
+
"train_samples_per_second": 8.533,
|
| 15 |
+
"train_steps_per_second": 0.267
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_cer": 0.31446325281076837,
|
| 4 |
+
"eval_loss": 0.9786815643310547,
|
| 5 |
+
"eval_runtime": 151.6459,
|
| 6 |
+
"eval_samples": 4961,
|
| 7 |
+
"eval_samples_per_second": 32.714,
|
| 8 |
+
"eval_steps_per_second": 4.095,
|
| 9 |
+
"eval_wer": 1.0
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"total_flos": 1.1567283350118728e+19,
|
| 4 |
+
"train_loss": 1.009270377192937,
|
| 5 |
+
"train_runtime": 42299.7955,
|
| 6 |
+
"train_samples": 12032,
|
| 7 |
+
"train_samples_per_second": 8.533,
|
| 8 |
+
"train_steps_per_second": 0.267
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1316 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7032522559165955,
|
| 3 |
+
"best_model_checkpoint": "./Hubert-noisy_common_voice_debug/checkpoint-3700",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 11280,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.26595744680851063,
|
| 13 |
+
"eval_cer": 1.0538678168650442,
|
| 14 |
+
"eval_loss": 12.082180976867676,
|
| 15 |
+
"eval_runtime": 151.0989,
|
| 16 |
+
"eval_samples_per_second": 32.833,
|
| 17 |
+
"eval_steps_per_second": 4.11,
|
| 18 |
+
"eval_wer": 1.1570247933884297,
|
| 19 |
+
"step": 100
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.5319148936170213,
|
| 23 |
+
"eval_cer": 0.9816708830000044,
|
| 24 |
+
"eval_loss": 5.878890037536621,
|
| 25 |
+
"eval_runtime": 151.597,
|
| 26 |
+
"eval_samples_per_second": 32.725,
|
| 27 |
+
"eval_steps_per_second": 4.096,
|
| 28 |
+
"eval_wer": 1.0,
|
| 29 |
+
"step": 200
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.7978723404255319,
|
| 33 |
+
"eval_cer": 0.9816665437204115,
|
| 34 |
+
"eval_loss": 5.362728595733643,
|
| 35 |
+
"eval_runtime": 149.1154,
|
| 36 |
+
"eval_samples_per_second": 33.27,
|
| 37 |
+
"eval_steps_per_second": 4.165,
|
| 38 |
+
"eval_wer": 1.0,
|
| 39 |
+
"step": 300
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0638297872340425,
|
| 43 |
+
"eval_cer": 0.9816752222795971,
|
| 44 |
+
"eval_loss": 4.9316487312316895,
|
| 45 |
+
"eval_runtime": 149.2108,
|
| 46 |
+
"eval_samples_per_second": 33.248,
|
| 47 |
+
"eval_steps_per_second": 4.162,
|
| 48 |
+
"eval_wer": 1.0,
|
| 49 |
+
"step": 400
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 1.3297872340425532,
|
| 53 |
+
"grad_norm": 18.269367218017578,
|
| 54 |
+
"learning_rate": 1.1927999999999998e-05,
|
| 55 |
+
"loss": 6.372,
|
| 56 |
+
"step": 500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.3297872340425532,
|
| 60 |
+
"eval_cer": 0.9816752222795971,
|
| 61 |
+
"eval_loss": 4.455592632293701,
|
| 62 |
+
"eval_runtime": 149.4229,
|
| 63 |
+
"eval_samples_per_second": 33.201,
|
| 64 |
+
"eval_steps_per_second": 4.156,
|
| 65 |
+
"eval_wer": 1.0,
|
| 66 |
+
"step": 500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.5957446808510638,
|
| 70 |
+
"eval_cer": 0.9816752222795971,
|
| 71 |
+
"eval_loss": 3.989032506942749,
|
| 72 |
+
"eval_runtime": 148.7951,
|
| 73 |
+
"eval_samples_per_second": 33.341,
|
| 74 |
+
"eval_steps_per_second": 4.174,
|
| 75 |
+
"eval_wer": 1.0,
|
| 76 |
+
"step": 600
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 1.8617021276595744,
|
| 80 |
+
"eval_cer": 0.9816622044408188,
|
| 81 |
+
"eval_loss": 3.57344126701355,
|
| 82 |
+
"eval_runtime": 148.5583,
|
| 83 |
+
"eval_samples_per_second": 33.394,
|
| 84 |
+
"eval_steps_per_second": 4.18,
|
| 85 |
+
"eval_wer": 1.0,
|
| 86 |
+
"step": 700
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.127659574468085,
|
| 90 |
+
"eval_cer": 0.9816708830000044,
|
| 91 |
+
"eval_loss": 3.2931714057922363,
|
| 92 |
+
"eval_runtime": 148.3742,
|
| 93 |
+
"eval_samples_per_second": 33.436,
|
| 94 |
+
"eval_steps_per_second": 4.185,
|
| 95 |
+
"eval_wer": 1.0,
|
| 96 |
+
"step": 800
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 2.393617021276596,
|
| 100 |
+
"eval_cer": 0.9816839008387828,
|
| 101 |
+
"eval_loss": 3.1536319255828857,
|
| 102 |
+
"eval_runtime": 148.8499,
|
| 103 |
+
"eval_samples_per_second": 33.329,
|
| 104 |
+
"eval_steps_per_second": 4.172,
|
| 105 |
+
"eval_wer": 1.0,
|
| 106 |
+
"step": 900
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 2.6595744680851063,
|
| 110 |
+
"grad_norm": 1.548904299736023,
|
| 111 |
+
"learning_rate": 2.3928e-05,
|
| 112 |
+
"loss": 3.4101,
|
| 113 |
+
"step": 1000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.6595744680851063,
|
| 117 |
+
"eval_cer": 0.9816622044408188,
|
| 118 |
+
"eval_loss": 3.0483851432800293,
|
| 119 |
+
"eval_runtime": 148.9657,
|
| 120 |
+
"eval_samples_per_second": 33.303,
|
| 121 |
+
"eval_steps_per_second": 4.169,
|
| 122 |
+
"eval_wer": 1.0,
|
| 123 |
+
"step": 1000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.925531914893617,
|
| 127 |
+
"eval_cer": 0.9807552950059231,
|
| 128 |
+
"eval_loss": 2.8469536304473877,
|
| 129 |
+
"eval_runtime": 148.996,
|
| 130 |
+
"eval_samples_per_second": 33.296,
|
| 131 |
+
"eval_steps_per_second": 4.168,
|
| 132 |
+
"eval_wer": 1.0,
|
| 133 |
+
"step": 1100
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 3.1914893617021276,
|
| 137 |
+
"eval_cer": 0.8702078080996993,
|
| 138 |
+
"eval_loss": 2.521139144897461,
|
| 139 |
+
"eval_runtime": 149.4933,
|
| 140 |
+
"eval_samples_per_second": 33.185,
|
| 141 |
+
"eval_steps_per_second": 4.154,
|
| 142 |
+
"eval_wer": 1.0,
|
| 143 |
+
"step": 1200
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.4574468085106385,
|
| 147 |
+
"eval_cer": 0.5518348643758163,
|
| 148 |
+
"eval_loss": 2.035446882247925,
|
| 149 |
+
"eval_runtime": 150.325,
|
| 150 |
+
"eval_samples_per_second": 33.002,
|
| 151 |
+
"eval_steps_per_second": 4.131,
|
| 152 |
+
"eval_wer": 1.0,
|
| 153 |
+
"step": 1300
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 3.723404255319149,
|
| 157 |
+
"eval_cer": 0.44136548450226293,
|
| 158 |
+
"eval_loss": 1.678019404411316,
|
| 159 |
+
"eval_runtime": 150.3487,
|
| 160 |
+
"eval_samples_per_second": 32.997,
|
| 161 |
+
"eval_steps_per_second": 4.13,
|
| 162 |
+
"eval_wer": 1.0,
|
| 163 |
+
"step": 1400
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 3.9893617021276597,
|
| 167 |
+
"grad_norm": 4.6614580154418945,
|
| 168 |
+
"learning_rate": 3.5928e-05,
|
| 169 |
+
"loss": 2.3222,
|
| 170 |
+
"step": 1500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.9893617021276597,
|
| 174 |
+
"eval_cer": 0.43124194521225584,
|
| 175 |
+
"eval_loss": 1.5038928985595703,
|
| 176 |
+
"eval_runtime": 150.8122,
|
| 177 |
+
"eval_samples_per_second": 32.895,
|
| 178 |
+
"eval_steps_per_second": 4.118,
|
| 179 |
+
"eval_wer": 1.0,
|
| 180 |
+
"step": 1500
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 4.25531914893617,
|
| 184 |
+
"eval_cer": 0.396458279996355,
|
| 185 |
+
"eval_loss": 1.3418982028961182,
|
| 186 |
+
"eval_runtime": 150.4572,
|
| 187 |
+
"eval_samples_per_second": 32.973,
|
| 188 |
+
"eval_steps_per_second": 4.127,
|
| 189 |
+
"eval_wer": 1.0,
|
| 190 |
+
"step": 1600
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 4.5212765957446805,
|
| 194 |
+
"eval_cer": 0.3685610514942309,
|
| 195 |
+
"eval_loss": 1.2053722143173218,
|
| 196 |
+
"eval_runtime": 150.9364,
|
| 197 |
+
"eval_samples_per_second": 32.868,
|
| 198 |
+
"eval_steps_per_second": 4.114,
|
| 199 |
+
"eval_wer": 1.0,
|
| 200 |
+
"step": 1700
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.787234042553192,
|
| 204 |
+
"eval_cer": 0.3321067636351013,
|
| 205 |
+
"eval_loss": 1.0588372945785522,
|
| 206 |
+
"eval_runtime": 150.9127,
|
| 207 |
+
"eval_samples_per_second": 32.873,
|
| 208 |
+
"eval_steps_per_second": 4.115,
|
| 209 |
+
"eval_wer": 1.0,
|
| 210 |
+
"step": 1800
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.053191489361702,
|
| 214 |
+
"eval_cer": 0.31576069740901613,
|
| 215 |
+
"eval_loss": 0.9545913338661194,
|
| 216 |
+
"eval_runtime": 150.8858,
|
| 217 |
+
"eval_samples_per_second": 32.879,
|
| 218 |
+
"eval_steps_per_second": 4.116,
|
| 219 |
+
"eval_wer": 1.0,
|
| 220 |
+
"step": 1900
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 5.319148936170213,
|
| 224 |
+
"grad_norm": 4.519205570220947,
|
| 225 |
+
"learning_rate": 4.7928e-05,
|
| 226 |
+
"loss": 1.2343,
|
| 227 |
+
"step": 2000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 5.319148936170213,
|
| 231 |
+
"eval_cer": 0.3105969546935818,
|
| 232 |
+
"eval_loss": 0.9041823148727417,
|
| 233 |
+
"eval_runtime": 150.7252,
|
| 234 |
+
"eval_samples_per_second": 32.914,
|
| 235 |
+
"eval_steps_per_second": 4.12,
|
| 236 |
+
"eval_wer": 1.0,
|
| 237 |
+
"step": 2000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 5.585106382978723,
|
| 241 |
+
"eval_cer": 0.30879615366256896,
|
| 242 |
+
"eval_loss": 0.8747320771217346,
|
| 243 |
+
"eval_runtime": 151.661,
|
| 244 |
+
"eval_samples_per_second": 32.711,
|
| 245 |
+
"eval_steps_per_second": 4.095,
|
| 246 |
+
"eval_wer": 1.0,
|
| 247 |
+
"step": 2100
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 5.851063829787234,
|
| 251 |
+
"eval_cer": 0.2971885807518236,
|
| 252 |
+
"eval_loss": 0.8223557472229004,
|
| 253 |
+
"eval_runtime": 155.0175,
|
| 254 |
+
"eval_samples_per_second": 32.003,
|
| 255 |
+
"eval_steps_per_second": 4.006,
|
| 256 |
+
"eval_wer": 1.0,
|
| 257 |
+
"step": 2200
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 6.117021276595745,
|
| 261 |
+
"eval_cer": 0.2995751845278647,
|
| 262 |
+
"eval_loss": 0.8100646138191223,
|
| 263 |
+
"eval_runtime": 151.2405,
|
| 264 |
+
"eval_samples_per_second": 32.802,
|
| 265 |
+
"eval_steps_per_second": 4.106,
|
| 266 |
+
"eval_wer": 1.0,
|
| 267 |
+
"step": 2300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 6.382978723404255,
|
| 271 |
+
"eval_cer": 0.2969933131701475,
|
| 272 |
+
"eval_loss": 0.7891868948936462,
|
| 273 |
+
"eval_runtime": 151.3984,
|
| 274 |
+
"eval_samples_per_second": 32.768,
|
| 275 |
+
"eval_steps_per_second": 4.102,
|
| 276 |
+
"eval_wer": 1.0,
|
| 277 |
+
"step": 2400
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 6.648936170212766,
|
| 281 |
+
"grad_norm": 7.927753925323486,
|
| 282 |
+
"learning_rate": 5.9903999999999994e-05,
|
| 283 |
+
"loss": 0.8716,
|
| 284 |
+
"step": 2500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 6.648936170212766,
|
| 288 |
+
"eval_cer": 0.29149544592606735,
|
| 289 |
+
"eval_loss": 0.7661293148994446,
|
| 290 |
+
"eval_runtime": 152.5094,
|
| 291 |
+
"eval_samples_per_second": 32.529,
|
| 292 |
+
"eval_steps_per_second": 4.072,
|
| 293 |
+
"eval_wer": 1.0,
|
| 294 |
+
"step": 2500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 6.914893617021277,
|
| 298 |
+
"eval_cer": 0.28864019995400364,
|
| 299 |
+
"eval_loss": 0.7654036283493042,
|
| 300 |
+
"eval_runtime": 151.4659,
|
| 301 |
+
"eval_samples_per_second": 32.753,
|
| 302 |
+
"eval_steps_per_second": 4.1,
|
| 303 |
+
"eval_wer": 1.0,
|
| 304 |
+
"step": 2600
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 7.180851063829787,
|
| 308 |
+
"eval_cer": 0.28976841264813213,
|
| 309 |
+
"eval_loss": 0.767704963684082,
|
| 310 |
+
"eval_runtime": 151.3213,
|
| 311 |
+
"eval_samples_per_second": 32.785,
|
| 312 |
+
"eval_steps_per_second": 4.104,
|
| 313 |
+
"eval_wer": 1.0,
|
| 314 |
+
"step": 2700
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 7.446808510638298,
|
| 318 |
+
"eval_cer": 0.28610606067180727,
|
| 319 |
+
"eval_loss": 0.752834141254425,
|
| 320 |
+
"eval_runtime": 151.5246,
|
| 321 |
+
"eval_samples_per_second": 32.741,
|
| 322 |
+
"eval_steps_per_second": 4.098,
|
| 323 |
+
"eval_wer": 1.0,
|
| 324 |
+
"step": 2800
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 7.712765957446808,
|
| 328 |
+
"eval_cer": 0.28800232585386176,
|
| 329 |
+
"eval_loss": 0.7432949542999268,
|
| 330 |
+
"eval_runtime": 155.4404,
|
| 331 |
+
"eval_samples_per_second": 31.916,
|
| 332 |
+
"eval_steps_per_second": 3.995,
|
| 333 |
+
"eval_wer": 1.0,
|
| 334 |
+
"step": 2900
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 7.9787234042553195,
|
| 338 |
+
"grad_norm": 6.501748085021973,
|
| 339 |
+
"learning_rate": 7.1904e-05,
|
| 340 |
+
"loss": 0.7324,
|
| 341 |
+
"step": 3000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 7.9787234042553195,
|
| 345 |
+
"eval_cer": 0.2876595227660304,
|
| 346 |
+
"eval_loss": 0.7498476505279541,
|
| 347 |
+
"eval_runtime": 151.1874,
|
| 348 |
+
"eval_samples_per_second": 32.814,
|
| 349 |
+
"eval_steps_per_second": 4.107,
|
| 350 |
+
"eval_wer": 1.0,
|
| 351 |
+
"step": 3000
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 8.24468085106383,
|
| 355 |
+
"eval_cer": 0.2827474582669785,
|
| 356 |
+
"eval_loss": 0.7267456650733948,
|
| 357 |
+
"eval_runtime": 151.7111,
|
| 358 |
+
"eval_samples_per_second": 32.7,
|
| 359 |
+
"eval_steps_per_second": 4.093,
|
| 360 |
+
"eval_wer": 1.0,
|
| 361 |
+
"step": 3100
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 8.51063829787234,
|
| 365 |
+
"eval_cer": 0.28125474608705464,
|
| 366 |
+
"eval_loss": 0.7319472432136536,
|
| 367 |
+
"eval_runtime": 151.6994,
|
| 368 |
+
"eval_samples_per_second": 32.703,
|
| 369 |
+
"eval_steps_per_second": 4.094,
|
| 370 |
+
"eval_wer": 1.0,
|
| 371 |
+
"step": 3200
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 8.77659574468085,
|
| 375 |
+
"eval_cer": 0.28822796839268744,
|
| 376 |
+
"eval_loss": 0.7477809190750122,
|
| 377 |
+
"eval_runtime": 151.7004,
|
| 378 |
+
"eval_samples_per_second": 32.703,
|
| 379 |
+
"eval_steps_per_second": 4.094,
|
| 380 |
+
"eval_wer": 1.0,
|
| 381 |
+
"step": 3300
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 9.042553191489361,
|
| 385 |
+
"eval_cer": 0.2814543529483235,
|
| 386 |
+
"eval_loss": 0.7336912751197815,
|
| 387 |
+
"eval_runtime": 151.8383,
|
| 388 |
+
"eval_samples_per_second": 32.673,
|
| 389 |
+
"eval_steps_per_second": 4.09,
|
| 390 |
+
"eval_wer": 1.0,
|
| 391 |
+
"step": 3400
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 9.308510638297872,
|
| 395 |
+
"grad_norm": 6.825859546661377,
|
| 396 |
+
"learning_rate": 8.390399999999999e-05,
|
| 397 |
+
"loss": 0.6486,
|
| 398 |
+
"step": 3500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 9.308510638297872,
|
| 402 |
+
"eval_cer": 0.28514274060220524,
|
| 403 |
+
"eval_loss": 0.734130322933197,
|
| 404 |
+
"eval_runtime": 151.4163,
|
| 405 |
+
"eval_samples_per_second": 32.764,
|
| 406 |
+
"eval_steps_per_second": 4.101,
|
| 407 |
+
"eval_wer": 1.0,
|
| 408 |
+
"step": 3500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 9.574468085106384,
|
| 412 |
+
"eval_cer": 0.2803218009746022,
|
| 413 |
+
"eval_loss": 0.74192875623703,
|
| 414 |
+
"eval_runtime": 151.3965,
|
| 415 |
+
"eval_samples_per_second": 32.768,
|
| 416 |
+
"eval_steps_per_second": 4.102,
|
| 417 |
+
"eval_wer": 1.0,
|
| 418 |
+
"step": 3600
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 9.840425531914894,
|
| 422 |
+
"eval_cer": 0.277323358775976,
|
| 423 |
+
"eval_loss": 0.7032522559165955,
|
| 424 |
+
"eval_runtime": 151.1826,
|
| 425 |
+
"eval_samples_per_second": 32.815,
|
| 426 |
+
"eval_steps_per_second": 4.108,
|
| 427 |
+
"eval_wer": 0.9997984277363435,
|
| 428 |
+
"step": 3700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 10.106382978723405,
|
| 432 |
+
"eval_cer": 0.2829253687302834,
|
| 433 |
+
"eval_loss": 0.7326785922050476,
|
| 434 |
+
"eval_runtime": 151.1178,
|
| 435 |
+
"eval_samples_per_second": 32.829,
|
| 436 |
+
"eval_steps_per_second": 4.109,
|
| 437 |
+
"eval_wer": 1.0,
|
| 438 |
+
"step": 3800
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 10.372340425531915,
|
| 442 |
+
"eval_cer": 0.2855332757655574,
|
| 443 |
+
"eval_loss": 0.7554399967193604,
|
| 444 |
+
"eval_runtime": 151.3763,
|
| 445 |
+
"eval_samples_per_second": 32.773,
|
| 446 |
+
"eval_steps_per_second": 4.102,
|
| 447 |
+
"eval_wer": 0.9997984277363435,
|
| 448 |
+
"step": 3900
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 10.638297872340425,
|
| 452 |
+
"grad_norm": 5.65977144241333,
|
| 453 |
+
"learning_rate": 9.5904e-05,
|
| 454 |
+
"loss": 0.6034,
|
| 455 |
+
"step": 4000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 10.638297872340425,
|
| 459 |
+
"eval_cer": 0.2841360277366752,
|
| 460 |
+
"eval_loss": 0.7361173033714294,
|
| 461 |
+
"eval_runtime": 151.2103,
|
| 462 |
+
"eval_samples_per_second": 32.809,
|
| 463 |
+
"eval_steps_per_second": 4.107,
|
| 464 |
+
"eval_wer": 1.0,
|
| 465 |
+
"step": 4000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 10.904255319148936,
|
| 469 |
+
"eval_cer": 0.2833202431732284,
|
| 470 |
+
"eval_loss": 0.745939314365387,
|
| 471 |
+
"eval_runtime": 151.41,
|
| 472 |
+
"eval_samples_per_second": 32.765,
|
| 473 |
+
"eval_steps_per_second": 4.101,
|
| 474 |
+
"eval_wer": 1.0,
|
| 475 |
+
"step": 4100
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 11.170212765957446,
|
| 479 |
+
"eval_cer": 0.2801352119521117,
|
| 480 |
+
"eval_loss": 0.7383919358253479,
|
| 481 |
+
"eval_runtime": 151.4933,
|
| 482 |
+
"eval_samples_per_second": 32.747,
|
| 483 |
+
"eval_steps_per_second": 4.099,
|
| 484 |
+
"eval_wer": 1.0,
|
| 485 |
+
"step": 4200
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 11.436170212765958,
|
| 489 |
+
"eval_cer": 0.27755767987398733,
|
| 490 |
+
"eval_loss": 0.7336695194244385,
|
| 491 |
+
"eval_runtime": 151.6319,
|
| 492 |
+
"eval_samples_per_second": 32.717,
|
| 493 |
+
"eval_steps_per_second": 4.095,
|
| 494 |
+
"eval_wer": 1.0,
|
| 495 |
+
"step": 4300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 11.702127659574469,
|
| 499 |
+
"eval_cer": 0.2819099773055677,
|
| 500 |
+
"eval_loss": 0.7572408318519592,
|
| 501 |
+
"eval_runtime": 151.2617,
|
| 502 |
+
"eval_samples_per_second": 32.797,
|
| 503 |
+
"eval_steps_per_second": 4.105,
|
| 504 |
+
"eval_wer": 1.0,
|
| 505 |
+
"step": 4400
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 11.96808510638298,
|
| 509 |
+
"grad_norm": 5.250299453735352,
|
| 510 |
+
"learning_rate": 0.00010790399999999999,
|
| 511 |
+
"loss": 0.5687,
|
| 512 |
+
"step": 4500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 11.96808510638298,
|
| 516 |
+
"eval_cer": 0.2824393694158896,
|
| 517 |
+
"eval_loss": 0.7522115707397461,
|
| 518 |
+
"eval_runtime": 151.5116,
|
| 519 |
+
"eval_samples_per_second": 32.743,
|
| 520 |
+
"eval_steps_per_second": 4.099,
|
| 521 |
+
"eval_wer": 1.0,
|
| 522 |
+
"step": 4500
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 12.23404255319149,
|
| 526 |
+
"eval_cer": 0.27894624934368395,
|
| 527 |
+
"eval_loss": 0.7491241097450256,
|
| 528 |
+
"eval_runtime": 151.5087,
|
| 529 |
+
"eval_samples_per_second": 32.744,
|
| 530 |
+
"eval_steps_per_second": 4.099,
|
| 531 |
+
"eval_wer": 1.0,
|
| 532 |
+
"step": 4600
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 12.5,
|
| 536 |
+
"eval_cer": 0.2831987433446299,
|
| 537 |
+
"eval_loss": 0.7484813928604126,
|
| 538 |
+
"eval_runtime": 151.4358,
|
| 539 |
+
"eval_samples_per_second": 32.76,
|
| 540 |
+
"eval_steps_per_second": 4.101,
|
| 541 |
+
"eval_wer": 1.0,
|
| 542 |
+
"step": 4700
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 12.76595744680851,
|
| 546 |
+
"eval_cer": 0.2849040802246011,
|
| 547 |
+
"eval_loss": 0.7622763514518738,
|
| 548 |
+
"eval_runtime": 151.3314,
|
| 549 |
+
"eval_samples_per_second": 32.782,
|
| 550 |
+
"eval_steps_per_second": 4.104,
|
| 551 |
+
"eval_wer": 1.0,
|
| 552 |
+
"step": 4800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 13.03191489361702,
|
| 556 |
+
"eval_cer": 0.285936828767688,
|
| 557 |
+
"eval_loss": 0.7829335331916809,
|
| 558 |
+
"eval_runtime": 151.6153,
|
| 559 |
+
"eval_samples_per_second": 32.721,
|
| 560 |
+
"eval_steps_per_second": 4.096,
|
| 561 |
+
"eval_wer": 1.0,
|
| 562 |
+
"step": 4900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 13.297872340425531,
|
| 566 |
+
"grad_norm": 4.461331844329834,
|
| 567 |
+
"learning_rate": 0.00011990399999999998,
|
| 568 |
+
"loss": 0.5255,
|
| 569 |
+
"step": 5000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 13.297872340425531,
|
| 573 |
+
"eval_cer": 0.28200110217701657,
|
| 574 |
+
"eval_loss": 0.7818682789802551,
|
| 575 |
+
"eval_runtime": 151.2326,
|
| 576 |
+
"eval_samples_per_second": 32.804,
|
| 577 |
+
"eval_steps_per_second": 4.106,
|
| 578 |
+
"eval_wer": 1.0,
|
| 579 |
+
"step": 5000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 13.563829787234042,
|
| 583 |
+
"eval_cer": 0.2824393694158896,
|
| 584 |
+
"eval_loss": 0.7782630324363708,
|
| 585 |
+
"eval_runtime": 151.6654,
|
| 586 |
+
"eval_samples_per_second": 32.71,
|
| 587 |
+
"eval_steps_per_second": 4.095,
|
| 588 |
+
"eval_wer": 0.9997984277363435,
|
| 589 |
+
"step": 5100
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 13.829787234042554,
|
| 593 |
+
"eval_cer": 0.2840275457468551,
|
| 594 |
+
"eval_loss": 0.7653459906578064,
|
| 595 |
+
"eval_runtime": 151.0873,
|
| 596 |
+
"eval_samples_per_second": 32.835,
|
| 597 |
+
"eval_steps_per_second": 4.11,
|
| 598 |
+
"eval_wer": 1.0,
|
| 599 |
+
"step": 5200
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 14.095744680851064,
|
| 603 |
+
"eval_cer": 0.28220938759747105,
|
| 604 |
+
"eval_loss": 0.7816305756568909,
|
| 605 |
+
"eval_runtime": 151.1619,
|
| 606 |
+
"eval_samples_per_second": 32.819,
|
| 607 |
+
"eval_steps_per_second": 4.108,
|
| 608 |
+
"eval_wer": 1.0,
|
| 609 |
+
"step": 5300
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 14.361702127659575,
|
| 613 |
+
"eval_cer": 0.2824133337383328,
|
| 614 |
+
"eval_loss": 0.7607858180999756,
|
| 615 |
+
"eval_runtime": 150.8484,
|
| 616 |
+
"eval_samples_per_second": 32.887,
|
| 617 |
+
"eval_steps_per_second": 4.117,
|
| 618 |
+
"eval_wer": 1.0,
|
| 619 |
+
"step": 5400
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 14.627659574468085,
|
| 623 |
+
"grad_norm": 4.656615257263184,
|
| 624 |
+
"learning_rate": 0.000131904,
|
| 625 |
+
"loss": 0.5016,
|
| 626 |
+
"step": 5500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 14.627659574468085,
|
| 630 |
+
"eval_cer": 0.28411433133871117,
|
| 631 |
+
"eval_loss": 0.7712321281433105,
|
| 632 |
+
"eval_runtime": 151.045,
|
| 633 |
+
"eval_samples_per_second": 32.845,
|
| 634 |
+
"eval_steps_per_second": 4.111,
|
| 635 |
+
"eval_wer": 0.9997984277363435,
|
| 636 |
+
"step": 5500
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 14.893617021276595,
|
| 640 |
+
"eval_cer": 0.286396792404525,
|
| 641 |
+
"eval_loss": 0.7712020874023438,
|
| 642 |
+
"eval_runtime": 151.0967,
|
| 643 |
+
"eval_samples_per_second": 32.833,
|
| 644 |
+
"eval_steps_per_second": 4.11,
|
| 645 |
+
"eval_wer": 1.0,
|
| 646 |
+
"step": 5600
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 15.159574468085106,
|
| 650 |
+
"eval_cer": 0.285060294289942,
|
| 651 |
+
"eval_loss": 0.8153163194656372,
|
| 652 |
+
"eval_runtime": 151.4841,
|
| 653 |
+
"eval_samples_per_second": 32.749,
|
| 654 |
+
"eval_steps_per_second": 4.099,
|
| 655 |
+
"eval_wer": 0.999596855472687,
|
| 656 |
+
"step": 5700
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 15.425531914893616,
|
| 660 |
+
"eval_cer": 0.2852338654736541,
|
| 661 |
+
"eval_loss": 0.8161126375198364,
|
| 662 |
+
"eval_runtime": 150.9671,
|
| 663 |
+
"eval_samples_per_second": 32.861,
|
| 664 |
+
"eval_steps_per_second": 4.113,
|
| 665 |
+
"eval_wer": 0.9997984277363435,
|
| 666 |
+
"step": 5800
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 15.691489361702128,
|
| 670 |
+
"eval_cer": 0.28826702190902265,
|
| 671 |
+
"eval_loss": 0.7910758852958679,
|
| 672 |
+
"eval_runtime": 151.0378,
|
| 673 |
+
"eval_samples_per_second": 32.846,
|
| 674 |
+
"eval_steps_per_second": 4.112,
|
| 675 |
+
"eval_wer": 1.0,
|
| 676 |
+
"step": 5900
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 15.957446808510639,
|
| 680 |
+
"grad_norm": 5.0891499519348145,
|
| 681 |
+
"learning_rate": 0.00014390399999999998,
|
| 682 |
+
"loss": 0.4821,
|
| 683 |
+
"step": 6000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 15.957446808510639,
|
| 687 |
+
"eval_cer": 0.28232654814647673,
|
| 688 |
+
"eval_loss": 0.7926108241081238,
|
| 689 |
+
"eval_runtime": 151.2593,
|
| 690 |
+
"eval_samples_per_second": 32.798,
|
| 691 |
+
"eval_steps_per_second": 4.106,
|
| 692 |
+
"eval_wer": 1.0,
|
| 693 |
+
"step": 6000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 16.22340425531915,
|
| 697 |
+
"eval_cer": 0.28674393477194915,
|
| 698 |
+
"eval_loss": 0.814721405506134,
|
| 699 |
+
"eval_runtime": 150.8029,
|
| 700 |
+
"eval_samples_per_second": 32.897,
|
| 701 |
+
"eval_steps_per_second": 4.118,
|
| 702 |
+
"eval_wer": 1.0,
|
| 703 |
+
"step": 6100
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 16.48936170212766,
|
| 707 |
+
"eval_cer": 0.28262161915878725,
|
| 708 |
+
"eval_loss": 0.7699744701385498,
|
| 709 |
+
"eval_runtime": 150.7429,
|
| 710 |
+
"eval_samples_per_second": 32.91,
|
| 711 |
+
"eval_steps_per_second": 4.12,
|
| 712 |
+
"eval_wer": 1.0,
|
| 713 |
+
"step": 6200
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 16.75531914893617,
|
| 717 |
+
"eval_cer": 0.2909530359769671,
|
| 718 |
+
"eval_loss": 0.8119201064109802,
|
| 719 |
+
"eval_runtime": 150.6966,
|
| 720 |
+
"eval_samples_per_second": 32.92,
|
| 721 |
+
"eval_steps_per_second": 4.121,
|
| 722 |
+
"eval_wer": 1.0,
|
| 723 |
+
"step": 6300
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 17.02127659574468,
|
| 727 |
+
"eval_cer": 0.2845873128143266,
|
| 728 |
+
"eval_loss": 0.835509181022644,
|
| 729 |
+
"eval_runtime": 151.1321,
|
| 730 |
+
"eval_samples_per_second": 32.826,
|
| 731 |
+
"eval_steps_per_second": 4.109,
|
| 732 |
+
"eval_wer": 1.0,
|
| 733 |
+
"step": 6400
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 17.28723404255319,
|
| 737 |
+
"grad_norm": 4.005585670471191,
|
| 738 |
+
"learning_rate": 0.00015587999999999998,
|
| 739 |
+
"loss": 0.4503,
|
| 740 |
+
"step": 6500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 17.28723404255319,
|
| 744 |
+
"eval_cer": 0.28592381092890956,
|
| 745 |
+
"eval_loss": 0.793612539768219,
|
| 746 |
+
"eval_runtime": 150.9291,
|
| 747 |
+
"eval_samples_per_second": 32.87,
|
| 748 |
+
"eval_steps_per_second": 4.115,
|
| 749 |
+
"eval_wer": 0.9997984277363435,
|
| 750 |
+
"step": 6500
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 17.5531914893617,
|
| 754 |
+
"eval_cer": 0.29515779790239227,
|
| 755 |
+
"eval_loss": 0.7976419925689697,
|
| 756 |
+
"eval_runtime": 169.2668,
|
| 757 |
+
"eval_samples_per_second": 29.309,
|
| 758 |
+
"eval_steps_per_second": 3.669,
|
| 759 |
+
"eval_wer": 0.9997984277363435,
|
| 760 |
+
"step": 6600
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 17.819148936170212,
|
| 764 |
+
"eval_cer": 0.2901546085318915,
|
| 765 |
+
"eval_loss": 0.8274039626121521,
|
| 766 |
+
"eval_runtime": 158.3816,
|
| 767 |
+
"eval_samples_per_second": 31.323,
|
| 768 |
+
"eval_steps_per_second": 3.921,
|
| 769 |
+
"eval_wer": 0.9997984277363435,
|
| 770 |
+
"step": 6700
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 18.085106382978722,
|
| 774 |
+
"eval_cer": 0.2885490750825548,
|
| 775 |
+
"eval_loss": 0.903409481048584,
|
| 776 |
+
"eval_runtime": 151.19,
|
| 777 |
+
"eval_samples_per_second": 32.813,
|
| 778 |
+
"eval_steps_per_second": 4.107,
|
| 779 |
+
"eval_wer": 0.9997984277363435,
|
| 780 |
+
"step": 6800
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 18.351063829787233,
|
| 784 |
+
"eval_cer": 0.28819325415594504,
|
| 785 |
+
"eval_loss": 0.8065901398658752,
|
| 786 |
+
"eval_runtime": 152.3883,
|
| 787 |
+
"eval_samples_per_second": 32.555,
|
| 788 |
+
"eval_steps_per_second": 4.075,
|
| 789 |
+
"eval_wer": 0.9997984277363435,
|
| 790 |
+
"step": 6900
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 18.617021276595743,
|
| 794 |
+
"grad_norm": 4.851210117340088,
|
| 795 |
+
"learning_rate": 0.000167856,
|
| 796 |
+
"loss": 0.4435,
|
| 797 |
+
"step": 7000
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 18.617021276595743,
|
| 801 |
+
"eval_cer": 0.29208124867109564,
|
| 802 |
+
"eval_loss": 0.849502682685852,
|
| 803 |
+
"eval_runtime": 171.4215,
|
| 804 |
+
"eval_samples_per_second": 28.94,
|
| 805 |
+
"eval_steps_per_second": 3.623,
|
| 806 |
+
"eval_wer": 1.0,
|
| 807 |
+
"step": 7000
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 18.882978723404257,
|
| 811 |
+
"eval_cer": 0.2896121985827913,
|
| 812 |
+
"eval_loss": 0.8448098301887512,
|
| 813 |
+
"eval_runtime": 151.0183,
|
| 814 |
+
"eval_samples_per_second": 32.85,
|
| 815 |
+
"eval_steps_per_second": 4.112,
|
| 816 |
+
"eval_wer": 0.9997984277363435,
|
| 817 |
+
"step": 7100
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 19.148936170212767,
|
| 821 |
+
"eval_cer": 0.29039760818908844,
|
| 822 |
+
"eval_loss": 0.877412736415863,
|
| 823 |
+
"eval_runtime": 155.1268,
|
| 824 |
+
"eval_samples_per_second": 31.98,
|
| 825 |
+
"eval_steps_per_second": 4.003,
|
| 826 |
+
"eval_wer": 1.0,
|
| 827 |
+
"step": 7200
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 19.414893617021278,
|
| 831 |
+
"eval_cer": 0.29732309841920046,
|
| 832 |
+
"eval_loss": 0.8293471336364746,
|
| 833 |
+
"eval_runtime": 151.2836,
|
| 834 |
+
"eval_samples_per_second": 32.793,
|
| 835 |
+
"eval_steps_per_second": 4.105,
|
| 836 |
+
"eval_wer": 0.9997984277363435,
|
| 837 |
+
"step": 7300
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 19.680851063829788,
|
| 841 |
+
"eval_cer": 0.2924804623936334,
|
| 842 |
+
"eval_loss": 0.8037762641906738,
|
| 843 |
+
"eval_runtime": 151.1502,
|
| 844 |
+
"eval_samples_per_second": 32.822,
|
| 845 |
+
"eval_steps_per_second": 4.108,
|
| 846 |
+
"eval_wer": 1.0,
|
| 847 |
+
"step": 7400
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 19.9468085106383,
|
| 851 |
+
"grad_norm": 3.5915985107421875,
|
| 852 |
+
"learning_rate": 0.000179856,
|
| 853 |
+
"loss": 0.4457,
|
| 854 |
+
"step": 7500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 19.9468085106383,
|
| 858 |
+
"eval_cer": 0.29077512551366225,
|
| 859 |
+
"eval_loss": 0.806207537651062,
|
| 860 |
+
"eval_runtime": 154.4071,
|
| 861 |
+
"eval_samples_per_second": 32.129,
|
| 862 |
+
"eval_steps_per_second": 4.022,
|
| 863 |
+
"eval_wer": 0.9997984277363435,
|
| 864 |
+
"step": 7500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 20.21276595744681,
|
| 868 |
+
"eval_cer": 0.29176448126082105,
|
| 869 |
+
"eval_loss": 0.8740444779396057,
|
| 870 |
+
"eval_runtime": 154.5779,
|
| 871 |
+
"eval_samples_per_second": 32.094,
|
| 872 |
+
"eval_steps_per_second": 4.017,
|
| 873 |
+
"eval_wer": 1.0,
|
| 874 |
+
"step": 7600
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 20.47872340425532,
|
| 878 |
+
"eval_cer": 0.2977179728621454,
|
| 879 |
+
"eval_loss": 0.848852276802063,
|
| 880 |
+
"eval_runtime": 173.0214,
|
| 881 |
+
"eval_samples_per_second": 28.673,
|
| 882 |
+
"eval_steps_per_second": 3.589,
|
| 883 |
+
"eval_wer": 1.0,
|
| 884 |
+
"step": 7700
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 20.74468085106383,
|
| 888 |
+
"eval_cer": 0.29728404490286525,
|
| 889 |
+
"eval_loss": 0.8606219291687012,
|
| 890 |
+
"eval_runtime": 151.0098,
|
| 891 |
+
"eval_samples_per_second": 32.852,
|
| 892 |
+
"eval_steps_per_second": 4.112,
|
| 893 |
+
"eval_wer": 1.0,
|
| 894 |
+
"step": 7800
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 21.01063829787234,
|
| 898 |
+
"eval_cer": 0.29262799789978866,
|
| 899 |
+
"eval_loss": 0.8141107559204102,
|
| 900 |
+
"eval_runtime": 155.5633,
|
| 901 |
+
"eval_samples_per_second": 31.891,
|
| 902 |
+
"eval_steps_per_second": 3.992,
|
| 903 |
+
"eval_wer": 0.9997984277363435,
|
| 904 |
+
"step": 7900
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 21.27659574468085,
|
| 908 |
+
"grad_norm": 5.225797176361084,
|
| 909 |
+
"learning_rate": 0.000191856,
|
| 910 |
+
"loss": 0.4252,
|
| 911 |
+
"step": 8000
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 21.27659574468085,
|
| 915 |
+
"eval_cer": 0.29841225759699375,
|
| 916 |
+
"eval_loss": 0.8832473158836365,
|
| 917 |
+
"eval_runtime": 161.5174,
|
| 918 |
+
"eval_samples_per_second": 30.715,
|
| 919 |
+
"eval_steps_per_second": 3.845,
|
| 920 |
+
"eval_wer": 0.9997984277363435,
|
| 921 |
+
"step": 8000
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 21.54255319148936,
|
| 925 |
+
"eval_cer": 0.29451124524306477,
|
| 926 |
+
"eval_loss": 0.8589781522750854,
|
| 927 |
+
"eval_runtime": 150.7991,
|
| 928 |
+
"eval_samples_per_second": 32.898,
|
| 929 |
+
"eval_steps_per_second": 4.118,
|
| 930 |
+
"eval_wer": 0.9997984277363435,
|
| 931 |
+
"step": 8100
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 21.80851063829787,
|
| 935 |
+
"eval_cer": 0.2939644960143717,
|
| 936 |
+
"eval_loss": 0.8304488062858582,
|
| 937 |
+
"eval_runtime": 156.9786,
|
| 938 |
+
"eval_samples_per_second": 31.603,
|
| 939 |
+
"eval_steps_per_second": 3.956,
|
| 940 |
+
"eval_wer": 0.9997984277363435,
|
| 941 |
+
"step": 8200
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 22.074468085106382,
|
| 945 |
+
"eval_cer": 0.29739686617227806,
|
| 946 |
+
"eval_loss": 0.873356819152832,
|
| 947 |
+
"eval_runtime": 151.6885,
|
| 948 |
+
"eval_samples_per_second": 32.705,
|
| 949 |
+
"eval_steps_per_second": 4.094,
|
| 950 |
+
"eval_wer": 0.9997984277363435,
|
| 951 |
+
"step": 8300
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 22.340425531914892,
|
| 955 |
+
"eval_cer": 0.292992497385584,
|
| 956 |
+
"eval_loss": 0.8416581153869629,
|
| 957 |
+
"eval_runtime": 171.0572,
|
| 958 |
+
"eval_samples_per_second": 29.002,
|
| 959 |
+
"eval_steps_per_second": 3.63,
|
| 960 |
+
"eval_wer": 0.9997984277363435,
|
| 961 |
+
"step": 8400
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 22.606382978723403,
|
| 965 |
+
"grad_norm": 6.305420875549316,
|
| 966 |
+
"learning_rate": 0.00020385599999999998,
|
| 967 |
+
"loss": 0.418,
|
| 968 |
+
"step": 8500
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 22.606382978723403,
|
| 972 |
+
"eval_cer": 0.2993235063114822,
|
| 973 |
+
"eval_loss": 0.9387206435203552,
|
| 974 |
+
"eval_runtime": 154.7978,
|
| 975 |
+
"eval_samples_per_second": 32.048,
|
| 976 |
+
"eval_steps_per_second": 4.012,
|
| 977 |
+
"eval_wer": 1.0,
|
| 978 |
+
"step": 8500
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 22.872340425531917,
|
| 982 |
+
"eval_cer": 0.2996055594850143,
|
| 983 |
+
"eval_loss": 0.8810231685638428,
|
| 984 |
+
"eval_runtime": 154.5989,
|
| 985 |
+
"eval_samples_per_second": 32.089,
|
| 986 |
+
"eval_steps_per_second": 4.017,
|
| 987 |
+
"eval_wer": 1.0,
|
| 988 |
+
"step": 8600
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 23.138297872340427,
|
| 992 |
+
"eval_cer": 0.3074119234724651,
|
| 993 |
+
"eval_loss": 0.9089604020118713,
|
| 994 |
+
"eval_runtime": 151.1053,
|
| 995 |
+
"eval_samples_per_second": 32.831,
|
| 996 |
+
"eval_steps_per_second": 4.11,
|
| 997 |
+
"eval_wer": 1.0,
|
| 998 |
+
"step": 8700
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 23.404255319148938,
|
| 1002 |
+
"eval_cer": 0.31074015092014423,
|
| 1003 |
+
"eval_loss": 0.8992817997932434,
|
| 1004 |
+
"eval_runtime": 154.115,
|
| 1005 |
+
"eval_samples_per_second": 32.19,
|
| 1006 |
+
"eval_steps_per_second": 4.029,
|
| 1007 |
+
"eval_wer": 0.9997984277363435,
|
| 1008 |
+
"step": 8800
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 23.670212765957448,
|
| 1012 |
+
"eval_cer": 0.3032852685797104,
|
| 1013 |
+
"eval_loss": 0.8724159598350525,
|
| 1014 |
+
"eval_runtime": 152.578,
|
| 1015 |
+
"eval_samples_per_second": 32.515,
|
| 1016 |
+
"eval_steps_per_second": 4.07,
|
| 1017 |
+
"eval_wer": 1.0,
|
| 1018 |
+
"step": 8900
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 23.93617021276596,
|
| 1022 |
+
"grad_norm": 3.3926920890808105,
|
| 1023 |
+
"learning_rate": 0.000215856,
|
| 1024 |
+
"loss": 0.424,
|
| 1025 |
+
"step": 9000
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 23.93617021276596,
|
| 1029 |
+
"eval_cer": 0.30421387441257003,
|
| 1030 |
+
"eval_loss": 0.8894767761230469,
|
| 1031 |
+
"eval_runtime": 168.9525,
|
| 1032 |
+
"eval_samples_per_second": 29.363,
|
| 1033 |
+
"eval_steps_per_second": 3.676,
|
| 1034 |
+
"eval_wer": 0.9997984277363435,
|
| 1035 |
+
"step": 9000
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 24.20212765957447,
|
| 1039 |
+
"eval_cer": 0.3013890033976559,
|
| 1040 |
+
"eval_loss": 0.8863051533699036,
|
| 1041 |
+
"eval_runtime": 151.1238,
|
| 1042 |
+
"eval_samples_per_second": 32.827,
|
| 1043 |
+
"eval_steps_per_second": 4.109,
|
| 1044 |
+
"eval_wer": 1.0,
|
| 1045 |
+
"step": 9100
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 24.46808510638298,
|
| 1049 |
+
"eval_cer": 0.3112391680733165,
|
| 1050 |
+
"eval_loss": 0.9255392551422119,
|
| 1051 |
+
"eval_runtime": 151.1678,
|
| 1052 |
+
"eval_samples_per_second": 32.818,
|
| 1053 |
+
"eval_steps_per_second": 4.108,
|
| 1054 |
+
"eval_wer": 0.9997984277363435,
|
| 1055 |
+
"step": 9200
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 24.73404255319149,
|
| 1059 |
+
"eval_cer": 0.3010635574281958,
|
| 1060 |
+
"eval_loss": 0.9398043155670166,
|
| 1061 |
+
"eval_runtime": 151.9435,
|
| 1062 |
+
"eval_samples_per_second": 32.65,
|
| 1063 |
+
"eval_steps_per_second": 4.087,
|
| 1064 |
+
"eval_wer": 0.9997984277363435,
|
| 1065 |
+
"step": 9300
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 25.0,
|
| 1069 |
+
"eval_cer": 0.30710383462137614,
|
| 1070 |
+
"eval_loss": 0.8763216137886047,
|
| 1071 |
+
"eval_runtime": 157.5659,
|
| 1072 |
+
"eval_samples_per_second": 31.485,
|
| 1073 |
+
"eval_steps_per_second": 3.941,
|
| 1074 |
+
"eval_wer": 0.9997984277363435,
|
| 1075 |
+
"step": 9400
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 25.26595744680851,
|
| 1079 |
+
"grad_norm": 9.72305679321289,
|
| 1080 |
+
"learning_rate": 0.00022785599999999997,
|
| 1081 |
+
"loss": 0.4122,
|
| 1082 |
+
"step": 9500
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 25.26595744680851,
|
| 1086 |
+
"eval_cer": 0.3090174569218018,
|
| 1087 |
+
"eval_loss": 0.9353150129318237,
|
| 1088 |
+
"eval_runtime": 152.5497,
|
| 1089 |
+
"eval_samples_per_second": 32.521,
|
| 1090 |
+
"eval_steps_per_second": 4.071,
|
| 1091 |
+
"eval_wer": 1.0,
|
| 1092 |
+
"step": 9500
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 25.53191489361702,
|
| 1096 |
+
"eval_cer": 0.31283602296346763,
|
| 1097 |
+
"eval_loss": 0.9381543397903442,
|
| 1098 |
+
"eval_runtime": 171.9179,
|
| 1099 |
+
"eval_samples_per_second": 28.857,
|
| 1100 |
+
"eval_steps_per_second": 3.612,
|
| 1101 |
+
"eval_wer": 1.0,
|
| 1102 |
+
"step": 9600
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 25.79787234042553,
|
| 1106 |
+
"eval_cer": 0.3101673660138944,
|
| 1107 |
+
"eval_loss": 0.9294571876525879,
|
| 1108 |
+
"eval_runtime": 151.6573,
|
| 1109 |
+
"eval_samples_per_second": 32.712,
|
| 1110 |
+
"eval_steps_per_second": 4.095,
|
| 1111 |
+
"eval_wer": 0.9997984277363435,
|
| 1112 |
+
"step": 9700
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 26.06382978723404,
|
| 1116 |
+
"eval_cer": 0.3091953673851067,
|
| 1117 |
+
"eval_loss": 0.9286412596702576,
|
| 1118 |
+
"eval_runtime": 151.2277,
|
| 1119 |
+
"eval_samples_per_second": 32.805,
|
| 1120 |
+
"eval_steps_per_second": 4.106,
|
| 1121 |
+
"eval_wer": 0.9997984277363435,
|
| 1122 |
+
"step": 9800
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 26.329787234042552,
|
| 1126 |
+
"eval_cer": 0.3013759855588775,
|
| 1127 |
+
"eval_loss": 0.9140869379043579,
|
| 1128 |
+
"eval_runtime": 151.0836,
|
| 1129 |
+
"eval_samples_per_second": 32.836,
|
| 1130 |
+
"eval_steps_per_second": 4.11,
|
| 1131 |
+
"eval_wer": 1.0,
|
| 1132 |
+
"step": 9900
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 26.595744680851062,
|
| 1136 |
+
"grad_norm": 4.485568046569824,
|
| 1137 |
+
"learning_rate": 0.000239856,
|
| 1138 |
+
"loss": 0.4146,
|
| 1139 |
+
"step": 10000
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 26.595744680851062,
|
| 1143 |
+
"eval_cer": 0.3125756661878995,
|
| 1144 |
+
"eval_loss": 0.942582905292511,
|
| 1145 |
+
"eval_runtime": 151.6657,
|
| 1146 |
+
"eval_samples_per_second": 32.71,
|
| 1147 |
+
"eval_steps_per_second": 4.095,
|
| 1148 |
+
"eval_wer": 1.0,
|
| 1149 |
+
"step": 10000
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 26.861702127659573,
|
| 1153 |
+
"eval_cer": 0.3032115008266328,
|
| 1154 |
+
"eval_loss": 0.8652423620223999,
|
| 1155 |
+
"eval_runtime": 150.762,
|
| 1156 |
+
"eval_samples_per_second": 32.906,
|
| 1157 |
+
"eval_steps_per_second": 4.119,
|
| 1158 |
+
"eval_wer": 1.0,
|
| 1159 |
+
"step": 10100
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 27.127659574468087,
|
| 1163 |
+
"eval_cer": 0.31053186549968975,
|
| 1164 |
+
"eval_loss": 0.9288522601127625,
|
| 1165 |
+
"eval_runtime": 151.4303,
|
| 1166 |
+
"eval_samples_per_second": 32.761,
|
| 1167 |
+
"eval_steps_per_second": 4.101,
|
| 1168 |
+
"eval_wer": 1.0,
|
| 1169 |
+
"step": 10200
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 27.393617021276597,
|
| 1173 |
+
"eval_cer": 0.31025849088534324,
|
| 1174 |
+
"eval_loss": 0.9458552002906799,
|
| 1175 |
+
"eval_runtime": 151.6351,
|
| 1176 |
+
"eval_samples_per_second": 32.717,
|
| 1177 |
+
"eval_steps_per_second": 4.095,
|
| 1178 |
+
"eval_wer": 1.0,
|
| 1179 |
+
"step": 10300
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 27.659574468085108,
|
| 1183 |
+
"eval_cer": 0.3123847378858162,
|
| 1184 |
+
"eval_loss": 0.9137027859687805,
|
| 1185 |
+
"eval_runtime": 151.0095,
|
| 1186 |
+
"eval_samples_per_second": 32.852,
|
| 1187 |
+
"eval_steps_per_second": 4.112,
|
| 1188 |
+
"eval_wer": 0.9997984277363435,
|
| 1189 |
+
"step": 10400
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 27.925531914893618,
|
| 1193 |
+
"grad_norm": 9.41711711883545,
|
| 1194 |
+
"learning_rate": 0.000251856,
|
| 1195 |
+
"loss": 0.416,
|
| 1196 |
+
"step": 10500
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 27.925531914893618,
|
| 1200 |
+
"eval_cer": 0.3099504020342543,
|
| 1201 |
+
"eval_loss": 0.9305452704429626,
|
| 1202 |
+
"eval_runtime": 151.4404,
|
| 1203 |
+
"eval_samples_per_second": 32.759,
|
| 1204 |
+
"eval_steps_per_second": 4.101,
|
| 1205 |
+
"eval_wer": 1.0,
|
| 1206 |
+
"step": 10500
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 28.19148936170213,
|
| 1210 |
+
"eval_cer": 0.30706912038463374,
|
| 1211 |
+
"eval_loss": 0.9589177370071411,
|
| 1212 |
+
"eval_runtime": 152.6851,
|
| 1213 |
+
"eval_samples_per_second": 32.492,
|
| 1214 |
+
"eval_steps_per_second": 4.067,
|
| 1215 |
+
"eval_wer": 1.0,
|
| 1216 |
+
"step": 10600
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 28.45744680851064,
|
| 1220 |
+
"eval_cer": 0.30605372895991806,
|
| 1221 |
+
"eval_loss": 0.92755526304245,
|
| 1222 |
+
"eval_runtime": 151.1956,
|
| 1223 |
+
"eval_samples_per_second": 32.812,
|
| 1224 |
+
"eval_steps_per_second": 4.107,
|
| 1225 |
+
"eval_wer": 0.9997984277363435,
|
| 1226 |
+
"step": 10700
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 28.72340425531915,
|
| 1230 |
+
"eval_cer": 0.3078067979154101,
|
| 1231 |
+
"eval_loss": 0.9259202480316162,
|
| 1232 |
+
"eval_runtime": 151.1107,
|
| 1233 |
+
"eval_samples_per_second": 32.83,
|
| 1234 |
+
"eval_steps_per_second": 4.11,
|
| 1235 |
+
"eval_wer": 1.0,
|
| 1236 |
+
"step": 10800
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 28.98936170212766,
|
| 1240 |
+
"eval_cer": 0.3149796270823118,
|
| 1241 |
+
"eval_loss": 0.9287381768226624,
|
| 1242 |
+
"eval_runtime": 151.0264,
|
| 1243 |
+
"eval_samples_per_second": 32.849,
|
| 1244 |
+
"eval_steps_per_second": 4.112,
|
| 1245 |
+
"eval_wer": 1.0,
|
| 1246 |
+
"step": 10900
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 29.25531914893617,
|
| 1250 |
+
"grad_norm": 6.876596927642822,
|
| 1251 |
+
"learning_rate": 0.000263832,
|
| 1252 |
+
"loss": 0.4078,
|
| 1253 |
+
"step": 11000
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 29.25531914893617,
|
| 1257 |
+
"eval_cer": 0.3165591248540917,
|
| 1258 |
+
"eval_loss": 0.9346364736557007,
|
| 1259 |
+
"eval_runtime": 151.4411,
|
| 1260 |
+
"eval_samples_per_second": 32.759,
|
| 1261 |
+
"eval_steps_per_second": 4.101,
|
| 1262 |
+
"eval_wer": 0.9997984277363435,
|
| 1263 |
+
"step": 11000
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 29.52127659574468,
|
| 1267 |
+
"eval_cer": 0.31648969638060687,
|
| 1268 |
+
"eval_loss": 0.9558972120285034,
|
| 1269 |
+
"eval_runtime": 151.7115,
|
| 1270 |
+
"eval_samples_per_second": 32.7,
|
| 1271 |
+
"eval_steps_per_second": 4.093,
|
| 1272 |
+
"eval_wer": 1.0,
|
| 1273 |
+
"step": 11100
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 29.78723404255319,
|
| 1277 |
+
"eval_cer": 0.31732717734201765,
|
| 1278 |
+
"eval_loss": 0.9669720530509949,
|
| 1279 |
+
"eval_runtime": 151.1358,
|
| 1280 |
+
"eval_samples_per_second": 32.825,
|
| 1281 |
+
"eval_steps_per_second": 4.109,
|
| 1282 |
+
"eval_wer": 1.0,
|
| 1283 |
+
"step": 11200
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 30.0,
|
| 1287 |
+
"step": 11280,
|
| 1288 |
+
"total_flos": 1.1567283350118728e+19,
|
| 1289 |
+
"train_loss": 1.009270377192937,
|
| 1290 |
+
"train_runtime": 42299.7955,
|
| 1291 |
+
"train_samples_per_second": 8.533,
|
| 1292 |
+
"train_steps_per_second": 0.267
|
| 1293 |
+
}
|
| 1294 |
+
],
|
| 1295 |
+
"logging_steps": 500,
|
| 1296 |
+
"max_steps": 11280,
|
| 1297 |
+
"num_input_tokens_seen": 0,
|
| 1298 |
+
"num_train_epochs": 30,
|
| 1299 |
+
"save_steps": 400,
|
| 1300 |
+
"stateful_callbacks": {
|
| 1301 |
+
"TrainerControl": {
|
| 1302 |
+
"args": {
|
| 1303 |
+
"should_epoch_stop": false,
|
| 1304 |
+
"should_evaluate": false,
|
| 1305 |
+
"should_log": false,
|
| 1306 |
+
"should_save": true,
|
| 1307 |
+
"should_training_stop": true
|
| 1308 |
+
},
|
| 1309 |
+
"attributes": {}
|
| 1310 |
+
}
|
| 1311 |
+
},
|
| 1312 |
+
"total_flos": 1.1567283350118728e+19,
|
| 1313 |
+
"train_batch_size": 16,
|
| 1314 |
+
"trial_name": null,
|
| 1315 |
+
"trial_params": null
|
| 1316 |
+
}
|