End of training
Browse files- README.md +8 -6
- all_results.json +12 -12
- eval_results.json +7 -7
- train_results.json +6 -6
- trainer_state.json +1255 -22
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/wav2vec2-large-xlsr-53
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- audiofolder
|
|
@@ -14,15 +16,15 @@ model-index:
|
|
| 14 |
name: Automatic Speech Recognition
|
| 15 |
type: automatic-speech-recognition
|
| 16 |
dataset:
|
| 17 |
-
name:
|
| 18 |
type: audiofolder
|
| 19 |
config: default
|
| 20 |
split: validation
|
| 21 |
-
args:
|
| 22 |
metrics:
|
| 23 |
- name: Wer
|
| 24 |
type: wer
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -30,10 +32,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# em_ctc
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 2.
|
| 36 |
-
- Wer: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/wav2vec2-large-xlsr-53
|
| 4 |
tags:
|
| 5 |
+
- automatic-speech-recognition
|
| 6 |
+
- wav_sub-P001
|
| 7 |
- generated_from_trainer
|
| 8 |
datasets:
|
| 9 |
- audiofolder
|
|
|
|
| 16 |
name: Automatic Speech Recognition
|
| 17 |
type: automatic-speech-recognition
|
| 18 |
dataset:
|
| 19 |
+
name: WAV_SUB-P001 - TR
|
| 20 |
type: audiofolder
|
| 21 |
config: default
|
| 22 |
split: validation
|
| 23 |
+
args: 'Config: tr, Training split: train+validation, Eval split: test'
|
| 24 |
metrics:
|
| 25 |
- name: Wer
|
| 26 |
type: wer
|
| 27 |
+
value: 0.9843253066787824
|
| 28 |
---
|
| 29 |
|
| 30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
# em_ctc
|
| 34 |
|
| 35 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the WAV_SUB-P001 - TR dataset.
|
| 36 |
It achieves the following results on the evaluation set:
|
| 37 |
+
- Loss: 2.7286
|
| 38 |
+
- Wer: 0.9843
|
| 39 |
|
| 40 |
## Model description
|
| 41 |
|
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 2.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"eval_wer":
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples":
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 19.98,
|
| 3 |
+
"eval_loss": 2.728630542755127,
|
| 4 |
+
"eval_runtime": 35.1738,
|
| 5 |
+
"eval_samples": 2201,
|
| 6 |
+
"eval_samples_per_second": 62.575,
|
| 7 |
+
"eval_steps_per_second": 7.847,
|
| 8 |
+
"eval_wer": 0.9843253066787824,
|
| 9 |
+
"train_loss": 2.556843002393288,
|
| 10 |
+
"train_runtime": 17303.3516,
|
| 11 |
+
"train_samples": 19819,
|
| 12 |
+
"train_samples_per_second": 22.908,
|
| 13 |
+
"train_steps_per_second": 0.715
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 2.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"eval_wer":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 19.98,
|
| 3 |
+
"eval_loss": 2.728630542755127,
|
| 4 |
+
"eval_runtime": 35.1738,
|
| 5 |
+
"eval_samples": 2201,
|
| 6 |
+
"eval_samples_per_second": 62.575,
|
| 7 |
+
"eval_steps_per_second": 7.847,
|
| 8 |
+
"eval_wer": 0.9843253066787824
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 19.98,
|
| 3 |
+
"train_loss": 2.556843002393288,
|
| 4 |
+
"train_runtime": 17303.3516,
|
| 5 |
+
"train_samples": 19819,
|
| 6 |
+
"train_samples_per_second": 22.908,
|
| 7 |
+
"train_steps_per_second": 0.715
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,46 +1,1279 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 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.
|
| 13 |
-
"eval_loss": 3.
|
| 14 |
-
"eval_runtime":
|
| 15 |
-
"eval_samples_per_second":
|
| 16 |
-
"eval_steps_per_second":
|
| 17 |
"eval_wer": 1.0,
|
| 18 |
"step": 100
|
| 19 |
},
|
| 20 |
{
|
| 21 |
-
"epoch": 0.
|
| 22 |
-
"eval_loss": 2.
|
| 23 |
-
"eval_runtime":
|
| 24 |
-
"eval_samples_per_second":
|
| 25 |
-
"eval_steps_per_second":
|
| 26 |
"eval_wer": 1.0,
|
| 27 |
"step": 200
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch":
|
| 31 |
-
"
|
| 32 |
-
"
|
| 33 |
-
"
|
| 34 |
-
"
|
| 35 |
-
"
|
| 36 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
}
|
| 38 |
],
|
| 39 |
"logging_steps": 500,
|
| 40 |
-
"max_steps":
|
| 41 |
-
"num_train_epochs":
|
| 42 |
"save_steps": 400,
|
| 43 |
-
"total_flos": 1.
|
| 44 |
"trial_name": null,
|
| 45 |
"trial_params": null
|
| 46 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.983857949959646,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 12380,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.16,
|
| 13 |
+
"eval_loss": 3.898963689804077,
|
| 14 |
+
"eval_runtime": 35.3167,
|
| 15 |
+
"eval_samples_per_second": 62.322,
|
| 16 |
+
"eval_steps_per_second": 7.815,
|
| 17 |
"eval_wer": 1.0,
|
| 18 |
"step": 100
|
| 19 |
},
|
| 20 |
{
|
| 21 |
+
"epoch": 0.32,
|
| 22 |
+
"eval_loss": 2.8678269386291504,
|
| 23 |
+
"eval_runtime": 35.3815,
|
| 24 |
+
"eval_samples_per_second": 62.208,
|
| 25 |
+
"eval_steps_per_second": 7.801,
|
| 26 |
"eval_wer": 1.0,
|
| 27 |
"step": 200
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.48,
|
| 31 |
+
"eval_loss": 2.7794852256774902,
|
| 32 |
+
"eval_runtime": 35.4013,
|
| 33 |
+
"eval_samples_per_second": 62.173,
|
| 34 |
+
"eval_steps_per_second": 7.796,
|
| 35 |
+
"eval_wer": 1.0,
|
| 36 |
+
"step": 300
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.65,
|
| 40 |
+
"eval_loss": 2.7316112518310547,
|
| 41 |
+
"eval_runtime": 35.4325,
|
| 42 |
+
"eval_samples_per_second": 62.118,
|
| 43 |
+
"eval_steps_per_second": 7.789,
|
| 44 |
+
"eval_wer": 1.0,
|
| 45 |
+
"step": 400
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.81,
|
| 49 |
+
"learning_rate": 0.00029699999999999996,
|
| 50 |
+
"loss": 4.389,
|
| 51 |
+
"step": 500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.81,
|
| 55 |
+
"eval_loss": 2.688910722732544,
|
| 56 |
+
"eval_runtime": 35.4199,
|
| 57 |
+
"eval_samples_per_second": 62.14,
|
| 58 |
+
"eval_steps_per_second": 7.792,
|
| 59 |
+
"eval_wer": 0.9784189004997729,
|
| 60 |
+
"step": 500
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.97,
|
| 64 |
+
"eval_loss": 3.272650957107544,
|
| 65 |
+
"eval_runtime": 35.3276,
|
| 66 |
+
"eval_samples_per_second": 62.303,
|
| 67 |
+
"eval_steps_per_second": 7.813,
|
| 68 |
+
"eval_wer": 0.9784189004997729,
|
| 69 |
+
"step": 600
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.13,
|
| 73 |
+
"eval_loss": 2.7057435512542725,
|
| 74 |
+
"eval_runtime": 35.3617,
|
| 75 |
+
"eval_samples_per_second": 62.243,
|
| 76 |
+
"eval_steps_per_second": 7.805,
|
| 77 |
+
"eval_wer": 0.9784189004997729,
|
| 78 |
+
"step": 700
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.29,
|
| 82 |
+
"eval_loss": 2.852546215057373,
|
| 83 |
+
"eval_runtime": 35.394,
|
| 84 |
+
"eval_samples_per_second": 62.186,
|
| 85 |
+
"eval_steps_per_second": 7.798,
|
| 86 |
+
"eval_wer": 0.9963652885052249,
|
| 87 |
+
"step": 800
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.45,
|
| 91 |
+
"eval_loss": 2.668452024459839,
|
| 92 |
+
"eval_runtime": 35.1413,
|
| 93 |
+
"eval_samples_per_second": 62.633,
|
| 94 |
+
"eval_steps_per_second": 7.854,
|
| 95 |
+
"eval_wer": 0.9968196274420718,
|
| 96 |
+
"step": 900
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.61,
|
| 100 |
+
"learning_rate": 0.0002875,
|
| 101 |
+
"loss": 2.5649,
|
| 102 |
+
"step": 1000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.61,
|
| 106 |
+
"eval_loss": 2.740328788757324,
|
| 107 |
+
"eval_runtime": 35.5622,
|
| 108 |
+
"eval_samples_per_second": 61.892,
|
| 109 |
+
"eval_steps_per_second": 7.761,
|
| 110 |
+
"eval_wer": 1.0,
|
| 111 |
+
"step": 1000
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.78,
|
| 115 |
+
"eval_loss": 2.7789676189422607,
|
| 116 |
+
"eval_runtime": 35.3028,
|
| 117 |
+
"eval_samples_per_second": 62.346,
|
| 118 |
+
"eval_steps_per_second": 7.818,
|
| 119 |
+
"eval_wer": 1.0,
|
| 120 |
+
"step": 1100
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.94,
|
| 124 |
+
"eval_loss": 2.812997579574585,
|
| 125 |
+
"eval_runtime": 35.2664,
|
| 126 |
+
"eval_samples_per_second": 62.411,
|
| 127 |
+
"eval_steps_per_second": 7.826,
|
| 128 |
+
"eval_wer": 0.9786460699681963,
|
| 129 |
+
"step": 1200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.1,
|
| 133 |
+
"eval_loss": 2.803067207336426,
|
| 134 |
+
"eval_runtime": 35.2577,
|
| 135 |
+
"eval_samples_per_second": 62.426,
|
| 136 |
+
"eval_steps_per_second": 7.828,
|
| 137 |
+
"eval_wer": 1.0,
|
| 138 |
+
"step": 1300
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.26,
|
| 142 |
+
"eval_loss": 2.9682765007019043,
|
| 143 |
+
"eval_runtime": 35.4719,
|
| 144 |
+
"eval_samples_per_second": 62.049,
|
| 145 |
+
"eval_steps_per_second": 7.781,
|
| 146 |
+
"eval_wer": 1.0,
|
| 147 |
+
"step": 1400
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.42,
|
| 151 |
+
"learning_rate": 0.00027489898989898986,
|
| 152 |
+
"loss": 2.5421,
|
| 153 |
+
"step": 1500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.42,
|
| 157 |
+
"eval_loss": 2.9459009170532227,
|
| 158 |
+
"eval_runtime": 35.5607,
|
| 159 |
+
"eval_samples_per_second": 61.894,
|
| 160 |
+
"eval_steps_per_second": 7.761,
|
| 161 |
+
"eval_wer": 0.9784189004997729,
|
| 162 |
+
"step": 1500
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 2.58,
|
| 166 |
+
"eval_loss": 2.7052440643310547,
|
| 167 |
+
"eval_runtime": 35.4157,
|
| 168 |
+
"eval_samples_per_second": 62.148,
|
| 169 |
+
"eval_steps_per_second": 7.793,
|
| 170 |
+
"eval_wer": 0.9784189004997729,
|
| 171 |
+
"step": 1600
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.74,
|
| 175 |
+
"eval_loss": 2.7878549098968506,
|
| 176 |
+
"eval_runtime": 35.5231,
|
| 177 |
+
"eval_samples_per_second": 61.96,
|
| 178 |
+
"eval_steps_per_second": 7.77,
|
| 179 |
+
"eval_wer": 0.9786460699681963,
|
| 180 |
+
"step": 1700
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 2.91,
|
| 184 |
+
"eval_loss": 2.7955620288848877,
|
| 185 |
+
"eval_runtime": 35.85,
|
| 186 |
+
"eval_samples_per_second": 61.395,
|
| 187 |
+
"eval_steps_per_second": 7.699,
|
| 188 |
+
"eval_wer": 1.0,
|
| 189 |
+
"step": 1800
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 3.07,
|
| 193 |
+
"eval_loss": 2.776027202606201,
|
| 194 |
+
"eval_runtime": 35.7619,
|
| 195 |
+
"eval_samples_per_second": 61.546,
|
| 196 |
+
"eval_steps_per_second": 7.718,
|
| 197 |
+
"eval_wer": 0.9784189004997729,
|
| 198 |
+
"step": 1900
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 3.23,
|
| 202 |
+
"learning_rate": 0.00026227272727272726,
|
| 203 |
+
"loss": 2.5357,
|
| 204 |
+
"step": 2000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.23,
|
| 208 |
+
"eval_loss": 2.859441041946411,
|
| 209 |
+
"eval_runtime": 35.6034,
|
| 210 |
+
"eval_samples_per_second": 61.82,
|
| 211 |
+
"eval_steps_per_second": 7.752,
|
| 212 |
+
"eval_wer": 0.9995456610631531,
|
| 213 |
+
"step": 2000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 3.39,
|
| 217 |
+
"eval_loss": 2.9047751426696777,
|
| 218 |
+
"eval_runtime": 35.41,
|
| 219 |
+
"eval_samples_per_second": 62.158,
|
| 220 |
+
"eval_steps_per_second": 7.794,
|
| 221 |
+
"eval_wer": 0.97955474784189,
|
| 222 |
+
"step": 2100
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.55,
|
| 226 |
+
"eval_loss": 3.0097830295562744,
|
| 227 |
+
"eval_runtime": 36.2831,
|
| 228 |
+
"eval_samples_per_second": 60.662,
|
| 229 |
+
"eval_steps_per_second": 7.607,
|
| 230 |
+
"eval_wer": 0.9784189004997729,
|
| 231 |
+
"step": 2200
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 3.71,
|
| 235 |
+
"eval_loss": 2.7079336643218994,
|
| 236 |
+
"eval_runtime": 35.6106,
|
| 237 |
+
"eval_samples_per_second": 61.807,
|
| 238 |
+
"eval_steps_per_second": 7.75,
|
| 239 |
+
"eval_wer": 0.9784189004997729,
|
| 240 |
+
"step": 2300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.87,
|
| 244 |
+
"eval_loss": 3.2402844429016113,
|
| 245 |
+
"eval_runtime": 35.4272,
|
| 246 |
+
"eval_samples_per_second": 62.127,
|
| 247 |
+
"eval_steps_per_second": 7.791,
|
| 248 |
+
"eval_wer": 1.0,
|
| 249 |
+
"step": 2400
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 4.04,
|
| 253 |
+
"learning_rate": 0.00024967171717171717,
|
| 254 |
+
"loss": 2.5203,
|
| 255 |
+
"step": 2500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.04,
|
| 259 |
+
"eval_loss": 3.0476486682891846,
|
| 260 |
+
"eval_runtime": 35.3746,
|
| 261 |
+
"eval_samples_per_second": 62.22,
|
| 262 |
+
"eval_steps_per_second": 7.802,
|
| 263 |
+
"eval_wer": 0.9784189004997729,
|
| 264 |
+
"step": 2500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 4.2,
|
| 268 |
+
"eval_loss": 2.850980520248413,
|
| 269 |
+
"eval_runtime": 35.1192,
|
| 270 |
+
"eval_samples_per_second": 62.672,
|
| 271 |
+
"eval_steps_per_second": 7.859,
|
| 272 |
+
"eval_wer": 1.0,
|
| 273 |
+
"step": 2600
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 4.36,
|
| 277 |
+
"eval_loss": 2.790703296661377,
|
| 278 |
+
"eval_runtime": 34.998,
|
| 279 |
+
"eval_samples_per_second": 62.889,
|
| 280 |
+
"eval_steps_per_second": 7.886,
|
| 281 |
+
"eval_wer": 0.9784189004997729,
|
| 282 |
+
"step": 2700
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 4.52,
|
| 286 |
+
"eval_loss": 2.7485764026641846,
|
| 287 |
+
"eval_runtime": 35.3947,
|
| 288 |
+
"eval_samples_per_second": 62.184,
|
| 289 |
+
"eval_steps_per_second": 7.798,
|
| 290 |
+
"eval_wer": 0.9784189004997729,
|
| 291 |
+
"step": 2800
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 4.68,
|
| 295 |
+
"eval_loss": 3.1700756549835205,
|
| 296 |
+
"eval_runtime": 35.059,
|
| 297 |
+
"eval_samples_per_second": 62.78,
|
| 298 |
+
"eval_steps_per_second": 7.872,
|
| 299 |
+
"eval_wer": 1.0,
|
| 300 |
+
"step": 2900
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.84,
|
| 304 |
+
"learning_rate": 0.00023704545454545452,
|
| 305 |
+
"loss": 2.5191,
|
| 306 |
+
"step": 3000
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 4.84,
|
| 310 |
+
"eval_loss": 2.9528985023498535,
|
| 311 |
+
"eval_runtime": 34.7932,
|
| 312 |
+
"eval_samples_per_second": 63.259,
|
| 313 |
+
"eval_steps_per_second": 7.933,
|
| 314 |
+
"eval_wer": 0.9784189004997729,
|
| 315 |
+
"step": 3000
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 5.0,
|
| 319 |
+
"eval_loss": 3.1192333698272705,
|
| 320 |
+
"eval_runtime": 35.2754,
|
| 321 |
+
"eval_samples_per_second": 62.395,
|
| 322 |
+
"eval_steps_per_second": 7.824,
|
| 323 |
+
"eval_wer": 0.9650159018627896,
|
| 324 |
+
"step": 3100
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 5.17,
|
| 328 |
+
"eval_loss": 2.8595774173736572,
|
| 329 |
+
"eval_runtime": 34.9676,
|
| 330 |
+
"eval_samples_per_second": 62.944,
|
| 331 |
+
"eval_steps_per_second": 7.893,
|
| 332 |
+
"eval_wer": 1.0,
|
| 333 |
+
"step": 3200
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 5.33,
|
| 337 |
+
"eval_loss": 2.919292449951172,
|
| 338 |
+
"eval_runtime": 34.9757,
|
| 339 |
+
"eval_samples_per_second": 62.929,
|
| 340 |
+
"eval_steps_per_second": 7.891,
|
| 341 |
+
"eval_wer": 1.0,
|
| 342 |
+
"step": 3300
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 5.49,
|
| 346 |
+
"eval_loss": 3.0366740226745605,
|
| 347 |
+
"eval_runtime": 35.2556,
|
| 348 |
+
"eval_samples_per_second": 62.43,
|
| 349 |
+
"eval_steps_per_second": 7.829,
|
| 350 |
+
"eval_wer": 0.9784189004997729,
|
| 351 |
+
"step": 3400
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 5.65,
|
| 355 |
+
"learning_rate": 0.00022441919191919192,
|
| 356 |
+
"loss": 2.5422,
|
| 357 |
+
"step": 3500
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 5.65,
|
| 361 |
+
"eval_loss": 2.9161534309387207,
|
| 362 |
+
"eval_runtime": 35.3286,
|
| 363 |
+
"eval_samples_per_second": 62.301,
|
| 364 |
+
"eval_steps_per_second": 7.812,
|
| 365 |
+
"eval_wer": 0.9784189004997729,
|
| 366 |
+
"step": 3500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 5.81,
|
| 370 |
+
"eval_loss": 3.033392906188965,
|
| 371 |
+
"eval_runtime": 35.2038,
|
| 372 |
+
"eval_samples_per_second": 62.522,
|
| 373 |
+
"eval_steps_per_second": 7.84,
|
| 374 |
+
"eval_wer": 1.0,
|
| 375 |
+
"step": 3600
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 5.97,
|
| 379 |
+
"eval_loss": 2.8513946533203125,
|
| 380 |
+
"eval_runtime": 35.344,
|
| 381 |
+
"eval_samples_per_second": 62.274,
|
| 382 |
+
"eval_steps_per_second": 7.809,
|
| 383 |
+
"eval_wer": 0.9784189004997729,
|
| 384 |
+
"step": 3700
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 6.13,
|
| 388 |
+
"eval_loss": 2.9653778076171875,
|
| 389 |
+
"eval_runtime": 35.1238,
|
| 390 |
+
"eval_samples_per_second": 62.664,
|
| 391 |
+
"eval_steps_per_second": 7.858,
|
| 392 |
+
"eval_wer": 1.0,
|
| 393 |
+
"step": 3800
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.3,
|
| 397 |
+
"eval_loss": 3.261630058288574,
|
| 398 |
+
"eval_runtime": 35.0149,
|
| 399 |
+
"eval_samples_per_second": 62.859,
|
| 400 |
+
"eval_steps_per_second": 7.882,
|
| 401 |
+
"eval_wer": 0.9784189004997729,
|
| 402 |
+
"step": 3900
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 6.46,
|
| 406 |
+
"learning_rate": 0.00021179292929292926,
|
| 407 |
+
"loss": 2.5062,
|
| 408 |
+
"step": 4000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 6.46,
|
| 412 |
+
"eval_loss": 3.33202862739563,
|
| 413 |
+
"eval_runtime": 34.9678,
|
| 414 |
+
"eval_samples_per_second": 62.944,
|
| 415 |
+
"eval_steps_per_second": 7.893,
|
| 416 |
+
"eval_wer": 0.9793275783734666,
|
| 417 |
+
"step": 4000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 6.62,
|
| 421 |
+
"eval_loss": 2.714141845703125,
|
| 422 |
+
"eval_runtime": 35.0566,
|
| 423 |
+
"eval_samples_per_second": 62.784,
|
| 424 |
+
"eval_steps_per_second": 7.873,
|
| 425 |
+
"eval_wer": 0.9784189004997729,
|
| 426 |
+
"step": 4100
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 6.78,
|
| 430 |
+
"eval_loss": 3.2108442783355713,
|
| 431 |
+
"eval_runtime": 34.9263,
|
| 432 |
+
"eval_samples_per_second": 63.018,
|
| 433 |
+
"eval_steps_per_second": 7.902,
|
| 434 |
+
"eval_wer": 0.9784189004997729,
|
| 435 |
+
"step": 4200
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 6.94,
|
| 439 |
+
"eval_loss": 3.0015201568603516,
|
| 440 |
+
"eval_runtime": 35.4639,
|
| 441 |
+
"eval_samples_per_second": 62.063,
|
| 442 |
+
"eval_steps_per_second": 7.783,
|
| 443 |
+
"eval_wer": 0.9784189004997729,
|
| 444 |
+
"step": 4300
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 7.1,
|
| 448 |
+
"eval_loss": 3.024397134780884,
|
| 449 |
+
"eval_runtime": 35.8959,
|
| 450 |
+
"eval_samples_per_second": 61.316,
|
| 451 |
+
"eval_steps_per_second": 7.689,
|
| 452 |
+
"eval_wer": 1.0,
|
| 453 |
+
"step": 4400
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 7.26,
|
| 457 |
+
"learning_rate": 0.00019916666666666664,
|
| 458 |
+
"loss": 2.5114,
|
| 459 |
+
"step": 4500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 7.26,
|
| 463 |
+
"eval_loss": 2.874232053756714,
|
| 464 |
+
"eval_runtime": 35.3093,
|
| 465 |
+
"eval_samples_per_second": 62.335,
|
| 466 |
+
"eval_steps_per_second": 7.817,
|
| 467 |
+
"eval_wer": 0.9784189004997729,
|
| 468 |
+
"step": 4500
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 7.43,
|
| 472 |
+
"eval_loss": 3.147094488143921,
|
| 473 |
+
"eval_runtime": 35.1341,
|
| 474 |
+
"eval_samples_per_second": 62.646,
|
| 475 |
+
"eval_steps_per_second": 7.856,
|
| 476 |
+
"eval_wer": 0.9784189004997729,
|
| 477 |
+
"step": 4600
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 7.59,
|
| 481 |
+
"eval_loss": 2.7005865573883057,
|
| 482 |
+
"eval_runtime": 35.1317,
|
| 483 |
+
"eval_samples_per_second": 62.65,
|
| 484 |
+
"eval_steps_per_second": 7.856,
|
| 485 |
+
"eval_wer": 0.9772830531576556,
|
| 486 |
+
"step": 4700
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 7.75,
|
| 490 |
+
"eval_loss": 3.118854284286499,
|
| 491 |
+
"eval_runtime": 35.077,
|
| 492 |
+
"eval_samples_per_second": 62.748,
|
| 493 |
+
"eval_steps_per_second": 7.868,
|
| 494 |
+
"eval_wer": 1.0,
|
| 495 |
+
"step": 4800
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 7.91,
|
| 499 |
+
"eval_loss": 3.1604197025299072,
|
| 500 |
+
"eval_runtime": 35.1065,
|
| 501 |
+
"eval_samples_per_second": 62.695,
|
| 502 |
+
"eval_steps_per_second": 7.862,
|
| 503 |
+
"eval_wer": 0.9784189004997729,
|
| 504 |
+
"step": 4900
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 8.07,
|
| 508 |
+
"learning_rate": 0.000186540404040404,
|
| 509 |
+
"loss": 2.5065,
|
| 510 |
+
"step": 5000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 8.07,
|
| 514 |
+
"eval_loss": 2.92966890335083,
|
| 515 |
+
"eval_runtime": 35.3426,
|
| 516 |
+
"eval_samples_per_second": 62.276,
|
| 517 |
+
"eval_steps_per_second": 7.809,
|
| 518 |
+
"eval_wer": 0.9784189004997729,
|
| 519 |
+
"step": 5000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 8.23,
|
| 523 |
+
"eval_loss": 3.0998189449310303,
|
| 524 |
+
"eval_runtime": 34.9856,
|
| 525 |
+
"eval_samples_per_second": 62.912,
|
| 526 |
+
"eval_steps_per_second": 7.889,
|
| 527 |
+
"eval_wer": 0.9784189004997729,
|
| 528 |
+
"step": 5100
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 8.39,
|
| 532 |
+
"eval_loss": 2.818410873413086,
|
| 533 |
+
"eval_runtime": 35.4037,
|
| 534 |
+
"eval_samples_per_second": 62.169,
|
| 535 |
+
"eval_steps_per_second": 7.796,
|
| 536 |
+
"eval_wer": 0.9843253066787824,
|
| 537 |
+
"step": 5200
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 8.56,
|
| 541 |
+
"eval_loss": 2.7132983207702637,
|
| 542 |
+
"eval_runtime": 35.1176,
|
| 543 |
+
"eval_samples_per_second": 62.675,
|
| 544 |
+
"eval_steps_per_second": 7.859,
|
| 545 |
+
"eval_wer": 0.9861426624261699,
|
| 546 |
+
"step": 5300
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 8.72,
|
| 550 |
+
"eval_loss": 2.7398695945739746,
|
| 551 |
+
"eval_runtime": 35.1472,
|
| 552 |
+
"eval_samples_per_second": 62.622,
|
| 553 |
+
"eval_steps_per_second": 7.853,
|
| 554 |
+
"eval_wer": 0.9811449341208541,
|
| 555 |
+
"step": 5400
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 8.88,
|
| 559 |
+
"learning_rate": 0.0001739141414141414,
|
| 560 |
+
"loss": 2.4956,
|
| 561 |
+
"step": 5500
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 8.88,
|
| 565 |
+
"eval_loss": 2.7186269760131836,
|
| 566 |
+
"eval_runtime": 35.2793,
|
| 567 |
+
"eval_samples_per_second": 62.388,
|
| 568 |
+
"eval_steps_per_second": 7.823,
|
| 569 |
+
"eval_wer": 0.9888686960472512,
|
| 570 |
+
"step": 5500
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 9.04,
|
| 574 |
+
"eval_loss": 2.9871914386749268,
|
| 575 |
+
"eval_runtime": 34.9607,
|
| 576 |
+
"eval_samples_per_second": 62.956,
|
| 577 |
+
"eval_steps_per_second": 7.895,
|
| 578 |
+
"eval_wer": 0.9954566106315311,
|
| 579 |
+
"step": 5600
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 9.2,
|
| 583 |
+
"eval_loss": 3.082458257675171,
|
| 584 |
+
"eval_runtime": 34.98,
|
| 585 |
+
"eval_samples_per_second": 62.922,
|
| 586 |
+
"eval_steps_per_second": 7.89,
|
| 587 |
+
"eval_wer": 0.9993184915947296,
|
| 588 |
+
"step": 5700
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 9.36,
|
| 592 |
+
"eval_loss": 3.0588910579681396,
|
| 593 |
+
"eval_runtime": 35.5843,
|
| 594 |
+
"eval_samples_per_second": 61.853,
|
| 595 |
+
"eval_steps_per_second": 7.756,
|
| 596 |
+
"eval_wer": 0.9854611540208996,
|
| 597 |
+
"step": 5800
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 9.52,
|
| 601 |
+
"eval_loss": 2.8176536560058594,
|
| 602 |
+
"eval_runtime": 35.3586,
|
| 603 |
+
"eval_samples_per_second": 62.248,
|
| 604 |
+
"eval_steps_per_second": 7.806,
|
| 605 |
+
"eval_wer": 0.9784189004997729,
|
| 606 |
+
"step": 5900
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 9.69,
|
| 610 |
+
"learning_rate": 0.00016128787878787879,
|
| 611 |
+
"loss": 2.4774,
|
| 612 |
+
"step": 6000
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 9.69,
|
| 616 |
+
"eval_loss": 2.8103556632995605,
|
| 617 |
+
"eval_runtime": 34.9221,
|
| 618 |
+
"eval_samples_per_second": 63.026,
|
| 619 |
+
"eval_steps_per_second": 7.903,
|
| 620 |
+
"eval_wer": 0.9993184915947296,
|
| 621 |
+
"step": 6000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 9.85,
|
| 625 |
+
"eval_loss": 2.9498021602630615,
|
| 626 |
+
"eval_runtime": 35.3693,
|
| 627 |
+
"eval_samples_per_second": 62.229,
|
| 628 |
+
"eval_steps_per_second": 7.803,
|
| 629 |
+
"eval_wer": 0.97955474784189,
|
| 630 |
+
"step": 6100
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 10.01,
|
| 634 |
+
"eval_loss": 3.000636577606201,
|
| 635 |
+
"eval_runtime": 35.1186,
|
| 636 |
+
"eval_samples_per_second": 62.673,
|
| 637 |
+
"eval_steps_per_second": 7.859,
|
| 638 |
+
"eval_wer": 0.9784189004997729,
|
| 639 |
+
"step": 6200
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 10.17,
|
| 643 |
+
"eval_loss": 2.809988260269165,
|
| 644 |
+
"eval_runtime": 35.4693,
|
| 645 |
+
"eval_samples_per_second": 62.054,
|
| 646 |
+
"eval_steps_per_second": 7.781,
|
| 647 |
+
"eval_wer": 0.9784189004997729,
|
| 648 |
+
"step": 6300
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 10.33,
|
| 652 |
+
"eval_loss": 3.1577372550964355,
|
| 653 |
+
"eval_runtime": 34.9862,
|
| 654 |
+
"eval_samples_per_second": 62.91,
|
| 655 |
+
"eval_steps_per_second": 7.889,
|
| 656 |
+
"eval_wer": 0.9786460699681963,
|
| 657 |
+
"step": 6400
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 10.49,
|
| 661 |
+
"learning_rate": 0.00014866161616161616,
|
| 662 |
+
"loss": 2.4689,
|
| 663 |
+
"step": 6500
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 10.49,
|
| 667 |
+
"eval_loss": 2.7814252376556396,
|
| 668 |
+
"eval_runtime": 35.2563,
|
| 669 |
+
"eval_samples_per_second": 62.429,
|
| 670 |
+
"eval_steps_per_second": 7.828,
|
| 671 |
+
"eval_wer": 0.9977283053157655,
|
| 672 |
+
"step": 6500
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 10.65,
|
| 676 |
+
"eval_loss": 2.72709059715271,
|
| 677 |
+
"eval_runtime": 35.1394,
|
| 678 |
+
"eval_samples_per_second": 62.636,
|
| 679 |
+
"eval_steps_per_second": 7.854,
|
| 680 |
+
"eval_wer": 0.983643798273512,
|
| 681 |
+
"step": 6600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 10.82,
|
| 685 |
+
"eval_loss": 2.840341567993164,
|
| 686 |
+
"eval_runtime": 35.568,
|
| 687 |
+
"eval_samples_per_second": 61.882,
|
| 688 |
+
"eval_steps_per_second": 7.76,
|
| 689 |
+
"eval_wer": 0.9784189004997729,
|
| 690 |
+
"step": 6700
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 10.98,
|
| 694 |
+
"eval_loss": 2.72574782371521,
|
| 695 |
+
"eval_runtime": 35.3787,
|
| 696 |
+
"eval_samples_per_second": 62.213,
|
| 697 |
+
"eval_steps_per_second": 7.801,
|
| 698 |
+
"eval_wer": 0.9997728305315765,
|
| 699 |
+
"step": 6800
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 11.14,
|
| 703 |
+
"eval_loss": 2.672818183898926,
|
| 704 |
+
"eval_runtime": 35.1351,
|
| 705 |
+
"eval_samples_per_second": 62.644,
|
| 706 |
+
"eval_steps_per_second": 7.855,
|
| 707 |
+
"eval_wer": 0.989777373920945,
|
| 708 |
+
"step": 6900
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 11.3,
|
| 712 |
+
"learning_rate": 0.0001360353535353535,
|
| 713 |
+
"loss": 2.486,
|
| 714 |
+
"step": 7000
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 11.3,
|
| 718 |
+
"eval_loss": 2.7347543239593506,
|
| 719 |
+
"eval_runtime": 35.0574,
|
| 720 |
+
"eval_samples_per_second": 62.783,
|
| 721 |
+
"eval_steps_per_second": 7.873,
|
| 722 |
+
"eval_wer": 0.9809177646524307,
|
| 723 |
+
"step": 7000
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 11.46,
|
| 727 |
+
"eval_loss": 2.705395460128784,
|
| 728 |
+
"eval_runtime": 35.0691,
|
| 729 |
+
"eval_samples_per_second": 62.762,
|
| 730 |
+
"eval_steps_per_second": 7.87,
|
| 731 |
+
"eval_wer": 0.9981826442526125,
|
| 732 |
+
"step": 7100
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 11.62,
|
| 736 |
+
"eval_loss": 2.725370168685913,
|
| 737 |
+
"eval_runtime": 35.6237,
|
| 738 |
+
"eval_samples_per_second": 61.785,
|
| 739 |
+
"eval_steps_per_second": 7.748,
|
| 740 |
+
"eval_wer": 0.9947751022262608,
|
| 741 |
+
"step": 7200
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 11.78,
|
| 745 |
+
"eval_loss": 2.74984073638916,
|
| 746 |
+
"eval_runtime": 36.5139,
|
| 747 |
+
"eval_samples_per_second": 60.278,
|
| 748 |
+
"eval_steps_per_second": 7.559,
|
| 749 |
+
"eval_wer": 0.9890958655156747,
|
| 750 |
+
"step": 7300
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 11.95,
|
| 754 |
+
"eval_loss": 2.707601308822632,
|
| 755 |
+
"eval_runtime": 34.9847,
|
| 756 |
+
"eval_samples_per_second": 62.913,
|
| 757 |
+
"eval_steps_per_second": 7.889,
|
| 758 |
+
"eval_wer": 0.989777373920945,
|
| 759 |
+
"step": 7400
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 12.11,
|
| 763 |
+
"learning_rate": 0.0001234090909090909,
|
| 764 |
+
"loss": 2.4616,
|
| 765 |
+
"step": 7500
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 12.11,
|
| 769 |
+
"eval_loss": 2.639796733856201,
|
| 770 |
+
"eval_runtime": 34.999,
|
| 771 |
+
"eval_samples_per_second": 62.888,
|
| 772 |
+
"eval_steps_per_second": 7.886,
|
| 773 |
+
"eval_wer": 0.9995456610631531,
|
| 774 |
+
"step": 7500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 12.27,
|
| 778 |
+
"eval_loss": 2.7625598907470703,
|
| 779 |
+
"eval_runtime": 35.1361,
|
| 780 |
+
"eval_samples_per_second": 62.642,
|
| 781 |
+
"eval_steps_per_second": 7.855,
|
| 782 |
+
"eval_wer": 0.9845524761472059,
|
| 783 |
+
"step": 7600
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 12.43,
|
| 787 |
+
"eval_loss": 2.6803925037384033,
|
| 788 |
+
"eval_runtime": 34.9233,
|
| 789 |
+
"eval_samples_per_second": 63.024,
|
| 790 |
+
"eval_steps_per_second": 7.903,
|
| 791 |
+
"eval_wer": 0.9813721035892776,
|
| 792 |
+
"step": 7700
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 12.59,
|
| 796 |
+
"eval_loss": 2.82122540473938,
|
| 797 |
+
"eval_runtime": 35.0324,
|
| 798 |
+
"eval_samples_per_second": 62.828,
|
| 799 |
+
"eval_steps_per_second": 7.878,
|
| 800 |
+
"eval_wer": 0.9834166288050886,
|
| 801 |
+
"step": 7800
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 12.75,
|
| 805 |
+
"eval_loss": 2.6534788608551025,
|
| 806 |
+
"eval_runtime": 35.3231,
|
| 807 |
+
"eval_samples_per_second": 62.311,
|
| 808 |
+
"eval_steps_per_second": 7.814,
|
| 809 |
+
"eval_wer": 0.995910949568378,
|
| 810 |
+
"step": 7900
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 12.91,
|
| 814 |
+
"learning_rate": 0.00011078282828282827,
|
| 815 |
+
"loss": 2.4573,
|
| 816 |
+
"step": 8000
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 12.91,
|
| 820 |
+
"eval_loss": 2.754741668701172,
|
| 821 |
+
"eval_runtime": 35.0413,
|
| 822 |
+
"eval_samples_per_second": 62.812,
|
| 823 |
+
"eval_steps_per_second": 7.876,
|
| 824 |
+
"eval_wer": 0.9993184915947296,
|
| 825 |
+
"step": 8000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 13.08,
|
| 829 |
+
"eval_loss": 2.7253224849700928,
|
| 830 |
+
"eval_runtime": 35.5654,
|
| 831 |
+
"eval_samples_per_second": 61.886,
|
| 832 |
+
"eval_steps_per_second": 7.76,
|
| 833 |
+
"eval_wer": 0.9797819173103135,
|
| 834 |
+
"step": 8100
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 13.24,
|
| 838 |
+
"eval_loss": 2.6850526332855225,
|
| 839 |
+
"eval_runtime": 35.1914,
|
| 840 |
+
"eval_samples_per_second": 62.544,
|
| 841 |
+
"eval_steps_per_second": 7.843,
|
| 842 |
+
"eval_wer": 0.9936392548841436,
|
| 843 |
+
"step": 8200
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 13.4,
|
| 847 |
+
"eval_loss": 2.7627077102661133,
|
| 848 |
+
"eval_runtime": 35.2815,
|
| 849 |
+
"eval_samples_per_second": 62.384,
|
| 850 |
+
"eval_steps_per_second": 7.823,
|
| 851 |
+
"eval_wer": 0.9906860517946388,
|
| 852 |
+
"step": 8300
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 13.56,
|
| 856 |
+
"eval_loss": 2.6607019901275635,
|
| 857 |
+
"eval_runtime": 34.8716,
|
| 858 |
+
"eval_samples_per_second": 63.117,
|
| 859 |
+
"eval_steps_per_second": 7.915,
|
| 860 |
+
"eval_wer": 0.985688323489323,
|
| 861 |
+
"step": 8400
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 13.72,
|
| 865 |
+
"learning_rate": 9.815656565656566e-05,
|
| 866 |
+
"loss": 2.4487,
|
| 867 |
+
"step": 8500
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 13.72,
|
| 871 |
+
"eval_loss": 2.6645030975341797,
|
| 872 |
+
"eval_runtime": 34.8917,
|
| 873 |
+
"eval_samples_per_second": 63.081,
|
| 874 |
+
"eval_steps_per_second": 7.91,
|
| 875 |
+
"eval_wer": 0.980009086778737,
|
| 876 |
+
"step": 8500
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 13.88,
|
| 880 |
+
"eval_loss": 2.7558000087738037,
|
| 881 |
+
"eval_runtime": 35.3637,
|
| 882 |
+
"eval_samples_per_second": 62.239,
|
| 883 |
+
"eval_steps_per_second": 7.805,
|
| 884 |
+
"eval_wer": 0.9972739663789186,
|
| 885 |
+
"step": 8600
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 14.04,
|
| 889 |
+
"eval_loss": 2.7665319442749023,
|
| 890 |
+
"eval_runtime": 35.5805,
|
| 891 |
+
"eval_samples_per_second": 61.86,
|
| 892 |
+
"eval_steps_per_second": 7.757,
|
| 893 |
+
"eval_wer": 0.9961381190368015,
|
| 894 |
+
"step": 8700
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 14.21,
|
| 898 |
+
"eval_loss": 2.7696964740753174,
|
| 899 |
+
"eval_runtime": 36.3477,
|
| 900 |
+
"eval_samples_per_second": 60.554,
|
| 901 |
+
"eval_steps_per_second": 7.593,
|
| 902 |
+
"eval_wer": 0.9827351203998183,
|
| 903 |
+
"step": 8800
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 14.37,
|
| 907 |
+
"eval_loss": 2.8530843257904053,
|
| 908 |
+
"eval_runtime": 35.2199,
|
| 909 |
+
"eval_samples_per_second": 62.493,
|
| 910 |
+
"eval_steps_per_second": 7.836,
|
| 911 |
+
"eval_wer": 0.991821899136756,
|
| 912 |
+
"step": 8900
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 14.53,
|
| 916 |
+
"learning_rate": 8.555555555555556e-05,
|
| 917 |
+
"loss": 2.4416,
|
| 918 |
+
"step": 9000
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 14.53,
|
| 922 |
+
"eval_loss": 2.8973543643951416,
|
| 923 |
+
"eval_runtime": 35.1322,
|
| 924 |
+
"eval_samples_per_second": 62.649,
|
| 925 |
+
"eval_steps_per_second": 7.856,
|
| 926 |
+
"eval_wer": 0.9920490686051795,
|
| 927 |
+
"step": 9000
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 14.69,
|
| 931 |
+
"eval_loss": 2.7307872772216797,
|
| 932 |
+
"eval_runtime": 35.6885,
|
| 933 |
+
"eval_samples_per_second": 61.673,
|
| 934 |
+
"eval_steps_per_second": 7.734,
|
| 935 |
+
"eval_wer": 0.9975011358473421,
|
| 936 |
+
"step": 9100
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 14.85,
|
| 940 |
+
"eval_loss": 2.7919318675994873,
|
| 941 |
+
"eval_runtime": 35.3942,
|
| 942 |
+
"eval_samples_per_second": 62.185,
|
| 943 |
+
"eval_steps_per_second": 7.798,
|
| 944 |
+
"eval_wer": 0.981599273057701,
|
| 945 |
+
"step": 9200
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 15.01,
|
| 949 |
+
"eval_loss": 2.660515308380127,
|
| 950 |
+
"eval_runtime": 35.5069,
|
| 951 |
+
"eval_samples_per_second": 61.988,
|
| 952 |
+
"eval_steps_per_second": 7.773,
|
| 953 |
+
"eval_wer": 0.9893230349840981,
|
| 954 |
+
"step": 9300
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 15.17,
|
| 958 |
+
"eval_loss": 2.605780839920044,
|
| 959 |
+
"eval_runtime": 35.3171,
|
| 960 |
+
"eval_samples_per_second": 62.321,
|
| 961 |
+
"eval_steps_per_second": 7.815,
|
| 962 |
+
"eval_wer": 0.981599273057701,
|
| 963 |
+
"step": 9400
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 15.33,
|
| 967 |
+
"learning_rate": 7.292929292929292e-05,
|
| 968 |
+
"loss": 2.4405,
|
| 969 |
+
"step": 9500
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 15.33,
|
| 973 |
+
"eval_loss": 2.63655161857605,
|
| 974 |
+
"eval_runtime": 35.0912,
|
| 975 |
+
"eval_samples_per_second": 62.722,
|
| 976 |
+
"eval_steps_per_second": 7.865,
|
| 977 |
+
"eval_wer": 0.9911403907314857,
|
| 978 |
+
"step": 9500
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 15.5,
|
| 982 |
+
"eval_loss": 2.565321207046509,
|
| 983 |
+
"eval_runtime": 35.6382,
|
| 984 |
+
"eval_samples_per_second": 61.76,
|
| 985 |
+
"eval_steps_per_second": 7.745,
|
| 986 |
+
"eval_wer": 0.9818264425261245,
|
| 987 |
+
"step": 9600
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 15.66,
|
| 991 |
+
"eval_loss": 2.7025833129882812,
|
| 992 |
+
"eval_runtime": 35.355,
|
| 993 |
+
"eval_samples_per_second": 62.254,
|
| 994 |
+
"eval_steps_per_second": 7.807,
|
| 995 |
+
"eval_wer": 0.9806905951840073,
|
| 996 |
+
"step": 9700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 15.82,
|
| 1000 |
+
"eval_loss": 2.7358055114746094,
|
| 1001 |
+
"eval_runtime": 35.0441,
|
| 1002 |
+
"eval_samples_per_second": 62.807,
|
| 1003 |
+
"eval_steps_per_second": 7.876,
|
| 1004 |
+
"eval_wer": 0.97955474784189,
|
| 1005 |
+
"step": 9800
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 15.98,
|
| 1009 |
+
"eval_loss": 2.695399522781372,
|
| 1010 |
+
"eval_runtime": 35.2995,
|
| 1011 |
+
"eval_samples_per_second": 62.352,
|
| 1012 |
+
"eval_steps_per_second": 7.819,
|
| 1013 |
+
"eval_wer": 0.9847796456156293,
|
| 1014 |
+
"step": 9900
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 16.14,
|
| 1018 |
+
"learning_rate": 6.0303030303030296e-05,
|
| 1019 |
+
"loss": 2.4352,
|
| 1020 |
+
"step": 10000
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 16.14,
|
| 1024 |
+
"eval_loss": 2.661024570465088,
|
| 1025 |
+
"eval_runtime": 35.3284,
|
| 1026 |
+
"eval_samples_per_second": 62.301,
|
| 1027 |
+
"eval_steps_per_second": 7.812,
|
| 1028 |
+
"eval_wer": 0.985688323489323,
|
| 1029 |
+
"step": 10000
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 16.3,
|
| 1033 |
+
"eval_loss": 2.7686407566070557,
|
| 1034 |
+
"eval_runtime": 35.066,
|
| 1035 |
+
"eval_samples_per_second": 62.767,
|
| 1036 |
+
"eval_steps_per_second": 7.871,
|
| 1037 |
+
"eval_wer": 0.9811449341208541,
|
| 1038 |
+
"step": 10100
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 16.46,
|
| 1042 |
+
"eval_loss": 2.775808572769165,
|
| 1043 |
+
"eval_runtime": 34.9312,
|
| 1044 |
+
"eval_samples_per_second": 63.01,
|
| 1045 |
+
"eval_steps_per_second": 7.901,
|
| 1046 |
+
"eval_wer": 0.9797819173103135,
|
| 1047 |
+
"step": 10200
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 16.63,
|
| 1051 |
+
"eval_loss": 2.751512050628662,
|
| 1052 |
+
"eval_runtime": 35.1597,
|
| 1053 |
+
"eval_samples_per_second": 62.6,
|
| 1054 |
+
"eval_steps_per_second": 7.85,
|
| 1055 |
+
"eval_wer": 0.9847796456156293,
|
| 1056 |
+
"step": 10300
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 16.79,
|
| 1060 |
+
"eval_loss": 2.726439952850342,
|
| 1061 |
+
"eval_runtime": 35.8311,
|
| 1062 |
+
"eval_samples_per_second": 61.427,
|
| 1063 |
+
"eval_steps_per_second": 7.703,
|
| 1064 |
+
"eval_wer": 0.9911403907314857,
|
| 1065 |
+
"step": 10400
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 16.95,
|
| 1069 |
+
"learning_rate": 4.767676767676767e-05,
|
| 1070 |
+
"loss": 2.4354,
|
| 1071 |
+
"step": 10500
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 16.95,
|
| 1075 |
+
"eval_loss": 2.7038702964782715,
|
| 1076 |
+
"eval_runtime": 35.4185,
|
| 1077 |
+
"eval_samples_per_second": 62.143,
|
| 1078 |
+
"eval_steps_per_second": 7.793,
|
| 1079 |
+
"eval_wer": 0.9791004089050431,
|
| 1080 |
+
"step": 10500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 17.11,
|
| 1084 |
+
"eval_loss": 2.7579615116119385,
|
| 1085 |
+
"eval_runtime": 35.1081,
|
| 1086 |
+
"eval_samples_per_second": 62.692,
|
| 1087 |
+
"eval_steps_per_second": 7.861,
|
| 1088 |
+
"eval_wer": 0.9843253066787824,
|
| 1089 |
+
"step": 10600
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 17.27,
|
| 1093 |
+
"eval_loss": 2.7186856269836426,
|
| 1094 |
+
"eval_runtime": 35.331,
|
| 1095 |
+
"eval_samples_per_second": 62.297,
|
| 1096 |
+
"eval_steps_per_second": 7.812,
|
| 1097 |
+
"eval_wer": 0.9854611540208996,
|
| 1098 |
+
"step": 10700
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 17.43,
|
| 1102 |
+
"eval_loss": 2.7545225620269775,
|
| 1103 |
+
"eval_runtime": 35.2267,
|
| 1104 |
+
"eval_samples_per_second": 62.481,
|
| 1105 |
+
"eval_steps_per_second": 7.835,
|
| 1106 |
+
"eval_wer": 0.9797819173103135,
|
| 1107 |
+
"step": 10800
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 17.59,
|
| 1111 |
+
"eval_loss": 2.745208501815796,
|
| 1112 |
+
"eval_runtime": 35.2608,
|
| 1113 |
+
"eval_samples_per_second": 62.421,
|
| 1114 |
+
"eval_steps_per_second": 7.827,
|
| 1115 |
+
"eval_wer": 0.9809177646524307,
|
| 1116 |
+
"step": 10900
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 17.76,
|
| 1120 |
+
"learning_rate": 3.5050505050505044e-05,
|
| 1121 |
+
"loss": 2.4321,
|
| 1122 |
+
"step": 11000
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 17.76,
|
| 1126 |
+
"eval_loss": 2.680445432662964,
|
| 1127 |
+
"eval_runtime": 35.0774,
|
| 1128 |
+
"eval_samples_per_second": 62.747,
|
| 1129 |
+
"eval_steps_per_second": 7.868,
|
| 1130 |
+
"eval_wer": 0.983643798273512,
|
| 1131 |
+
"step": 11000
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 17.92,
|
| 1135 |
+
"eval_loss": 2.6586036682128906,
|
| 1136 |
+
"eval_runtime": 35.3748,
|
| 1137 |
+
"eval_samples_per_second": 62.219,
|
| 1138 |
+
"eval_steps_per_second": 7.802,
|
| 1139 |
+
"eval_wer": 0.9890958655156747,
|
| 1140 |
+
"step": 11100
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 18.08,
|
| 1144 |
+
"eval_loss": 2.680509328842163,
|
| 1145 |
+
"eval_runtime": 35.5491,
|
| 1146 |
+
"eval_samples_per_second": 61.914,
|
| 1147 |
+
"eval_steps_per_second": 7.764,
|
| 1148 |
+
"eval_wer": 0.9829622898682417,
|
| 1149 |
+
"step": 11200
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 18.24,
|
| 1153 |
+
"eval_loss": 2.662594795227051,
|
| 1154 |
+
"eval_runtime": 35.3779,
|
| 1155 |
+
"eval_samples_per_second": 62.214,
|
| 1156 |
+
"eval_steps_per_second": 7.801,
|
| 1157 |
+
"eval_wer": 0.9870513402998637,
|
| 1158 |
+
"step": 11300
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 18.4,
|
| 1162 |
+
"eval_loss": 2.70019268989563,
|
| 1163 |
+
"eval_runtime": 35.0267,
|
| 1164 |
+
"eval_samples_per_second": 62.838,
|
| 1165 |
+
"eval_steps_per_second": 7.88,
|
| 1166 |
+
"eval_wer": 0.9809177646524307,
|
| 1167 |
+
"step": 11400
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 18.56,
|
| 1171 |
+
"learning_rate": 2.242424242424242e-05,
|
| 1172 |
+
"loss": 2.4193,
|
| 1173 |
+
"step": 11500
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 18.56,
|
| 1177 |
+
"eval_loss": 2.7054154872894287,
|
| 1178 |
+
"eval_runtime": 35.0708,
|
| 1179 |
+
"eval_samples_per_second": 62.759,
|
| 1180 |
+
"eval_steps_per_second": 7.87,
|
| 1181 |
+
"eval_wer": 0.9838709677419355,
|
| 1182 |
+
"step": 11500
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 18.72,
|
| 1186 |
+
"eval_loss": 2.717078447341919,
|
| 1187 |
+
"eval_runtime": 35.1262,
|
| 1188 |
+
"eval_samples_per_second": 62.66,
|
| 1189 |
+
"eval_steps_per_second": 7.857,
|
| 1190 |
+
"eval_wer": 0.9900045433893685,
|
| 1191 |
+
"step": 11600
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 18.89,
|
| 1195 |
+
"eval_loss": 2.712236166000366,
|
| 1196 |
+
"eval_runtime": 35.0347,
|
| 1197 |
+
"eval_samples_per_second": 62.823,
|
| 1198 |
+
"eval_steps_per_second": 7.878,
|
| 1199 |
+
"eval_wer": 0.9852339845524761,
|
| 1200 |
+
"step": 11700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 19.05,
|
| 1204 |
+
"eval_loss": 2.705820083618164,
|
| 1205 |
+
"eval_runtime": 35.6595,
|
| 1206 |
+
"eval_samples_per_second": 61.723,
|
| 1207 |
+
"eval_steps_per_second": 7.74,
|
| 1208 |
+
"eval_wer": 0.9870513402998637,
|
| 1209 |
+
"step": 11800
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 19.21,
|
| 1213 |
+
"eval_loss": 2.7003977298736572,
|
| 1214 |
+
"eval_runtime": 35.7161,
|
| 1215 |
+
"eval_samples_per_second": 61.625,
|
| 1216 |
+
"eval_steps_per_second": 7.728,
|
| 1217 |
+
"eval_wer": 0.9838709677419355,
|
| 1218 |
+
"step": 11900
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 19.37,
|
| 1222 |
+
"learning_rate": 9.797979797979796e-06,
|
| 1223 |
+
"loss": 2.4276,
|
| 1224 |
+
"step": 12000
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 19.37,
|
| 1228 |
+
"eval_loss": 2.725003719329834,
|
| 1229 |
+
"eval_runtime": 35.3082,
|
| 1230 |
+
"eval_samples_per_second": 62.337,
|
| 1231 |
+
"eval_steps_per_second": 7.817,
|
| 1232 |
+
"eval_wer": 0.9852339845524761,
|
| 1233 |
+
"step": 12000
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 19.53,
|
| 1237 |
+
"eval_loss": 2.7126049995422363,
|
| 1238 |
+
"eval_runtime": 35.1508,
|
| 1239 |
+
"eval_samples_per_second": 62.616,
|
| 1240 |
+
"eval_steps_per_second": 7.852,
|
| 1241 |
+
"eval_wer": 0.9861426624261699,
|
| 1242 |
+
"step": 12100
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 19.69,
|
| 1246 |
+
"eval_loss": 2.738818645477295,
|
| 1247 |
+
"eval_runtime": 35.2814,
|
| 1248 |
+
"eval_samples_per_second": 62.384,
|
| 1249 |
+
"eval_steps_per_second": 7.823,
|
| 1250 |
+
"eval_wer": 0.9834166288050886,
|
| 1251 |
+
"step": 12200
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 19.85,
|
| 1255 |
+
"eval_loss": 2.7310702800750732,
|
| 1256 |
+
"eval_runtime": 35.2924,
|
| 1257 |
+
"eval_samples_per_second": 62.365,
|
| 1258 |
+
"eval_steps_per_second": 7.82,
|
| 1259 |
+
"eval_wer": 0.9850068150840527,
|
| 1260 |
+
"step": 12300
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 19.98,
|
| 1264 |
+
"step": 12380,
|
| 1265 |
+
"total_flos": 1.642150502507821e+19,
|
| 1266 |
+
"train_loss": 2.556843002393288,
|
| 1267 |
+
"train_runtime": 17303.3516,
|
| 1268 |
+
"train_samples_per_second": 22.908,
|
| 1269 |
+
"train_steps_per_second": 0.715
|
| 1270 |
}
|
| 1271 |
],
|
| 1272 |
"logging_steps": 500,
|
| 1273 |
+
"max_steps": 12380,
|
| 1274 |
+
"num_train_epochs": 20,
|
| 1275 |
"save_steps": 400,
|
| 1276 |
+
"total_flos": 1.642150502507821e+19,
|
| 1277 |
"trial_name": null,
|
| 1278 |
"trial_params": null
|
| 1279 |
}
|