End of training
Browse files- README.md +11 -7
- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +1752 -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 |
datasets:
|
| 8 |
- common_voice_13_0
|
|
@@ -15,15 +19,15 @@ model-index:
|
|
| 15 |
name: Automatic Speech Recognition
|
| 16 |
type: automatic-speech-recognition
|
| 17 |
dataset:
|
| 18 |
-
name:
|
| 19 |
type: common_voice_13_0
|
| 20 |
config: ja
|
| 21 |
split: test
|
| 22 |
-
args: ja
|
| 23 |
metrics:
|
| 24 |
- name: Wer
|
| 25 |
type: wer
|
| 26 |
-
value: 0.
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -31,11 +35,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 31 |
|
| 32 |
# Hubert-common_voice-ja-demo-roma-debug-40epochs-cosine
|
| 33 |
|
| 34 |
-
This model is a fine-tuned version of [rinna/japanese-hubert-base](https://huggingface.co/rinna/japanese-hubert-base) on the
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
-
- Wer: 0.
|
| 38 |
-
- Cer: 0.
|
| 39 |
|
| 40 |
## Model description
|
| 41 |
|
|
|
|
| 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 |
+
- mozilla-foundation/common_voice_13_0
|
| 10 |
- generated_from_trainer
|
| 11 |
datasets:
|
| 12 |
- common_voice_13_0
|
|
|
|
| 19 |
name: Automatic Speech Recognition
|
| 20 |
type: automatic-speech-recognition
|
| 21 |
dataset:
|
| 22 |
+
name: MOZILLA-FOUNDATION/COMMON_VOICE_13_0 - JA
|
| 23 |
type: common_voice_13_0
|
| 24 |
config: ja
|
| 25 |
split: test
|
| 26 |
+
args: 'Config: ja, Training split: train+validation, Eval split: test'
|
| 27 |
metrics:
|
| 28 |
- name: Wer
|
| 29 |
type: wer
|
| 30 |
+
value: 0.9989917322040734
|
| 31 |
---
|
| 32 |
|
| 33 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 35 |
|
| 36 |
# Hubert-common_voice-ja-demo-roma-debug-40epochs-cosine
|
| 37 |
|
| 38 |
+
This model is a fine-tuned version of [rinna/japanese-hubert-base](https://huggingface.co/rinna/japanese-hubert-base) on the MOZILLA-FOUNDATION/COMMON_VOICE_13_0 - JA dataset.
|
| 39 |
It achieves the following results on the evaluation set:
|
| 40 |
+
- Loss: 0.5318
|
| 41 |
+
- Wer: 0.9990
|
| 42 |
+
- Cer: 0.1993
|
| 43 |
|
| 44 |
## Model description
|
| 45 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_cer": 0.19931834260319645,
|
| 4 |
+
"eval_loss": 0.5317806005477905,
|
| 5 |
+
"eval_runtime": 212.2678,
|
| 6 |
+
"eval_samples": 4959,
|
| 7 |
+
"eval_samples_per_second": 23.362,
|
| 8 |
+
"eval_steps_per_second": 2.921,
|
| 9 |
+
"eval_wer": 0.9989917322040734,
|
| 10 |
+
"total_flos": 2.1018408189762245e+19,
|
| 11 |
+
"train_loss": 1.5314149432993949,
|
| 12 |
+
"train_runtime": 82631.5291,
|
| 13 |
+
"train_samples": 12031,
|
| 14 |
+
"train_samples_per_second": 5.824,
|
| 15 |
+
"train_steps_per_second": 0.182
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_cer": 0.19931834260319645,
|
| 4 |
+
"eval_loss": 0.5317806005477905,
|
| 5 |
+
"eval_runtime": 212.2678,
|
| 6 |
+
"eval_samples": 4959,
|
| 7 |
+
"eval_samples_per_second": 23.362,
|
| 8 |
+
"eval_steps_per_second": 2.921,
|
| 9 |
+
"eval_wer": 0.9989917322040734
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"total_flos": 2.1018408189762245e+19,
|
| 4 |
+
"train_loss": 1.5314149432993949,
|
| 5 |
+
"train_runtime": 82631.5291,
|
| 6 |
+
"train_samples": 12031,
|
| 7 |
+
"train_samples_per_second": 5.824,
|
| 8 |
+
"train_steps_per_second": 0.182
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1752 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.441910982131958,
|
| 3 |
+
"best_model_checkpoint": "./Hubert-common_voice-ja-demo-roma-debug-40epochs-cosine/checkpoint-9700",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 15040,
|
| 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": 3.477914534858724,
|
| 14 |
+
"eval_loss": 16.817241668701172,
|
| 15 |
+
"eval_runtime": 221.1037,
|
| 16 |
+
"eval_samples_per_second": 22.428,
|
| 17 |
+
"eval_steps_per_second": 2.804,
|
| 18 |
+
"eval_wer": 2.9493849566444847,
|
| 19 |
+
"step": 100
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.5319148936170213,
|
| 23 |
+
"eval_cer": 2.9561261222527526,
|
| 24 |
+
"eval_loss": 16.550180435180664,
|
| 25 |
+
"eval_runtime": 213.5908,
|
| 26 |
+
"eval_samples_per_second": 23.217,
|
| 27 |
+
"eval_steps_per_second": 2.903,
|
| 28 |
+
"eval_wer": 2.7606372252470255,
|
| 29 |
+
"step": 200
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.7978723404255319,
|
| 33 |
+
"eval_cer": 1.6838657478883865,
|
| 34 |
+
"eval_loss": 15.8340425491333,
|
| 35 |
+
"eval_runtime": 213.9597,
|
| 36 |
+
"eval_samples_per_second": 23.177,
|
| 37 |
+
"eval_steps_per_second": 2.898,
|
| 38 |
+
"eval_wer": 1.9100625126033475,
|
| 39 |
+
"step": 300
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0638297872340425,
|
| 43 |
+
"eval_cer": 0.9275934445195761,
|
| 44 |
+
"eval_loss": 13.291926383972168,
|
| 45 |
+
"eval_runtime": 211.7589,
|
| 46 |
+
"eval_samples_per_second": 23.418,
|
| 47 |
+
"eval_steps_per_second": 2.928,
|
| 48 |
+
"eval_wer": 1.0,
|
| 49 |
+
"step": 400
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 1.3297872340425532,
|
| 53 |
+
"grad_norm": 41.734230041503906,
|
| 54 |
+
"learning_rate": 1.1904e-06,
|
| 55 |
+
"loss": 12.6588,
|
| 56 |
+
"step": 500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.3297872340425532,
|
| 60 |
+
"eval_cer": 0.9276012616915579,
|
| 61 |
+
"eval_loss": 7.879184246063232,
|
| 62 |
+
"eval_runtime": 213.1723,
|
| 63 |
+
"eval_samples_per_second": 23.263,
|
| 64 |
+
"eval_steps_per_second": 2.908,
|
| 65 |
+
"eval_wer": 1.0,
|
| 66 |
+
"step": 500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.5957446808510638,
|
| 70 |
+
"eval_cer": 0.9276012616915579,
|
| 71 |
+
"eval_loss": 6.154232501983643,
|
| 72 |
+
"eval_runtime": 212.0332,
|
| 73 |
+
"eval_samples_per_second": 23.388,
|
| 74 |
+
"eval_steps_per_second": 2.924,
|
| 75 |
+
"eval_wer": 1.0,
|
| 76 |
+
"step": 600
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 1.8617021276595744,
|
| 80 |
+
"eval_cer": 0.9276012616915579,
|
| 81 |
+
"eval_loss": 5.775670051574707,
|
| 82 |
+
"eval_runtime": 210.8145,
|
| 83 |
+
"eval_samples_per_second": 23.523,
|
| 84 |
+
"eval_steps_per_second": 2.941,
|
| 85 |
+
"eval_wer": 1.0,
|
| 86 |
+
"step": 700
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.127659574468085,
|
| 90 |
+
"eval_cer": 0.9276012616915579,
|
| 91 |
+
"eval_loss": 5.6188063621521,
|
| 92 |
+
"eval_runtime": 211.2888,
|
| 93 |
+
"eval_samples_per_second": 23.47,
|
| 94 |
+
"eval_steps_per_second": 2.934,
|
| 95 |
+
"eval_wer": 1.0,
|
| 96 |
+
"step": 800
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 2.393617021276596,
|
| 100 |
+
"eval_cer": 0.9276012616915579,
|
| 101 |
+
"eval_loss": 5.475332260131836,
|
| 102 |
+
"eval_runtime": 208.2147,
|
| 103 |
+
"eval_samples_per_second": 23.817,
|
| 104 |
+
"eval_steps_per_second": 2.978,
|
| 105 |
+
"eval_wer": 1.0,
|
| 106 |
+
"step": 900
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 2.6595744680851063,
|
| 110 |
+
"grad_norm": 41.6483154296875,
|
| 111 |
+
"learning_rate": 2.3904e-06,
|
| 112 |
+
"loss": 5.2353,
|
| 113 |
+
"step": 1000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.6595744680851063,
|
| 117 |
+
"eval_cer": 0.9276012616915579,
|
| 118 |
+
"eval_loss": 5.3238911628723145,
|
| 119 |
+
"eval_runtime": 208.8415,
|
| 120 |
+
"eval_samples_per_second": 23.745,
|
| 121 |
+
"eval_steps_per_second": 2.969,
|
| 122 |
+
"eval_wer": 1.0,
|
| 123 |
+
"step": 1000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.925531914893617,
|
| 127 |
+
"eval_cer": 0.9276012616915579,
|
| 128 |
+
"eval_loss": 5.167562961578369,
|
| 129 |
+
"eval_runtime": 207.9905,
|
| 130 |
+
"eval_samples_per_second": 23.842,
|
| 131 |
+
"eval_steps_per_second": 2.981,
|
| 132 |
+
"eval_wer": 1.0,
|
| 133 |
+
"step": 1100
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 3.1914893617021276,
|
| 137 |
+
"eval_cer": 0.9276012616915579,
|
| 138 |
+
"eval_loss": 5.008359432220459,
|
| 139 |
+
"eval_runtime": 208.6683,
|
| 140 |
+
"eval_samples_per_second": 23.765,
|
| 141 |
+
"eval_steps_per_second": 2.971,
|
| 142 |
+
"eval_wer": 1.0,
|
| 143 |
+
"step": 1200
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.4574468085106385,
|
| 147 |
+
"eval_cer": 0.9276012616915579,
|
| 148 |
+
"eval_loss": 4.840206146240234,
|
| 149 |
+
"eval_runtime": 207.1918,
|
| 150 |
+
"eval_samples_per_second": 23.934,
|
| 151 |
+
"eval_steps_per_second": 2.992,
|
| 152 |
+
"eval_wer": 1.0,
|
| 153 |
+
"step": 1300
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 3.723404255319149,
|
| 157 |
+
"eval_cer": 0.9276012616915579,
|
| 158 |
+
"eval_loss": 4.670183181762695,
|
| 159 |
+
"eval_runtime": 206.5028,
|
| 160 |
+
"eval_samples_per_second": 24.014,
|
| 161 |
+
"eval_steps_per_second": 3.002,
|
| 162 |
+
"eval_wer": 1.0,
|
| 163 |
+
"step": 1400
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 3.9893617021276597,
|
| 167 |
+
"grad_norm": 25.022253036499023,
|
| 168 |
+
"learning_rate": 3.5904e-06,
|
| 169 |
+
"loss": 4.4502,
|
| 170 |
+
"step": 1500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.9893617021276597,
|
| 174 |
+
"eval_cer": 0.9276012616915579,
|
| 175 |
+
"eval_loss": 4.495745658874512,
|
| 176 |
+
"eval_runtime": 206.573,
|
| 177 |
+
"eval_samples_per_second": 24.006,
|
| 178 |
+
"eval_steps_per_second": 3.001,
|
| 179 |
+
"eval_wer": 1.0,
|
| 180 |
+
"step": 1500
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 4.25531914893617,
|
| 184 |
+
"eval_cer": 0.9276012616915579,
|
| 185 |
+
"eval_loss": 4.321905136108398,
|
| 186 |
+
"eval_runtime": 208.0558,
|
| 187 |
+
"eval_samples_per_second": 23.835,
|
| 188 |
+
"eval_steps_per_second": 2.98,
|
| 189 |
+
"eval_wer": 1.0,
|
| 190 |
+
"step": 1600
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 4.5212765957446805,
|
| 194 |
+
"eval_cer": 0.9276012616915579,
|
| 195 |
+
"eval_loss": 4.150215148925781,
|
| 196 |
+
"eval_runtime": 213.3593,
|
| 197 |
+
"eval_samples_per_second": 23.242,
|
| 198 |
+
"eval_steps_per_second": 2.906,
|
| 199 |
+
"eval_wer": 1.0,
|
| 200 |
+
"step": 1700
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.787234042553192,
|
| 204 |
+
"eval_cer": 0.9276012616915579,
|
| 205 |
+
"eval_loss": 3.9856085777282715,
|
| 206 |
+
"eval_runtime": 207.2181,
|
| 207 |
+
"eval_samples_per_second": 23.931,
|
| 208 |
+
"eval_steps_per_second": 2.992,
|
| 209 |
+
"eval_wer": 1.0,
|
| 210 |
+
"step": 1800
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.053191489361702,
|
| 214 |
+
"eval_cer": 0.9276012616915579,
|
| 215 |
+
"eval_loss": 3.834258794784546,
|
| 216 |
+
"eval_runtime": 206.1057,
|
| 217 |
+
"eval_samples_per_second": 24.06,
|
| 218 |
+
"eval_steps_per_second": 3.008,
|
| 219 |
+
"eval_wer": 1.0,
|
| 220 |
+
"step": 1900
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 5.319148936170213,
|
| 224 |
+
"grad_norm": 12.537736892700195,
|
| 225 |
+
"learning_rate": 4.7904e-06,
|
| 226 |
+
"loss": 3.7863,
|
| 227 |
+
"step": 2000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 5.319148936170213,
|
| 231 |
+
"eval_cer": 0.9276012616915579,
|
| 232 |
+
"eval_loss": 3.690730333328247,
|
| 233 |
+
"eval_runtime": 207.5053,
|
| 234 |
+
"eval_samples_per_second": 23.898,
|
| 235 |
+
"eval_steps_per_second": 2.988,
|
| 236 |
+
"eval_wer": 1.0,
|
| 237 |
+
"step": 2000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 5.585106382978723,
|
| 241 |
+
"eval_cer": 0.9276012616915579,
|
| 242 |
+
"eval_loss": 3.5544023513793945,
|
| 243 |
+
"eval_runtime": 207.2908,
|
| 244 |
+
"eval_samples_per_second": 23.923,
|
| 245 |
+
"eval_steps_per_second": 2.991,
|
| 246 |
+
"eval_wer": 1.0,
|
| 247 |
+
"step": 2100
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 5.851063829787234,
|
| 251 |
+
"eval_cer": 0.9276012616915579,
|
| 252 |
+
"eval_loss": 3.4332330226898193,
|
| 253 |
+
"eval_runtime": 208.1267,
|
| 254 |
+
"eval_samples_per_second": 23.827,
|
| 255 |
+
"eval_steps_per_second": 2.979,
|
| 256 |
+
"eval_wer": 1.0,
|
| 257 |
+
"step": 2200
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 6.117021276595745,
|
| 261 |
+
"eval_cer": 0.9275817187616037,
|
| 262 |
+
"eval_loss": 3.3063294887542725,
|
| 263 |
+
"eval_runtime": 207.0236,
|
| 264 |
+
"eval_samples_per_second": 23.954,
|
| 265 |
+
"eval_steps_per_second": 2.995,
|
| 266 |
+
"eval_wer": 1.0,
|
| 267 |
+
"step": 2300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 6.382978723404255,
|
| 271 |
+
"eval_cer": 0.9276012616915579,
|
| 272 |
+
"eval_loss": 3.2074904441833496,
|
| 273 |
+
"eval_runtime": 207.8656,
|
| 274 |
+
"eval_samples_per_second": 23.857,
|
| 275 |
+
"eval_steps_per_second": 2.983,
|
| 276 |
+
"eval_wer": 1.0,
|
| 277 |
+
"step": 2400
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 6.648936170212766,
|
| 281 |
+
"grad_norm": 6.3363356590271,
|
| 282 |
+
"learning_rate": 5.9904e-06,
|
| 283 |
+
"loss": 3.2473,
|
| 284 |
+
"step": 2500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 6.648936170212766,
|
| 288 |
+
"eval_cer": 0.9276012616915579,
|
| 289 |
+
"eval_loss": 3.1271653175354004,
|
| 290 |
+
"eval_runtime": 209.3454,
|
| 291 |
+
"eval_samples_per_second": 23.688,
|
| 292 |
+
"eval_steps_per_second": 2.962,
|
| 293 |
+
"eval_wer": 1.0,
|
| 294 |
+
"step": 2500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 6.914893617021277,
|
| 298 |
+
"eval_cer": 0.9276012616915579,
|
| 299 |
+
"eval_loss": 3.0657362937927246,
|
| 300 |
+
"eval_runtime": 209.9086,
|
| 301 |
+
"eval_samples_per_second": 23.625,
|
| 302 |
+
"eval_steps_per_second": 2.954,
|
| 303 |
+
"eval_wer": 1.0,
|
| 304 |
+
"step": 2600
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 7.180851063829787,
|
| 308 |
+
"eval_cer": 0.9276012616915579,
|
| 309 |
+
"eval_loss": 3.0163941383361816,
|
| 310 |
+
"eval_runtime": 209.8668,
|
| 311 |
+
"eval_samples_per_second": 23.629,
|
| 312 |
+
"eval_steps_per_second": 2.954,
|
| 313 |
+
"eval_wer": 1.0,
|
| 314 |
+
"step": 2700
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 7.446808510638298,
|
| 318 |
+
"eval_cer": 0.9276012616915579,
|
| 319 |
+
"eval_loss": 2.9747605323791504,
|
| 320 |
+
"eval_runtime": 209.2203,
|
| 321 |
+
"eval_samples_per_second": 23.702,
|
| 322 |
+
"eval_steps_per_second": 2.963,
|
| 323 |
+
"eval_wer": 1.0,
|
| 324 |
+
"step": 2800
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 7.712765957446808,
|
| 328 |
+
"eval_cer": 0.9276012616915579,
|
| 329 |
+
"eval_loss": 2.9446794986724854,
|
| 330 |
+
"eval_runtime": 208.1917,
|
| 331 |
+
"eval_samples_per_second": 23.819,
|
| 332 |
+
"eval_steps_per_second": 2.978,
|
| 333 |
+
"eval_wer": 1.0,
|
| 334 |
+
"step": 2900
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 7.9787234042553195,
|
| 338 |
+
"grad_norm": 1.2298308610916138,
|
| 339 |
+
"learning_rate": 7.190400000000001e-06,
|
| 340 |
+
"loss": 2.9649,
|
| 341 |
+
"step": 3000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 7.9787234042553195,
|
| 345 |
+
"eval_cer": 0.9276012616915579,
|
| 346 |
+
"eval_loss": 2.918757915496826,
|
| 347 |
+
"eval_runtime": 208.2245,
|
| 348 |
+
"eval_samples_per_second": 23.816,
|
| 349 |
+
"eval_steps_per_second": 2.978,
|
| 350 |
+
"eval_wer": 1.0,
|
| 351 |
+
"step": 3000
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 8.24468085106383,
|
| 355 |
+
"eval_cer": 0.9276012616915579,
|
| 356 |
+
"eval_loss": 2.9005508422851562,
|
| 357 |
+
"eval_runtime": 205.5088,
|
| 358 |
+
"eval_samples_per_second": 24.13,
|
| 359 |
+
"eval_steps_per_second": 3.017,
|
| 360 |
+
"eval_wer": 1.0,
|
| 361 |
+
"step": 3100
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 8.51063829787234,
|
| 365 |
+
"eval_cer": 0.9276012616915579,
|
| 366 |
+
"eval_loss": 2.8877005577087402,
|
| 367 |
+
"eval_runtime": 205.8033,
|
| 368 |
+
"eval_samples_per_second": 24.096,
|
| 369 |
+
"eval_steps_per_second": 3.013,
|
| 370 |
+
"eval_wer": 1.0,
|
| 371 |
+
"step": 3200
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 8.77659574468085,
|
| 375 |
+
"eval_cer": 0.9276012616915579,
|
| 376 |
+
"eval_loss": 2.8712358474731445,
|
| 377 |
+
"eval_runtime": 206.9211,
|
| 378 |
+
"eval_samples_per_second": 23.966,
|
| 379 |
+
"eval_steps_per_second": 2.996,
|
| 380 |
+
"eval_wer": 1.0,
|
| 381 |
+
"step": 3300
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 9.042553191489361,
|
| 385 |
+
"eval_cer": 0.9276012616915579,
|
| 386 |
+
"eval_loss": 2.8528554439544678,
|
| 387 |
+
"eval_runtime": 205.6952,
|
| 388 |
+
"eval_samples_per_second": 24.108,
|
| 389 |
+
"eval_steps_per_second": 3.014,
|
| 390 |
+
"eval_wer": 1.0,
|
| 391 |
+
"step": 3400
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 9.308510638297872,
|
| 395 |
+
"grad_norm": 1.270645022392273,
|
| 396 |
+
"learning_rate": 8.3904e-06,
|
| 397 |
+
"loss": 2.8673,
|
| 398 |
+
"step": 3500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 9.308510638297872,
|
| 402 |
+
"eval_cer": 0.9276012616915579,
|
| 403 |
+
"eval_loss": 2.8438620567321777,
|
| 404 |
+
"eval_runtime": 206.4519,
|
| 405 |
+
"eval_samples_per_second": 24.02,
|
| 406 |
+
"eval_steps_per_second": 3.003,
|
| 407 |
+
"eval_wer": 1.0,
|
| 408 |
+
"step": 3500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 9.574468085106384,
|
| 412 |
+
"eval_cer": 0.9276012616915579,
|
| 413 |
+
"eval_loss": 2.8312649726867676,
|
| 414 |
+
"eval_runtime": 206.5101,
|
| 415 |
+
"eval_samples_per_second": 24.013,
|
| 416 |
+
"eval_steps_per_second": 3.002,
|
| 417 |
+
"eval_wer": 1.0,
|
| 418 |
+
"step": 3600
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 9.840425531914894,
|
| 422 |
+
"eval_cer": 0.9276012616915579,
|
| 423 |
+
"eval_loss": 2.8182146549224854,
|
| 424 |
+
"eval_runtime": 206.3851,
|
| 425 |
+
"eval_samples_per_second": 24.028,
|
| 426 |
+
"eval_steps_per_second": 3.004,
|
| 427 |
+
"eval_wer": 1.0,
|
| 428 |
+
"step": 3700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 10.106382978723405,
|
| 432 |
+
"eval_cer": 0.9276012616915579,
|
| 433 |
+
"eval_loss": 2.7310709953308105,
|
| 434 |
+
"eval_runtime": 207.1152,
|
| 435 |
+
"eval_samples_per_second": 23.943,
|
| 436 |
+
"eval_steps_per_second": 2.994,
|
| 437 |
+
"eval_wer": 1.0,
|
| 438 |
+
"step": 3800
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 10.372340425531915,
|
| 442 |
+
"eval_cer": 0.9276012616915579,
|
| 443 |
+
"eval_loss": 2.4996562004089355,
|
| 444 |
+
"eval_runtime": 207.3947,
|
| 445 |
+
"eval_samples_per_second": 23.911,
|
| 446 |
+
"eval_steps_per_second": 2.989,
|
| 447 |
+
"eval_wer": 1.0,
|
| 448 |
+
"step": 3900
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 10.638297872340425,
|
| 452 |
+
"grad_norm": 2.881579875946045,
|
| 453 |
+
"learning_rate": 9.5904e-06,
|
| 454 |
+
"loss": 2.6801,
|
| 455 |
+
"step": 4000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 10.638297872340425,
|
| 459 |
+
"eval_cer": 0.8951130949356452,
|
| 460 |
+
"eval_loss": 2.2397685050964355,
|
| 461 |
+
"eval_runtime": 207.9025,
|
| 462 |
+
"eval_samples_per_second": 23.853,
|
| 463 |
+
"eval_steps_per_second": 2.982,
|
| 464 |
+
"eval_wer": 1.0,
|
| 465 |
+
"step": 4000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 10.904255319148936,
|
| 469 |
+
"eval_cer": 0.6153560526408361,
|
| 470 |
+
"eval_loss": 1.9111369848251343,
|
| 471 |
+
"eval_runtime": 209.5566,
|
| 472 |
+
"eval_samples_per_second": 23.664,
|
| 473 |
+
"eval_steps_per_second": 2.959,
|
| 474 |
+
"eval_wer": 1.0,
|
| 475 |
+
"step": 4100
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 11.170212765957446,
|
| 479 |
+
"eval_cer": 0.4341266460032754,
|
| 480 |
+
"eval_loss": 1.5446799993515015,
|
| 481 |
+
"eval_runtime": 210.5656,
|
| 482 |
+
"eval_samples_per_second": 23.551,
|
| 483 |
+
"eval_steps_per_second": 2.944,
|
| 484 |
+
"eval_wer": 1.0,
|
| 485 |
+
"step": 4200
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 11.436170212765958,
|
| 489 |
+
"eval_cer": 0.3958537719809105,
|
| 490 |
+
"eval_loss": 1.3182002305984497,
|
| 491 |
+
"eval_runtime": 210.4243,
|
| 492 |
+
"eval_samples_per_second": 23.567,
|
| 493 |
+
"eval_steps_per_second": 2.946,
|
| 494 |
+
"eval_wer": 1.0,
|
| 495 |
+
"step": 4300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 11.702127659574469,
|
| 499 |
+
"eval_cer": 0.3706043064800447,
|
| 500 |
+
"eval_loss": 1.1702227592468262,
|
| 501 |
+
"eval_runtime": 210.384,
|
| 502 |
+
"eval_samples_per_second": 23.571,
|
| 503 |
+
"eval_steps_per_second": 2.947,
|
| 504 |
+
"eval_wer": 0.9995966928816293,
|
| 505 |
+
"step": 4400
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 11.96808510638298,
|
| 509 |
+
"grad_norm": 3.1133100986480713,
|
| 510 |
+
"learning_rate": 1.0790400000000001e-05,
|
| 511 |
+
"loss": 1.5214,
|
| 512 |
+
"step": 4500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 11.96808510638298,
|
| 516 |
+
"eval_cer": 0.32139911744128324,
|
| 517 |
+
"eval_loss": 1.0558491945266724,
|
| 518 |
+
"eval_runtime": 211.4587,
|
| 519 |
+
"eval_samples_per_second": 23.451,
|
| 520 |
+
"eval_steps_per_second": 2.932,
|
| 521 |
+
"eval_wer": 0.9991933857632587,
|
| 522 |
+
"step": 4500
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 12.23404255319149,
|
| 526 |
+
"eval_cer": 0.3023643036658628,
|
| 527 |
+
"eval_loss": 0.9717462658882141,
|
| 528 |
+
"eval_runtime": 211.083,
|
| 529 |
+
"eval_samples_per_second": 23.493,
|
| 530 |
+
"eval_steps_per_second": 2.937,
|
| 531 |
+
"eval_wer": 0.9987900786448881,
|
| 532 |
+
"step": 4600
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 12.5,
|
| 536 |
+
"eval_cer": 0.2873631506329955,
|
| 537 |
+
"eval_loss": 0.8958828449249268,
|
| 538 |
+
"eval_runtime": 215.6014,
|
| 539 |
+
"eval_samples_per_second": 23.001,
|
| 540 |
+
"eval_steps_per_second": 2.876,
|
| 541 |
+
"eval_wer": 0.9981851179673321,
|
| 542 |
+
"step": 4700
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 12.76595744680851,
|
| 546 |
+
"eval_cer": 0.2747306007105809,
|
| 547 |
+
"eval_loss": 0.8398593068122864,
|
| 548 |
+
"eval_runtime": 211.3172,
|
| 549 |
+
"eval_samples_per_second": 23.467,
|
| 550 |
+
"eval_steps_per_second": 2.934,
|
| 551 |
+
"eval_wer": 0.9977818108489614,
|
| 552 |
+
"step": 4800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 13.03191489361702,
|
| 556 |
+
"eval_cer": 0.2656900413137539,
|
| 557 |
+
"eval_loss": 0.7890844345092773,
|
| 558 |
+
"eval_runtime": 211.3236,
|
| 559 |
+
"eval_samples_per_second": 23.466,
|
| 560 |
+
"eval_steps_per_second": 2.934,
|
| 561 |
+
"eval_wer": 0.9973785037305909,
|
| 562 |
+
"step": 4900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 13.297872340425531,
|
| 566 |
+
"grad_norm": 2.623598098754883,
|
| 567 |
+
"learning_rate": 1.19904e-05,
|
| 568 |
+
"loss": 0.8719,
|
| 569 |
+
"step": 5000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 13.297872340425531,
|
| 573 |
+
"eval_cer": 0.25801748701372307,
|
| 574 |
+
"eval_loss": 0.7483934760093689,
|
| 575 |
+
"eval_runtime": 210.0979,
|
| 576 |
+
"eval_samples_per_second": 23.603,
|
| 577 |
+
"eval_steps_per_second": 2.951,
|
| 578 |
+
"eval_wer": 0.9979834644081468,
|
| 579 |
+
"step": 5000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 13.563829787234042,
|
| 583 |
+
"eval_cer": 0.2522914085371335,
|
| 584 |
+
"eval_loss": 0.7144805788993835,
|
| 585 |
+
"eval_runtime": 209.8962,
|
| 586 |
+
"eval_samples_per_second": 23.626,
|
| 587 |
+
"eval_steps_per_second": 2.954,
|
| 588 |
+
"eval_wer": 0.9975801572897761,
|
| 589 |
+
"step": 5100
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 13.829787234042554,
|
| 593 |
+
"eval_cer": 0.24809358718296481,
|
| 594 |
+
"eval_loss": 0.6851741075515747,
|
| 595 |
+
"eval_runtime": 209.9795,
|
| 596 |
+
"eval_samples_per_second": 23.617,
|
| 597 |
+
"eval_steps_per_second": 2.953,
|
| 598 |
+
"eval_wer": 0.9975801572897761,
|
| 599 |
+
"step": 5200
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 14.095744680851064,
|
| 603 |
+
"eval_cer": 0.24873068669947274,
|
| 604 |
+
"eval_loss": 0.6617541313171387,
|
| 605 |
+
"eval_runtime": 215.3335,
|
| 606 |
+
"eval_samples_per_second": 23.029,
|
| 607 |
+
"eval_steps_per_second": 2.879,
|
| 608 |
+
"eval_wer": 0.9979834644081468,
|
| 609 |
+
"step": 5300
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 14.361702127659575,
|
| 613 |
+
"eval_cer": 0.2476675513099626,
|
| 614 |
+
"eval_loss": 0.6399756073951721,
|
| 615 |
+
"eval_runtime": 210.9211,
|
| 616 |
+
"eval_samples_per_second": 23.511,
|
| 617 |
+
"eval_steps_per_second": 2.939,
|
| 618 |
+
"eval_wer": 0.9985884250857028,
|
| 619 |
+
"step": 5400
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 14.627659574468085,
|
| 623 |
+
"grad_norm": 3.613203763961792,
|
| 624 |
+
"learning_rate": 1.31904e-05,
|
| 625 |
+
"loss": 0.6568,
|
| 626 |
+
"step": 5500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 14.627659574468085,
|
| 630 |
+
"eval_cer": 0.2448690037405168,
|
| 631 |
+
"eval_loss": 0.6200099587440491,
|
| 632 |
+
"eval_runtime": 208.8016,
|
| 633 |
+
"eval_samples_per_second": 23.75,
|
| 634 |
+
"eval_steps_per_second": 2.969,
|
| 635 |
+
"eval_wer": 0.9987900786448881,
|
| 636 |
+
"step": 5500
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 14.893617021276595,
|
| 640 |
+
"eval_cer": 0.2420548218271076,
|
| 641 |
+
"eval_loss": 0.6031706929206848,
|
| 642 |
+
"eval_runtime": 208.8818,
|
| 643 |
+
"eval_samples_per_second": 23.741,
|
| 644 |
+
"eval_steps_per_second": 2.968,
|
| 645 |
+
"eval_wer": 0.9987900786448881,
|
| 646 |
+
"step": 5600
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 15.159574468085106,
|
| 650 |
+
"eval_cer": 0.2395025151750851,
|
| 651 |
+
"eval_loss": 0.5874983668327332,
|
| 652 |
+
"eval_runtime": 215.2428,
|
| 653 |
+
"eval_samples_per_second": 23.039,
|
| 654 |
+
"eval_steps_per_second": 2.88,
|
| 655 |
+
"eval_wer": 0.9983867715265174,
|
| 656 |
+
"step": 5700
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 15.425531914893616,
|
| 660 |
+
"eval_cer": 0.24089788037381715,
|
| 661 |
+
"eval_loss": 0.5776127576828003,
|
| 662 |
+
"eval_runtime": 208.795,
|
| 663 |
+
"eval_samples_per_second": 23.751,
|
| 664 |
+
"eval_steps_per_second": 2.969,
|
| 665 |
+
"eval_wer": 0.9989917322040734,
|
| 666 |
+
"step": 5800
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 15.691489361702128,
|
| 670 |
+
"eval_cer": 0.23602778222922294,
|
| 671 |
+
"eval_loss": 0.5617058873176575,
|
| 672 |
+
"eval_runtime": 209.2365,
|
| 673 |
+
"eval_samples_per_second": 23.7,
|
| 674 |
+
"eval_steps_per_second": 2.963,
|
| 675 |
+
"eval_wer": 0.999395039322444,
|
| 676 |
+
"step": 5900
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 15.957446808510639,
|
| 680 |
+
"grad_norm": 2.9839932918548584,
|
| 681 |
+
"learning_rate": 1.43904e-05,
|
| 682 |
+
"loss": 0.548,
|
| 683 |
+
"step": 6000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 15.957446808510639,
|
| 687 |
+
"eval_cer": 0.23474576602422542,
|
| 688 |
+
"eval_loss": 0.5485217571258545,
|
| 689 |
+
"eval_runtime": 212.04,
|
| 690 |
+
"eval_samples_per_second": 23.387,
|
| 691 |
+
"eval_steps_per_second": 2.924,
|
| 692 |
+
"eval_wer": 0.9981851179673321,
|
| 693 |
+
"step": 6000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 16.22340425531915,
|
| 697 |
+
"eval_cer": 0.23343248113130113,
|
| 698 |
+
"eval_loss": 0.5394352078437805,
|
| 699 |
+
"eval_runtime": 220.2211,
|
| 700 |
+
"eval_samples_per_second": 22.518,
|
| 701 |
+
"eval_steps_per_second": 2.815,
|
| 702 |
+
"eval_wer": 0.9987900786448881,
|
| 703 |
+
"step": 6100
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 16.48936170212766,
|
| 707 |
+
"eval_cer": 0.23171270329532884,
|
| 708 |
+
"eval_loss": 0.5321902632713318,
|
| 709 |
+
"eval_runtime": 221.3122,
|
| 710 |
+
"eval_samples_per_second": 22.407,
|
| 711 |
+
"eval_steps_per_second": 2.801,
|
| 712 |
+
"eval_wer": 0.9983867715265174,
|
| 713 |
+
"step": 6200
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 16.75531914893617,
|
| 717 |
+
"eval_cer": 0.23198630431468809,
|
| 718 |
+
"eval_loss": 0.5243075489997864,
|
| 719 |
+
"eval_runtime": 210.0529,
|
| 720 |
+
"eval_samples_per_second": 23.608,
|
| 721 |
+
"eval_steps_per_second": 2.952,
|
| 722 |
+
"eval_wer": 0.9969751966122202,
|
| 723 |
+
"step": 6300
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 17.02127659574468,
|
| 727 |
+
"eval_cer": 0.22724518950779177,
|
| 728 |
+
"eval_loss": 0.5120757222175598,
|
| 729 |
+
"eval_runtime": 207.2228,
|
| 730 |
+
"eval_samples_per_second": 23.931,
|
| 731 |
+
"eval_steps_per_second": 2.992,
|
| 732 |
+
"eval_wer": 0.9985884250857028,
|
| 733 |
+
"step": 6400
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 17.28723404255319,
|
| 737 |
+
"grad_norm": 2.14619779586792,
|
| 738 |
+
"learning_rate": 1.5590400000000002e-05,
|
| 739 |
+
"loss": 0.4681,
|
| 740 |
+
"step": 6500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 17.28723404255319,
|
| 744 |
+
"eval_cer": 0.22661981574925638,
|
| 745 |
+
"eval_loss": 0.5070440173149109,
|
| 746 |
+
"eval_runtime": 221.9297,
|
| 747 |
+
"eval_samples_per_second": 22.345,
|
| 748 |
+
"eval_steps_per_second": 2.794,
|
| 749 |
+
"eval_wer": 0.9989917322040734,
|
| 750 |
+
"step": 6500
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 17.5531914893617,
|
| 754 |
+
"eval_cer": 0.22627976876805278,
|
| 755 |
+
"eval_loss": 0.5013606548309326,
|
| 756 |
+
"eval_runtime": 207.1035,
|
| 757 |
+
"eval_samples_per_second": 23.945,
|
| 758 |
+
"eval_steps_per_second": 2.994,
|
| 759 |
+
"eval_wer": 0.9991933857632587,
|
| 760 |
+
"step": 6600
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 17.819148936170212,
|
| 764 |
+
"eval_cer": 0.2241613151610142,
|
| 765 |
+
"eval_loss": 0.4942595660686493,
|
| 766 |
+
"eval_runtime": 207.4255,
|
| 767 |
+
"eval_samples_per_second": 23.907,
|
| 768 |
+
"eval_steps_per_second": 2.989,
|
| 769 |
+
"eval_wer": 0.9985884250857028,
|
| 770 |
+
"step": 6700
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 18.085106382978722,
|
| 774 |
+
"eval_cer": 0.2228128529941723,
|
| 775 |
+
"eval_loss": 0.4929651618003845,
|
| 776 |
+
"eval_runtime": 209.5569,
|
| 777 |
+
"eval_samples_per_second": 23.664,
|
| 778 |
+
"eval_steps_per_second": 2.959,
|
| 779 |
+
"eval_wer": 0.9987900786448881,
|
| 780 |
+
"step": 6800
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 18.351063829787233,
|
| 784 |
+
"eval_cer": 0.2245013621422178,
|
| 785 |
+
"eval_loss": 0.4968652129173279,
|
| 786 |
+
"eval_runtime": 229.2892,
|
| 787 |
+
"eval_samples_per_second": 21.628,
|
| 788 |
+
"eval_steps_per_second": 2.704,
|
| 789 |
+
"eval_wer": 0.9985884250857028,
|
| 790 |
+
"step": 6900
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 18.617021276595743,
|
| 794 |
+
"grad_norm": 3.0399532318115234,
|
| 795 |
+
"learning_rate": 1.6790399999999998e-05,
|
| 796 |
+
"loss": 0.4198,
|
| 797 |
+
"step": 7000
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 18.617021276595743,
|
| 801 |
+
"eval_cer": 0.22252752621684052,
|
| 802 |
+
"eval_loss": 0.48828795552253723,
|
| 803 |
+
"eval_runtime": 211.562,
|
| 804 |
+
"eval_samples_per_second": 23.44,
|
| 805 |
+
"eval_steps_per_second": 2.931,
|
| 806 |
+
"eval_wer": 0.9985884250857028,
|
| 807 |
+
"step": 7000
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 18.882978723404257,
|
| 811 |
+
"eval_cer": 0.22148784234327548,
|
| 812 |
+
"eval_loss": 0.4804530441761017,
|
| 813 |
+
"eval_runtime": 212.7369,
|
| 814 |
+
"eval_samples_per_second": 23.31,
|
| 815 |
+
"eval_steps_per_second": 2.914,
|
| 816 |
+
"eval_wer": 0.9985884250857028,
|
| 817 |
+
"step": 7100
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 19.148936170212767,
|
| 821 |
+
"eval_cer": 0.22081556555284995,
|
| 822 |
+
"eval_loss": 0.47766780853271484,
|
| 823 |
+
"eval_runtime": 210.7245,
|
| 824 |
+
"eval_samples_per_second": 23.533,
|
| 825 |
+
"eval_steps_per_second": 2.942,
|
| 826 |
+
"eval_wer": 0.9983867715265174,
|
| 827 |
+
"step": 7200
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 19.414893617021278,
|
| 831 |
+
"eval_cer": 0.22092500596059364,
|
| 832 |
+
"eval_loss": 0.47176772356033325,
|
| 833 |
+
"eval_runtime": 209.4286,
|
| 834 |
+
"eval_samples_per_second": 23.679,
|
| 835 |
+
"eval_steps_per_second": 2.96,
|
| 836 |
+
"eval_wer": 0.9987900786448881,
|
| 837 |
+
"step": 7300
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 19.680851063829788,
|
| 841 |
+
"eval_cer": 0.21988532208702857,
|
| 842 |
+
"eval_loss": 0.47212347388267517,
|
| 843 |
+
"eval_runtime": 211.3146,
|
| 844 |
+
"eval_samples_per_second": 23.467,
|
| 845 |
+
"eval_steps_per_second": 2.934,
|
| 846 |
+
"eval_wer": 0.9983867715265174,
|
| 847 |
+
"step": 7400
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 19.9468085106383,
|
| 851 |
+
"grad_norm": 2.2681851387023926,
|
| 852 |
+
"learning_rate": 1.79904e-05,
|
| 853 |
+
"loss": 0.3795,
|
| 854 |
+
"step": 7500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 19.9468085106383,
|
| 858 |
+
"eval_cer": 0.22045206705570125,
|
| 859 |
+
"eval_loss": 0.4674960672855377,
|
| 860 |
+
"eval_runtime": 211.1631,
|
| 861 |
+
"eval_samples_per_second": 23.484,
|
| 862 |
+
"eval_steps_per_second": 2.936,
|
| 863 |
+
"eval_wer": 0.9983867715265174,
|
| 864 |
+
"step": 7500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 20.21276595744681,
|
| 868 |
+
"eval_cer": 0.21619561691166986,
|
| 869 |
+
"eval_loss": 0.46922406554222107,
|
| 870 |
+
"eval_runtime": 210.3069,
|
| 871 |
+
"eval_samples_per_second": 23.58,
|
| 872 |
+
"eval_steps_per_second": 2.948,
|
| 873 |
+
"eval_wer": 0.9987900786448881,
|
| 874 |
+
"step": 7600
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 20.47872340425532,
|
| 878 |
+
"eval_cer": 0.21727829523113423,
|
| 879 |
+
"eval_loss": 0.47320201992988586,
|
| 880 |
+
"eval_runtime": 208.0889,
|
| 881 |
+
"eval_samples_per_second": 23.831,
|
| 882 |
+
"eval_steps_per_second": 2.979,
|
| 883 |
+
"eval_wer": 0.9985884250857028,
|
| 884 |
+
"step": 7700
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 20.74468085106383,
|
| 888 |
+
"eval_cer": 0.217309563919061,
|
| 889 |
+
"eval_loss": 0.46535801887512207,
|
| 890 |
+
"eval_runtime": 208.908,
|
| 891 |
+
"eval_samples_per_second": 23.738,
|
| 892 |
+
"eval_steps_per_second": 2.968,
|
| 893 |
+
"eval_wer": 0.9981851179673321,
|
| 894 |
+
"step": 7800
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 21.01063829787234,
|
| 898 |
+
"eval_cer": 0.21583211841452118,
|
| 899 |
+
"eval_loss": 0.4556586742401123,
|
| 900 |
+
"eval_runtime": 207.993,
|
| 901 |
+
"eval_samples_per_second": 23.842,
|
| 902 |
+
"eval_steps_per_second": 2.981,
|
| 903 |
+
"eval_wer": 0.9985884250857028,
|
| 904 |
+
"step": 7900
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 21.27659574468085,
|
| 908 |
+
"grad_norm": 1.6410722732543945,
|
| 909 |
+
"learning_rate": 1.9190400000000002e-05,
|
| 910 |
+
"loss": 0.3504,
|
| 911 |
+
"step": 8000
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 21.27659574468085,
|
| 915 |
+
"eval_cer": 0.21436249008196304,
|
| 916 |
+
"eval_loss": 0.4562208950519562,
|
| 917 |
+
"eval_runtime": 209.2953,
|
| 918 |
+
"eval_samples_per_second": 23.694,
|
| 919 |
+
"eval_steps_per_second": 2.962,
|
| 920 |
+
"eval_wer": 0.9981851179673321,
|
| 921 |
+
"step": 8000
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 21.54255319148936,
|
| 925 |
+
"eval_cer": 0.2143742158399356,
|
| 926 |
+
"eval_loss": 0.4679461419582367,
|
| 927 |
+
"eval_runtime": 207.9541,
|
| 928 |
+
"eval_samples_per_second": 23.847,
|
| 929 |
+
"eval_steps_per_second": 2.981,
|
| 930 |
+
"eval_wer": 0.9981851179673321,
|
| 931 |
+
"step": 8100
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 21.80851063829787,
|
| 935 |
+
"eval_cer": 0.21689525380403132,
|
| 936 |
+
"eval_loss": 0.45842525362968445,
|
| 937 |
+
"eval_runtime": 208.594,
|
| 938 |
+
"eval_samples_per_second": 23.773,
|
| 939 |
+
"eval_steps_per_second": 2.972,
|
| 940 |
+
"eval_wer": 0.9989917322040734,
|
| 941 |
+
"step": 8200
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 22.074468085106382,
|
| 945 |
+
"eval_cer": 0.2133853435842515,
|
| 946 |
+
"eval_loss": 0.45605719089508057,
|
| 947 |
+
"eval_runtime": 209.0293,
|
| 948 |
+
"eval_samples_per_second": 23.724,
|
| 949 |
+
"eval_steps_per_second": 2.966,
|
| 950 |
+
"eval_wer": 0.9981851179673321,
|
| 951 |
+
"step": 8300
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 22.340425531914892,
|
| 955 |
+
"eval_cer": 0.21432731280804543,
|
| 956 |
+
"eval_loss": 0.45950454473495483,
|
| 957 |
+
"eval_runtime": 208.8441,
|
| 958 |
+
"eval_samples_per_second": 23.745,
|
| 959 |
+
"eval_steps_per_second": 2.969,
|
| 960 |
+
"eval_wer": 0.9987900786448881,
|
| 961 |
+
"step": 8400
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 22.606382978723403,
|
| 965 |
+
"grad_norm": 2.5605006217956543,
|
| 966 |
+
"learning_rate": 2.03904e-05,
|
| 967 |
+
"loss": 0.3134,
|
| 968 |
+
"step": 8500
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 22.606382978723403,
|
| 972 |
+
"eval_cer": 0.21549988860529926,
|
| 973 |
+
"eval_loss": 0.4543595314025879,
|
| 974 |
+
"eval_runtime": 207.7389,
|
| 975 |
+
"eval_samples_per_second": 23.871,
|
| 976 |
+
"eval_steps_per_second": 2.985,
|
| 977 |
+
"eval_wer": 0.9985884250857028,
|
| 978 |
+
"step": 8500
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 22.872340425531917,
|
| 982 |
+
"eval_cer": 0.21343615520213252,
|
| 983 |
+
"eval_loss": 0.4543818533420563,
|
| 984 |
+
"eval_runtime": 210.3188,
|
| 985 |
+
"eval_samples_per_second": 23.578,
|
| 986 |
+
"eval_steps_per_second": 2.948,
|
| 987 |
+
"eval_wer": 0.9983867715265174,
|
| 988 |
+
"step": 8600
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 23.138297872340427,
|
| 992 |
+
"eval_cer": 0.21289286174940492,
|
| 993 |
+
"eval_loss": 0.45519956946372986,
|
| 994 |
+
"eval_runtime": 210.0976,
|
| 995 |
+
"eval_samples_per_second": 23.603,
|
| 996 |
+
"eval_steps_per_second": 2.951,
|
| 997 |
+
"eval_wer": 0.9983867715265174,
|
| 998 |
+
"step": 8700
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 23.404255319148938,
|
| 1002 |
+
"eval_cer": 0.21207987586330893,
|
| 1003 |
+
"eval_loss": 0.45236727595329285,
|
| 1004 |
+
"eval_runtime": 209.5064,
|
| 1005 |
+
"eval_samples_per_second": 23.67,
|
| 1006 |
+
"eval_steps_per_second": 2.959,
|
| 1007 |
+
"eval_wer": 0.9983867715265174,
|
| 1008 |
+
"step": 8800
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 23.670212765957448,
|
| 1012 |
+
"eval_cer": 0.21132551876707564,
|
| 1013 |
+
"eval_loss": 0.45537006855010986,
|
| 1014 |
+
"eval_runtime": 209.1429,
|
| 1015 |
+
"eval_samples_per_second": 23.711,
|
| 1016 |
+
"eval_steps_per_second": 2.964,
|
| 1017 |
+
"eval_wer": 0.9985884250857028,
|
| 1018 |
+
"step": 8900
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 23.93617021276596,
|
| 1022 |
+
"grad_norm": 1.3679530620574951,
|
| 1023 |
+
"learning_rate": 2.15904e-05,
|
| 1024 |
+
"loss": 0.3014,
|
| 1025 |
+
"step": 9000
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 23.93617021276596,
|
| 1029 |
+
"eval_cer": 0.21025847479157464,
|
| 1030 |
+
"eval_loss": 0.46165505051612854,
|
| 1031 |
+
"eval_runtime": 209.5827,
|
| 1032 |
+
"eval_samples_per_second": 23.661,
|
| 1033 |
+
"eval_steps_per_second": 2.958,
|
| 1034 |
+
"eval_wer": 0.9981851179673321,
|
| 1035 |
+
"step": 9000
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 24.20212765957447,
|
| 1039 |
+
"eval_cer": 0.21297103346922183,
|
| 1040 |
+
"eval_loss": 0.4606345295906067,
|
| 1041 |
+
"eval_runtime": 209.6362,
|
| 1042 |
+
"eval_samples_per_second": 23.655,
|
| 1043 |
+
"eval_steps_per_second": 2.958,
|
| 1044 |
+
"eval_wer": 0.9977818108489614,
|
| 1045 |
+
"step": 9100
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 24.46808510638298,
|
| 1049 |
+
"eval_cer": 0.21053598439692472,
|
| 1050 |
+
"eval_loss": 0.4560868740081787,
|
| 1051 |
+
"eval_runtime": 209.0063,
|
| 1052 |
+
"eval_samples_per_second": 23.727,
|
| 1053 |
+
"eval_steps_per_second": 2.966,
|
| 1054 |
+
"eval_wer": 0.9973785037305909,
|
| 1055 |
+
"step": 9200
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 24.73404255319149,
|
| 1059 |
+
"eval_cer": 0.20888656110878767,
|
| 1060 |
+
"eval_loss": 0.45663759112358093,
|
| 1061 |
+
"eval_runtime": 211.6105,
|
| 1062 |
+
"eval_samples_per_second": 23.435,
|
| 1063 |
+
"eval_steps_per_second": 2.93,
|
| 1064 |
+
"eval_wer": 0.9983867715265174,
|
| 1065 |
+
"step": 9300
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 25.0,
|
| 1069 |
+
"eval_cer": 0.21188444656376662,
|
| 1070 |
+
"eval_loss": 0.4485580623149872,
|
| 1071 |
+
"eval_runtime": 212.8182,
|
| 1072 |
+
"eval_samples_per_second": 23.302,
|
| 1073 |
+
"eval_steps_per_second": 2.913,
|
| 1074 |
+
"eval_wer": 0.9989917322040734,
|
| 1075 |
+
"step": 9400
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 25.26595744680851,
|
| 1079 |
+
"grad_norm": 9.48050594329834,
|
| 1080 |
+
"learning_rate": 2.27904e-05,
|
| 1081 |
+
"loss": 0.2791,
|
| 1082 |
+
"step": 9500
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 25.26595744680851,
|
| 1086 |
+
"eval_cer": 0.21173982888210532,
|
| 1087 |
+
"eval_loss": 0.4541599452495575,
|
| 1088 |
+
"eval_runtime": 212.1403,
|
| 1089 |
+
"eval_samples_per_second": 23.376,
|
| 1090 |
+
"eval_steps_per_second": 2.923,
|
| 1091 |
+
"eval_wer": 0.9989917322040734,
|
| 1092 |
+
"step": 9500
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 25.53191489361702,
|
| 1096 |
+
"eval_cer": 0.20951193486732306,
|
| 1097 |
+
"eval_loss": 0.4539582133293152,
|
| 1098 |
+
"eval_runtime": 211.4226,
|
| 1099 |
+
"eval_samples_per_second": 23.455,
|
| 1100 |
+
"eval_steps_per_second": 2.933,
|
| 1101 |
+
"eval_wer": 0.9985884250857028,
|
| 1102 |
+
"step": 9600
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 25.79787234042553,
|
| 1106 |
+
"eval_cer": 0.20911325909625675,
|
| 1107 |
+
"eval_loss": 0.441910982131958,
|
| 1108 |
+
"eval_runtime": 213.904,
|
| 1109 |
+
"eval_samples_per_second": 23.183,
|
| 1110 |
+
"eval_steps_per_second": 2.898,
|
| 1111 |
+
"eval_wer": 0.9983867715265174,
|
| 1112 |
+
"step": 9700
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 26.06382978723404,
|
| 1116 |
+
"eval_cer": 0.20742474994821122,
|
| 1117 |
+
"eval_loss": 0.45685461163520813,
|
| 1118 |
+
"eval_runtime": 212.9009,
|
| 1119 |
+
"eval_samples_per_second": 23.293,
|
| 1120 |
+
"eval_steps_per_second": 2.912,
|
| 1121 |
+
"eval_wer": 0.9981851179673321,
|
| 1122 |
+
"step": 9800
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 26.329787234042552,
|
| 1126 |
+
"eval_cer": 0.20901163586049476,
|
| 1127 |
+
"eval_loss": 0.45425865054130554,
|
| 1128 |
+
"eval_runtime": 210.5454,
|
| 1129 |
+
"eval_samples_per_second": 23.553,
|
| 1130 |
+
"eval_steps_per_second": 2.945,
|
| 1131 |
+
"eval_wer": 0.9983867715265174,
|
| 1132 |
+
"step": 9900
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 26.595744680851062,
|
| 1136 |
+
"grad_norm": 2.0923683643341064,
|
| 1137 |
+
"learning_rate": 2.39904e-05,
|
| 1138 |
+
"loss": 0.2564,
|
| 1139 |
+
"step": 10000
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 26.595744680851062,
|
| 1143 |
+
"eval_cer": 0.20882402373293413,
|
| 1144 |
+
"eval_loss": 0.4689125120639801,
|
| 1145 |
+
"eval_runtime": 211.8844,
|
| 1146 |
+
"eval_samples_per_second": 23.404,
|
| 1147 |
+
"eval_steps_per_second": 2.926,
|
| 1148 |
+
"eval_wer": 0.9981851179673321,
|
| 1149 |
+
"step": 10000
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 26.861702127659573,
|
| 1153 |
+
"eval_cer": 0.20887483535081514,
|
| 1154 |
+
"eval_loss": 0.45904749631881714,
|
| 1155 |
+
"eval_runtime": 213.2034,
|
| 1156 |
+
"eval_samples_per_second": 23.259,
|
| 1157 |
+
"eval_steps_per_second": 2.908,
|
| 1158 |
+
"eval_wer": 0.9983867715265174,
|
| 1159 |
+
"step": 10100
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 27.127659574468087,
|
| 1163 |
+
"eval_cer": 0.2093282313257533,
|
| 1164 |
+
"eval_loss": 0.4985686242580414,
|
| 1165 |
+
"eval_runtime": 211.4066,
|
| 1166 |
+
"eval_samples_per_second": 23.457,
|
| 1167 |
+
"eval_steps_per_second": 2.933,
|
| 1168 |
+
"eval_wer": 0.9985884250857028,
|
| 1169 |
+
"step": 10200
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 27.393617021276597,
|
| 1173 |
+
"eval_cer": 0.20998096518622458,
|
| 1174 |
+
"eval_loss": 0.46933168172836304,
|
| 1175 |
+
"eval_runtime": 210.8303,
|
| 1176 |
+
"eval_samples_per_second": 23.521,
|
| 1177 |
+
"eval_steps_per_second": 2.941,
|
| 1178 |
+
"eval_wer": 0.9989917322040734,
|
| 1179 |
+
"step": 10300
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 27.659574468085108,
|
| 1183 |
+
"eval_cer": 0.208472250993758,
|
| 1184 |
+
"eval_loss": 0.5127565860748291,
|
| 1185 |
+
"eval_runtime": 210.3509,
|
| 1186 |
+
"eval_samples_per_second": 23.575,
|
| 1187 |
+
"eval_steps_per_second": 2.947,
|
| 1188 |
+
"eval_wer": 0.9981851179673321,
|
| 1189 |
+
"step": 10400
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 27.925531914893618,
|
| 1193 |
+
"grad_norm": 4.9774322509765625,
|
| 1194 |
+
"learning_rate": 2.51904e-05,
|
| 1195 |
+
"loss": 0.2449,
|
| 1196 |
+
"step": 10500
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 27.925531914893618,
|
| 1200 |
+
"eval_cer": 0.20994578791230697,
|
| 1201 |
+
"eval_loss": 0.4511733651161194,
|
| 1202 |
+
"eval_runtime": 213.3431,
|
| 1203 |
+
"eval_samples_per_second": 23.244,
|
| 1204 |
+
"eval_steps_per_second": 2.906,
|
| 1205 |
+
"eval_wer": 0.9983867715265174,
|
| 1206 |
+
"step": 10500
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 28.19148936170213,
|
| 1210 |
+
"eval_cer": 0.20911325909625675,
|
| 1211 |
+
"eval_loss": 0.4650718867778778,
|
| 1212 |
+
"eval_runtime": 212.7967,
|
| 1213 |
+
"eval_samples_per_second": 23.304,
|
| 1214 |
+
"eval_steps_per_second": 2.914,
|
| 1215 |
+
"eval_wer": 0.999395039322444,
|
| 1216 |
+
"step": 10600
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 28.45744680851064,
|
| 1220 |
+
"eval_cer": 0.206795467603685,
|
| 1221 |
+
"eval_loss": 0.46036314964294434,
|
| 1222 |
+
"eval_runtime": 211.0991,
|
| 1223 |
+
"eval_samples_per_second": 23.491,
|
| 1224 |
+
"eval_steps_per_second": 2.937,
|
| 1225 |
+
"eval_wer": 0.9983867715265174,
|
| 1226 |
+
"step": 10700
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 28.72340425531915,
|
| 1230 |
+
"eval_cer": 0.20801885501881984,
|
| 1231 |
+
"eval_loss": 0.46872854232788086,
|
| 1232 |
+
"eval_runtime": 211.1242,
|
| 1233 |
+
"eval_samples_per_second": 23.489,
|
| 1234 |
+
"eval_steps_per_second": 2.937,
|
| 1235 |
+
"eval_wer": 0.9989917322040734,
|
| 1236 |
+
"step": 10800
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 28.98936170212766,
|
| 1240 |
+
"eval_cer": 0.20636943173068278,
|
| 1241 |
+
"eval_loss": 0.468840092420578,
|
| 1242 |
+
"eval_runtime": 212.878,
|
| 1243 |
+
"eval_samples_per_second": 23.295,
|
| 1244 |
+
"eval_steps_per_second": 2.912,
|
| 1245 |
+
"eval_wer": 0.999395039322444,
|
| 1246 |
+
"step": 10900
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 29.25531914893617,
|
| 1250 |
+
"grad_norm": 4.162810802459717,
|
| 1251 |
+
"learning_rate": 2.63904e-05,
|
| 1252 |
+
"loss": 0.2258,
|
| 1253 |
+
"step": 11000
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 29.25531914893617,
|
| 1257 |
+
"eval_cer": 0.20923051667598214,
|
| 1258 |
+
"eval_loss": 0.4758950173854828,
|
| 1259 |
+
"eval_runtime": 212.8392,
|
| 1260 |
+
"eval_samples_per_second": 23.299,
|
| 1261 |
+
"eval_steps_per_second": 2.913,
|
| 1262 |
+
"eval_wer": 0.999395039322444,
|
| 1263 |
+
"step": 11000
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 29.52127659574468,
|
| 1267 |
+
"eval_cer": 0.20683064487760264,
|
| 1268 |
+
"eval_loss": 0.4816320836544037,
|
| 1269 |
+
"eval_runtime": 213.093,
|
| 1270 |
+
"eval_samples_per_second": 23.272,
|
| 1271 |
+
"eval_steps_per_second": 2.91,
|
| 1272 |
+
"eval_wer": 0.9987900786448881,
|
| 1273 |
+
"step": 11100
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 29.78723404255319,
|
| 1277 |
+
"eval_cer": 0.20534147361509028,
|
| 1278 |
+
"eval_loss": 0.4749976098537445,
|
| 1279 |
+
"eval_runtime": 212.0731,
|
| 1280 |
+
"eval_samples_per_second": 23.383,
|
| 1281 |
+
"eval_steps_per_second": 2.924,
|
| 1282 |
+
"eval_wer": 0.9987900786448881,
|
| 1283 |
+
"step": 11200
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 30.0531914893617,
|
| 1287 |
+
"eval_cer": 0.2047864544043901,
|
| 1288 |
+
"eval_loss": 0.47532856464385986,
|
| 1289 |
+
"eval_runtime": 213.0921,
|
| 1290 |
+
"eval_samples_per_second": 23.272,
|
| 1291 |
+
"eval_steps_per_second": 2.91,
|
| 1292 |
+
"eval_wer": 0.9985884250857028,
|
| 1293 |
+
"step": 11300
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 30.319148936170212,
|
| 1297 |
+
"eval_cer": 0.2059551216156531,
|
| 1298 |
+
"eval_loss": 0.48292842507362366,
|
| 1299 |
+
"eval_runtime": 212.9065,
|
| 1300 |
+
"eval_samples_per_second": 23.292,
|
| 1301 |
+
"eval_steps_per_second": 2.912,
|
| 1302 |
+
"eval_wer": 0.9991933857632587,
|
| 1303 |
+
"step": 11400
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 30.585106382978722,
|
| 1307 |
+
"grad_norm": 2.1135425567626953,
|
| 1308 |
+
"learning_rate": 2.7590400000000003e-05,
|
| 1309 |
+
"loss": 0.2124,
|
| 1310 |
+
"step": 11500
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 30.585106382978722,
|
| 1314 |
+
"eval_cer": 0.20811266108260015,
|
| 1315 |
+
"eval_loss": 0.4800090193748474,
|
| 1316 |
+
"eval_runtime": 215.1825,
|
| 1317 |
+
"eval_samples_per_second": 23.046,
|
| 1318 |
+
"eval_steps_per_second": 2.881,
|
| 1319 |
+
"eval_wer": 0.9985884250857028,
|
| 1320 |
+
"step": 11500
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 30.851063829787233,
|
| 1324 |
+
"eval_cer": 0.20605283626542426,
|
| 1325 |
+
"eval_loss": 0.5290209054946899,
|
| 1326 |
+
"eval_runtime": 213.4725,
|
| 1327 |
+
"eval_samples_per_second": 23.23,
|
| 1328 |
+
"eval_steps_per_second": 2.904,
|
| 1329 |
+
"eval_wer": 0.9989917322040734,
|
| 1330 |
+
"step": 11600
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 31.117021276595743,
|
| 1334 |
+
"eval_cer": 0.20547045695278818,
|
| 1335 |
+
"eval_loss": 0.5368531942367554,
|
| 1336 |
+
"eval_runtime": 214.9789,
|
| 1337 |
+
"eval_samples_per_second": 23.067,
|
| 1338 |
+
"eval_steps_per_second": 2.884,
|
| 1339 |
+
"eval_wer": 0.9987900786448881,
|
| 1340 |
+
"step": 11700
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 31.382978723404257,
|
| 1344 |
+
"eval_cer": 0.2040555488241019,
|
| 1345 |
+
"eval_loss": 0.5169620513916016,
|
| 1346 |
+
"eval_runtime": 214.7111,
|
| 1347 |
+
"eval_samples_per_second": 23.096,
|
| 1348 |
+
"eval_steps_per_second": 2.888,
|
| 1349 |
+
"eval_wer": 0.9977818108489614,
|
| 1350 |
+
"step": 11800
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 31.648936170212767,
|
| 1354 |
+
"eval_cer": 0.20704170852110831,
|
| 1355 |
+
"eval_loss": 0.5228501558303833,
|
| 1356 |
+
"eval_runtime": 216.5107,
|
| 1357 |
+
"eval_samples_per_second": 22.904,
|
| 1358 |
+
"eval_steps_per_second": 2.864,
|
| 1359 |
+
"eval_wer": 0.9989917322040734,
|
| 1360 |
+
"step": 11900
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 31.914893617021278,
|
| 1364 |
+
"grad_norm": 4.3140645027160645,
|
| 1365 |
+
"learning_rate": 2.87904e-05,
|
| 1366 |
+
"loss": 0.2007,
|
| 1367 |
+
"step": 12000
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 31.914893617021278,
|
| 1371 |
+
"eval_cer": 0.20597857313159817,
|
| 1372 |
+
"eval_loss": 0.5034534931182861,
|
| 1373 |
+
"eval_runtime": 215.0115,
|
| 1374 |
+
"eval_samples_per_second": 23.064,
|
| 1375 |
+
"eval_steps_per_second": 2.884,
|
| 1376 |
+
"eval_wer": 0.9985884250857028,
|
| 1377 |
+
"step": 12000
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 32.180851063829785,
|
| 1381 |
+
"eval_cer": 0.20488807764015213,
|
| 1382 |
+
"eval_loss": 0.5103185772895813,
|
| 1383 |
+
"eval_runtime": 214.9574,
|
| 1384 |
+
"eval_samples_per_second": 23.07,
|
| 1385 |
+
"eval_steps_per_second": 2.884,
|
| 1386 |
+
"eval_wer": 0.9973785037305909,
|
| 1387 |
+
"step": 12100
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 32.4468085106383,
|
| 1391 |
+
"eval_cer": 0.2032073856640883,
|
| 1392 |
+
"eval_loss": 0.48684972524642944,
|
| 1393 |
+
"eval_runtime": 215.7924,
|
| 1394 |
+
"eval_samples_per_second": 22.98,
|
| 1395 |
+
"eval_steps_per_second": 2.873,
|
| 1396 |
+
"eval_wer": 0.9971768501714056,
|
| 1397 |
+
"step": 12200
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 32.712765957446805,
|
| 1401 |
+
"eval_cer": 0.2042666124676076,
|
| 1402 |
+
"eval_loss": 0.4866673946380615,
|
| 1403 |
+
"eval_runtime": 215.5038,
|
| 1404 |
+
"eval_samples_per_second": 23.011,
|
| 1405 |
+
"eval_steps_per_second": 2.877,
|
| 1406 |
+
"eval_wer": 0.9995966928816293,
|
| 1407 |
+
"step": 12300
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 32.97872340425532,
|
| 1411 |
+
"eval_cer": 0.20395001700234905,
|
| 1412 |
+
"eval_loss": 0.5048738718032837,
|
| 1413 |
+
"eval_runtime": 220.2273,
|
| 1414 |
+
"eval_samples_per_second": 22.518,
|
| 1415 |
+
"eval_steps_per_second": 2.815,
|
| 1416 |
+
"eval_wer": 0.9981851179673321,
|
| 1417 |
+
"step": 12400
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 33.244680851063826,
|
| 1421 |
+
"grad_norm": 3.023064613342285,
|
| 1422 |
+
"learning_rate": 2.99904e-05,
|
| 1423 |
+
"loss": 0.1867,
|
| 1424 |
+
"step": 12500
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 33.244680851063826,
|
| 1428 |
+
"eval_cer": 0.20403600589414767,
|
| 1429 |
+
"eval_loss": 0.5125642418861389,
|
| 1430 |
+
"eval_runtime": 217.1156,
|
| 1431 |
+
"eval_samples_per_second": 22.84,
|
| 1432 |
+
"eval_steps_per_second": 2.856,
|
| 1433 |
+
"eval_wer": 0.9983867715265174,
|
| 1434 |
+
"step": 12500
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 33.51063829787234,
|
| 1438 |
+
"eval_cer": 0.20373895335884337,
|
| 1439 |
+
"eval_loss": 0.5321324467658997,
|
| 1440 |
+
"eval_runtime": 218.7742,
|
| 1441 |
+
"eval_samples_per_second": 22.667,
|
| 1442 |
+
"eval_steps_per_second": 2.834,
|
| 1443 |
+
"eval_wer": 0.9991933857632587,
|
| 1444 |
+
"step": 12600
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 33.776595744680854,
|
| 1448 |
+
"eval_cer": 0.20402428013617513,
|
| 1449 |
+
"eval_loss": 0.5186927914619446,
|
| 1450 |
+
"eval_runtime": 216.188,
|
| 1451 |
+
"eval_samples_per_second": 22.938,
|
| 1452 |
+
"eval_steps_per_second": 2.868,
|
| 1453 |
+
"eval_wer": 0.9977818108489614,
|
| 1454 |
+
"step": 12700
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 34.04255319148936,
|
| 1458 |
+
"eval_cer": 0.20636943173068278,
|
| 1459 |
+
"eval_loss": 0.5318589806556702,
|
| 1460 |
+
"eval_runtime": 217.8562,
|
| 1461 |
+
"eval_samples_per_second": 22.763,
|
| 1462 |
+
"eval_steps_per_second": 2.846,
|
| 1463 |
+
"eval_wer": 0.9989917322040734,
|
| 1464 |
+
"step": 12800
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 34.308510638297875,
|
| 1468 |
+
"eval_cer": 0.2040711831680653,
|
| 1469 |
+
"eval_loss": 0.5274991989135742,
|
| 1470 |
+
"eval_runtime": 218.7876,
|
| 1471 |
+
"eval_samples_per_second": 22.666,
|
| 1472 |
+
"eval_steps_per_second": 2.834,
|
| 1473 |
+
"eval_wer": 0.9979834644081468,
|
| 1474 |
+
"step": 12900
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 34.57446808510638,
|
| 1478 |
+
"grad_norm": 4.690342426300049,
|
| 1479 |
+
"learning_rate": 2.7264773627685888e-05,
|
| 1480 |
+
"loss": 0.1749,
|
| 1481 |
+
"step": 13000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 34.57446808510638,
|
| 1485 |
+
"eval_cer": 0.20429788115553435,
|
| 1486 |
+
"eval_loss": 0.543329656124115,
|
| 1487 |
+
"eval_runtime": 215.6846,
|
| 1488 |
+
"eval_samples_per_second": 22.992,
|
| 1489 |
+
"eval_steps_per_second": 2.875,
|
| 1490 |
+
"eval_wer": 0.9981851179673321,
|
| 1491 |
+
"step": 13000
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 34.840425531914896,
|
| 1495 |
+
"eval_cer": 0.2023240452301571,
|
| 1496 |
+
"eval_loss": 0.5093534588813782,
|
| 1497 |
+
"eval_runtime": 216.6521,
|
| 1498 |
+
"eval_samples_per_second": 22.889,
|
| 1499 |
+
"eval_steps_per_second": 2.862,
|
| 1500 |
+
"eval_wer": 0.9983867715265174,
|
| 1501 |
+
"step": 13100
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 35.1063829787234,
|
| 1505 |
+
"eval_cer": 0.20042838102459673,
|
| 1506 |
+
"eval_loss": 0.5363013744354248,
|
| 1507 |
+
"eval_runtime": 216.3501,
|
| 1508 |
+
"eval_samples_per_second": 22.921,
|
| 1509 |
+
"eval_steps_per_second": 2.866,
|
| 1510 |
+
"eval_wer": 0.9989917322040734,
|
| 1511 |
+
"step": 13200
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 35.37234042553192,
|
| 1515 |
+
"eval_cer": 0.20222633058038594,
|
| 1516 |
+
"eval_loss": 0.5330685377120972,
|
| 1517 |
+
"eval_runtime": 215.7086,
|
| 1518 |
+
"eval_samples_per_second": 22.989,
|
| 1519 |
+
"eval_steps_per_second": 2.874,
|
| 1520 |
+
"eval_wer": 0.999395039322444,
|
| 1521 |
+
"step": 13300
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 35.638297872340424,
|
| 1525 |
+
"eval_cer": 0.20093258861741586,
|
| 1526 |
+
"eval_loss": 0.5052955746650696,
|
| 1527 |
+
"eval_runtime": 213.6416,
|
| 1528 |
+
"eval_samples_per_second": 23.212,
|
| 1529 |
+
"eval_steps_per_second": 2.902,
|
| 1530 |
+
"eval_wer": 0.9989917322040734,
|
| 1531 |
+
"step": 13400
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 35.90425531914894,
|
| 1535 |
+
"grad_norm": 3.094069242477417,
|
| 1536 |
+
"learning_rate": 1.9986694366624682e-05,
|
| 1537 |
+
"loss": 0.1604,
|
| 1538 |
+
"step": 13500
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 35.90425531914894,
|
| 1542 |
+
"eval_cer": 0.2026210977654614,
|
| 1543 |
+
"eval_loss": 0.5156988501548767,
|
| 1544 |
+
"eval_runtime": 221.1798,
|
| 1545 |
+
"eval_samples_per_second": 22.421,
|
| 1546 |
+
"eval_steps_per_second": 2.803,
|
| 1547 |
+
"eval_wer": 0.9989917322040734,
|
| 1548 |
+
"step": 13500
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 36.170212765957444,
|
| 1552 |
+
"eval_cer": 0.20183938056729217,
|
| 1553 |
+
"eval_loss": 0.5299074649810791,
|
| 1554 |
+
"eval_runtime": 215.0871,
|
| 1555 |
+
"eval_samples_per_second": 23.056,
|
| 1556 |
+
"eval_steps_per_second": 2.883,
|
| 1557 |
+
"eval_wer": 0.9989917322040734,
|
| 1558 |
+
"step": 13600
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 36.43617021276596,
|
| 1562 |
+
"eval_cer": 0.20501706097785005,
|
| 1563 |
+
"eval_loss": 0.5117177963256836,
|
| 1564 |
+
"eval_runtime": 214.4648,
|
| 1565 |
+
"eval_samples_per_second": 23.123,
|
| 1566 |
+
"eval_steps_per_second": 2.891,
|
| 1567 |
+
"eval_wer": 0.9995966928816293,
|
| 1568 |
+
"step": 13700
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 36.702127659574465,
|
| 1572 |
+
"eval_cer": 0.2038327594226237,
|
| 1573 |
+
"eval_loss": 0.5067043900489807,
|
| 1574 |
+
"eval_runtime": 213.8511,
|
| 1575 |
+
"eval_samples_per_second": 23.189,
|
| 1576 |
+
"eval_steps_per_second": 2.899,
|
| 1577 |
+
"eval_wer": 0.999395039322444,
|
| 1578 |
+
"step": 13800
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 36.96808510638298,
|
| 1582 |
+
"eval_cer": 0.20278525837707692,
|
| 1583 |
+
"eval_loss": 0.4994255602359772,
|
| 1584 |
+
"eval_runtime": 216.1066,
|
| 1585 |
+
"eval_samples_per_second": 22.947,
|
| 1586 |
+
"eval_steps_per_second": 2.869,
|
| 1587 |
+
"eval_wer": 0.9995966928816293,
|
| 1588 |
+
"step": 13900
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 37.234042553191486,
|
| 1592 |
+
"grad_norm": 2.801684617996216,
|
| 1593 |
+
"learning_rate": 1.089715454232877e-05,
|
| 1594 |
+
"loss": 0.1412,
|
| 1595 |
+
"step": 14000
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 37.234042553191486,
|
| 1599 |
+
"eval_cer": 0.20241394270794655,
|
| 1600 |
+
"eval_loss": 0.5346146821975708,
|
| 1601 |
+
"eval_runtime": 215.4888,
|
| 1602 |
+
"eval_samples_per_second": 23.013,
|
| 1603 |
+
"eval_steps_per_second": 2.877,
|
| 1604 |
+
"eval_wer": 0.9983867715265174,
|
| 1605 |
+
"step": 14000
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 37.5,
|
| 1609 |
+
"eval_cer": 0.20152669368802448,
|
| 1610 |
+
"eval_loss": 0.5349961519241333,
|
| 1611 |
+
"eval_runtime": 213.0582,
|
| 1612 |
+
"eval_samples_per_second": 23.275,
|
| 1613 |
+
"eval_steps_per_second": 2.91,
|
| 1614 |
+
"eval_wer": 0.999395039322444,
|
| 1615 |
+
"step": 14100
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 37.765957446808514,
|
| 1619 |
+
"eval_cer": 0.20097949164930604,
|
| 1620 |
+
"eval_loss": 0.5236727595329285,
|
| 1621 |
+
"eval_runtime": 211.9375,
|
| 1622 |
+
"eval_samples_per_second": 23.398,
|
| 1623 |
+
"eval_steps_per_second": 2.925,
|
| 1624 |
+
"eval_wer": 0.9989917322040734,
|
| 1625 |
+
"step": 14200
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 38.03191489361702,
|
| 1629 |
+
"eval_cer": 0.19928316532927884,
|
| 1630 |
+
"eval_loss": 0.5304979085922241,
|
| 1631 |
+
"eval_runtime": 212.7349,
|
| 1632 |
+
"eval_samples_per_second": 23.311,
|
| 1633 |
+
"eval_steps_per_second": 2.914,
|
| 1634 |
+
"eval_wer": 0.9991933857632587,
|
| 1635 |
+
"step": 14300
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 38.297872340425535,
|
| 1639 |
+
"eval_cer": 0.19734059809182833,
|
| 1640 |
+
"eval_loss": 0.5309327244758606,
|
| 1641 |
+
"eval_runtime": 214.8976,
|
| 1642 |
+
"eval_samples_per_second": 23.076,
|
| 1643 |
+
"eval_steps_per_second": 2.885,
|
| 1644 |
+
"eval_wer": 0.9985884250857028,
|
| 1645 |
+
"step": 14400
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 38.56382978723404,
|
| 1649 |
+
"grad_norm": 1.7099072933197021,
|
| 1650 |
+
"learning_rate": 3.292383984744315e-06,
|
| 1651 |
+
"loss": 0.1286,
|
| 1652 |
+
"step": 14500
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 38.56382978723404,
|
| 1656 |
+
"eval_cer": 0.19924407946937037,
|
| 1657 |
+
"eval_loss": 0.5269559621810913,
|
| 1658 |
+
"eval_runtime": 214.5389,
|
| 1659 |
+
"eval_samples_per_second": 23.115,
|
| 1660 |
+
"eval_steps_per_second": 2.89,
|
| 1661 |
+
"eval_wer": 0.9991933857632587,
|
| 1662 |
+
"step": 14500
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 38.829787234042556,
|
| 1666 |
+
"eval_cer": 0.19992026484578673,
|
| 1667 |
+
"eval_loss": 0.5362734198570251,
|
| 1668 |
+
"eval_runtime": 221.7091,
|
| 1669 |
+
"eval_samples_per_second": 22.367,
|
| 1670 |
+
"eval_steps_per_second": 2.796,
|
| 1671 |
+
"eval_wer": 0.9989917322040734,
|
| 1672 |
+
"step": 14600
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 39.09574468085106,
|
| 1676 |
+
"eval_cer": 0.19989681332984166,
|
| 1677 |
+
"eval_loss": 0.5347363352775574,
|
| 1678 |
+
"eval_runtime": 213.4871,
|
| 1679 |
+
"eval_samples_per_second": 23.229,
|
| 1680 |
+
"eval_steps_per_second": 2.904,
|
| 1681 |
+
"eval_wer": 0.9989917322040734,
|
| 1682 |
+
"step": 14700
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 39.361702127659576,
|
| 1686 |
+
"eval_cer": 0.19991244767380506,
|
| 1687 |
+
"eval_loss": 0.5318508744239807,
|
| 1688 |
+
"eval_runtime": 214.7174,
|
| 1689 |
+
"eval_samples_per_second": 23.095,
|
| 1690 |
+
"eval_steps_per_second": 2.888,
|
| 1691 |
+
"eval_wer": 0.9989917322040734,
|
| 1692 |
+
"step": 14800
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 39.62765957446808,
|
| 1696 |
+
"eval_cer": 0.1994590516988669,
|
| 1697 |
+
"eval_loss": 0.5322226881980896,
|
| 1698 |
+
"eval_runtime": 213.8602,
|
| 1699 |
+
"eval_samples_per_second": 23.188,
|
| 1700 |
+
"eval_steps_per_second": 2.899,
|
| 1701 |
+
"eval_wer": 0.999395039322444,
|
| 1702 |
+
"step": 14900
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 39.8936170212766,
|
| 1706 |
+
"grad_norm": 2.6511354446411133,
|
| 1707 |
+
"learning_rate": 2.4271246633469913e-08,
|
| 1708 |
+
"loss": 0.1217,
|
| 1709 |
+
"step": 15000
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 39.8936170212766,
|
| 1713 |
+
"eval_cer": 0.19924017088337953,
|
| 1714 |
+
"eval_loss": 0.5321789383888245,
|
| 1715 |
+
"eval_runtime": 215.7329,
|
| 1716 |
+
"eval_samples_per_second": 22.987,
|
| 1717 |
+
"eval_steps_per_second": 2.874,
|
| 1718 |
+
"eval_wer": 0.9991933857632587,
|
| 1719 |
+
"step": 15000
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 40.0,
|
| 1723 |
+
"step": 15040,
|
| 1724 |
+
"total_flos": 2.1018408189762245e+19,
|
| 1725 |
+
"train_loss": 1.5314149432993949,
|
| 1726 |
+
"train_runtime": 82631.5291,
|
| 1727 |
+
"train_samples_per_second": 5.824,
|
| 1728 |
+
"train_steps_per_second": 0.182
|
| 1729 |
+
}
|
| 1730 |
+
],
|
| 1731 |
+
"logging_steps": 500,
|
| 1732 |
+
"max_steps": 15040,
|
| 1733 |
+
"num_input_tokens_seen": 0,
|
| 1734 |
+
"num_train_epochs": 40,
|
| 1735 |
+
"save_steps": 400,
|
| 1736 |
+
"stateful_callbacks": {
|
| 1737 |
+
"TrainerControl": {
|
| 1738 |
+
"args": {
|
| 1739 |
+
"should_epoch_stop": false,
|
| 1740 |
+
"should_evaluate": false,
|
| 1741 |
+
"should_log": false,
|
| 1742 |
+
"should_save": true,
|
| 1743 |
+
"should_training_stop": true
|
| 1744 |
+
},
|
| 1745 |
+
"attributes": {}
|
| 1746 |
+
}
|
| 1747 |
+
},
|
| 1748 |
+
"total_flos": 2.1018408189762245e+19,
|
| 1749 |
+
"train_batch_size": 16,
|
| 1750 |
+
"trial_name": null,
|
| 1751 |
+
"trial_params": null
|
| 1752 |
+
}
|