End of training
Browse files- README.md +9 -5
- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +897 -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
|
|
@@ -15,11 +19,11 @@ 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
|
|
@@ -31,11 +35,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 31 |
|
| 32 |
# Hubert-common_voice-phoneme-ctc_zero_infinity
|
| 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: 1.0
|
| 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
|
|
|
|
| 35 |
|
| 36 |
# Hubert-common_voice-phoneme-ctc_zero_infinity
|
| 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: 0.5230
|
| 41 |
- Wer: 1.0
|
| 42 |
+
- Cer: 0.1953
|
| 43 |
|
| 44 |
## Model description
|
| 45 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_cer": 0.19530663519242536,
|
| 4 |
+
"eval_loss": 0.5230427980422974,
|
| 5 |
+
"eval_runtime": 200.2116,
|
| 6 |
+
"eval_samples": 4961,
|
| 7 |
+
"eval_samples_per_second": 24.779,
|
| 8 |
+
"eval_steps_per_second": 3.102,
|
| 9 |
+
"eval_wer": 1.0,
|
| 10 |
+
"total_flos": 1.051128494332674e+19,
|
| 11 |
+
"train_loss": 1.2840657353401184,
|
| 12 |
+
"train_runtime": 37969.1145,
|
| 13 |
+
"train_samples": 12032,
|
| 14 |
+
"train_samples_per_second": 6.338,
|
| 15 |
+
"train_steps_per_second": 0.198
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_cer": 0.19530663519242536,
|
| 4 |
+
"eval_loss": 0.5230427980422974,
|
| 5 |
+
"eval_runtime": 200.2116,
|
| 6 |
+
"eval_samples": 4961,
|
| 7 |
+
"eval_samples_per_second": 24.779,
|
| 8 |
+
"eval_steps_per_second": 3.102,
|
| 9 |
+
"eval_wer": 1.0
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"total_flos": 1.051128494332674e+19,
|
| 4 |
+
"train_loss": 1.2840657353401184,
|
| 5 |
+
"train_runtime": 37969.1145,
|
| 6 |
+
"train_samples": 12032,
|
| 7 |
+
"train_samples_per_second": 6.338,
|
| 8 |
+
"train_steps_per_second": 0.198
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,897 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3724757134914398,
|
| 3 |
+
"best_model_checkpoint": "./Hubert-common_voice-phoneme-ctc_zero_infinity/checkpoint-4500",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 7520,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.26595744680851063,
|
| 13 |
+
"eval_cer": 1.81529422485279,
|
| 14 |
+
"eval_loss": 18.23810386657715,
|
| 15 |
+
"eval_runtime": 203.0311,
|
| 16 |
+
"eval_samples_per_second": 24.435,
|
| 17 |
+
"eval_steps_per_second": 3.059,
|
| 18 |
+
"eval_wer": 1.1471477524692602,
|
| 19 |
+
"step": 100
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.5319148936170213,
|
| 23 |
+
"eval_cer": 0.9816795615591899,
|
| 24 |
+
"eval_loss": 8.172632217407227,
|
| 25 |
+
"eval_runtime": 200.7558,
|
| 26 |
+
"eval_samples_per_second": 24.712,
|
| 27 |
+
"eval_steps_per_second": 3.093,
|
| 28 |
+
"eval_wer": 1.0,
|
| 29 |
+
"step": 200
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.7978723404255319,
|
| 33 |
+
"eval_cer": 0.9816708830000044,
|
| 34 |
+
"eval_loss": 6.93860387802124,
|
| 35 |
+
"eval_runtime": 199.4086,
|
| 36 |
+
"eval_samples_per_second": 24.879,
|
| 37 |
+
"eval_steps_per_second": 3.114,
|
| 38 |
+
"eval_wer": 1.0,
|
| 39 |
+
"step": 300
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0638297872340425,
|
| 43 |
+
"eval_cer": 0.9816795615591899,
|
| 44 |
+
"eval_loss": 6.238890171051025,
|
| 45 |
+
"eval_runtime": 199.5101,
|
| 46 |
+
"eval_samples_per_second": 24.866,
|
| 47 |
+
"eval_steps_per_second": 3.113,
|
| 48 |
+
"eval_wer": 1.0,
|
| 49 |
+
"step": 400
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 1.3297872340425532,
|
| 53 |
+
"grad_norm": 28.442243576049805,
|
| 54 |
+
"learning_rate": 1.1903999999999998e-05,
|
| 55 |
+
"loss": 8.8178,
|
| 56 |
+
"step": 500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.3297872340425532,
|
| 60 |
+
"eval_cer": 0.9816752222795971,
|
| 61 |
+
"eval_loss": 5.465348243713379,
|
| 62 |
+
"eval_runtime": 199.0401,
|
| 63 |
+
"eval_samples_per_second": 24.925,
|
| 64 |
+
"eval_steps_per_second": 3.12,
|
| 65 |
+
"eval_wer": 1.0,
|
| 66 |
+
"step": 500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.5957446808510638,
|
| 70 |
+
"eval_cer": 0.9816708830000044,
|
| 71 |
+
"eval_loss": 4.674495697021484,
|
| 72 |
+
"eval_runtime": 199.3585,
|
| 73 |
+
"eval_samples_per_second": 24.885,
|
| 74 |
+
"eval_steps_per_second": 3.115,
|
| 75 |
+
"eval_wer": 1.0,
|
| 76 |
+
"step": 600
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 1.8617021276595744,
|
| 80 |
+
"eval_cer": 0.9816622044408188,
|
| 81 |
+
"eval_loss": 3.9770586490631104,
|
| 82 |
+
"eval_runtime": 201.0574,
|
| 83 |
+
"eval_samples_per_second": 24.675,
|
| 84 |
+
"eval_steps_per_second": 3.089,
|
| 85 |
+
"eval_wer": 1.0,
|
| 86 |
+
"step": 700
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.127659574468085,
|
| 90 |
+
"eval_cer": 0.9816752222795971,
|
| 91 |
+
"eval_loss": 3.4578745365142822,
|
| 92 |
+
"eval_runtime": 200.5713,
|
| 93 |
+
"eval_samples_per_second": 24.734,
|
| 94 |
+
"eval_steps_per_second": 3.096,
|
| 95 |
+
"eval_wer": 1.0,
|
| 96 |
+
"step": 800
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 2.393617021276596,
|
| 100 |
+
"eval_cer": 0.9816795615591899,
|
| 101 |
+
"eval_loss": 3.174529552459717,
|
| 102 |
+
"eval_runtime": 200.8264,
|
| 103 |
+
"eval_samples_per_second": 24.703,
|
| 104 |
+
"eval_steps_per_second": 3.092,
|
| 105 |
+
"eval_wer": 1.0,
|
| 106 |
+
"step": 900
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 2.6595744680851063,
|
| 110 |
+
"grad_norm": 1.5112452507019043,
|
| 111 |
+
"learning_rate": 2.3903999999999997e-05,
|
| 112 |
+
"loss": 3.6858,
|
| 113 |
+
"step": 1000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.6595744680851063,
|
| 117 |
+
"eval_cer": 0.9816839008387828,
|
| 118 |
+
"eval_loss": 3.067502975463867,
|
| 119 |
+
"eval_runtime": 197.554,
|
| 120 |
+
"eval_samples_per_second": 25.112,
|
| 121 |
+
"eval_steps_per_second": 3.143,
|
| 122 |
+
"eval_wer": 1.0,
|
| 123 |
+
"step": 1000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.925531914893617,
|
| 127 |
+
"eval_cer": 0.9816795615591899,
|
| 128 |
+
"eval_loss": 3.0343146324157715,
|
| 129 |
+
"eval_runtime": 199.0478,
|
| 130 |
+
"eval_samples_per_second": 24.924,
|
| 131 |
+
"eval_steps_per_second": 3.12,
|
| 132 |
+
"eval_wer": 1.0,
|
| 133 |
+
"step": 1100
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 3.1914893617021276,
|
| 137 |
+
"eval_cer": 0.9816795615591899,
|
| 138 |
+
"eval_loss": 3.0101518630981445,
|
| 139 |
+
"eval_runtime": 204.5797,
|
| 140 |
+
"eval_samples_per_second": 24.25,
|
| 141 |
+
"eval_steps_per_second": 3.035,
|
| 142 |
+
"eval_wer": 1.0,
|
| 143 |
+
"step": 1200
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.4574468085106385,
|
| 147 |
+
"eval_cer": 0.9816882401183755,
|
| 148 |
+
"eval_loss": 2.992486000061035,
|
| 149 |
+
"eval_runtime": 198.4427,
|
| 150 |
+
"eval_samples_per_second": 25.0,
|
| 151 |
+
"eval_steps_per_second": 3.129,
|
| 152 |
+
"eval_wer": 1.0,
|
| 153 |
+
"step": 1300
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 3.723404255319149,
|
| 157 |
+
"eval_cer": 0.9366551965042763,
|
| 158 |
+
"eval_loss": 2.5595009326934814,
|
| 159 |
+
"eval_runtime": 208.4363,
|
| 160 |
+
"eval_samples_per_second": 23.801,
|
| 161 |
+
"eval_steps_per_second": 2.979,
|
| 162 |
+
"eval_wer": 1.0,
|
| 163 |
+
"step": 1400
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 3.9893617021276597,
|
| 167 |
+
"grad_norm": 3.490267515182495,
|
| 168 |
+
"learning_rate": 3.5903999999999994e-05,
|
| 169 |
+
"loss": 2.7891,
|
| 170 |
+
"step": 1500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.9893617021276597,
|
| 174 |
+
"eval_cer": 0.3741760792873167,
|
| 175 |
+
"eval_loss": 1.5431591272354126,
|
| 176 |
+
"eval_runtime": 201.614,
|
| 177 |
+
"eval_samples_per_second": 24.606,
|
| 178 |
+
"eval_steps_per_second": 3.08,
|
| 179 |
+
"eval_wer": 1.0,
|
| 180 |
+
"step": 1500
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 4.25531914893617,
|
| 184 |
+
"eval_cer": 0.2971755629130452,
|
| 185 |
+
"eval_loss": 1.0799192190170288,
|
| 186 |
+
"eval_runtime": 201.0045,
|
| 187 |
+
"eval_samples_per_second": 24.681,
|
| 188 |
+
"eval_steps_per_second": 3.089,
|
| 189 |
+
"eval_wer": 1.0,
|
| 190 |
+
"step": 1600
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 4.5212765957446805,
|
| 194 |
+
"eval_cer": 0.2638846098770682,
|
| 195 |
+
"eval_loss": 0.8669940829277039,
|
| 196 |
+
"eval_runtime": 202.8717,
|
| 197 |
+
"eval_samples_per_second": 24.454,
|
| 198 |
+
"eval_steps_per_second": 3.061,
|
| 199 |
+
"eval_wer": 1.0,
|
| 200 |
+
"step": 1700
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.787234042553192,
|
| 204 |
+
"eval_cer": 0.2558742997487557,
|
| 205 |
+
"eval_loss": 0.7350400686264038,
|
| 206 |
+
"eval_runtime": 202.705,
|
| 207 |
+
"eval_samples_per_second": 24.474,
|
| 208 |
+
"eval_steps_per_second": 3.064,
|
| 209 |
+
"eval_wer": 1.0,
|
| 210 |
+
"step": 1800
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.053191489361702,
|
| 214 |
+
"eval_cer": 0.24679652684061393,
|
| 215 |
+
"eval_loss": 0.6753000617027283,
|
| 216 |
+
"eval_runtime": 206.0047,
|
| 217 |
+
"eval_samples_per_second": 24.082,
|
| 218 |
+
"eval_steps_per_second": 3.014,
|
| 219 |
+
"eval_wer": 1.0,
|
| 220 |
+
"step": 1900
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 5.319148936170213,
|
| 224 |
+
"grad_norm": 2.341259717941284,
|
| 225 |
+
"learning_rate": 4.7903999999999994e-05,
|
| 226 |
+
"loss": 0.9179,
|
| 227 |
+
"step": 2000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 5.319148936170213,
|
| 231 |
+
"eval_cer": 0.23886866302456466,
|
| 232 |
+
"eval_loss": 0.6170761585235596,
|
| 233 |
+
"eval_runtime": 207.3253,
|
| 234 |
+
"eval_samples_per_second": 23.929,
|
| 235 |
+
"eval_steps_per_second": 2.995,
|
| 236 |
+
"eval_wer": 1.0,
|
| 237 |
+
"step": 2000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 5.585106382978723,
|
| 241 |
+
"eval_cer": 0.23860830624899654,
|
| 242 |
+
"eval_loss": 0.5866126418113708,
|
| 243 |
+
"eval_runtime": 206.9405,
|
| 244 |
+
"eval_samples_per_second": 23.973,
|
| 245 |
+
"eval_steps_per_second": 3.001,
|
| 246 |
+
"eval_wer": 1.0,
|
| 247 |
+
"step": 2100
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 5.851063829787234,
|
| 251 |
+
"eval_cer": 0.2389033772613071,
|
| 252 |
+
"eval_loss": 0.5648585557937622,
|
| 253 |
+
"eval_runtime": 207.0413,
|
| 254 |
+
"eval_samples_per_second": 23.961,
|
| 255 |
+
"eval_steps_per_second": 2.999,
|
| 256 |
+
"eval_wer": 1.0,
|
| 257 |
+
"step": 2200
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 6.117021276595745,
|
| 261 |
+
"eval_cer": 0.23212976181694314,
|
| 262 |
+
"eval_loss": 0.5367811918258667,
|
| 263 |
+
"eval_runtime": 209.1071,
|
| 264 |
+
"eval_samples_per_second": 23.725,
|
| 265 |
+
"eval_steps_per_second": 2.97,
|
| 266 |
+
"eval_wer": 1.0,
|
| 267 |
+
"step": 2300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 6.382978723404255,
|
| 271 |
+
"eval_cer": 0.22887530212234164,
|
| 272 |
+
"eval_loss": 0.5225070118904114,
|
| 273 |
+
"eval_runtime": 207.8275,
|
| 274 |
+
"eval_samples_per_second": 23.871,
|
| 275 |
+
"eval_steps_per_second": 2.988,
|
| 276 |
+
"eval_wer": 1.0,
|
| 277 |
+
"step": 2400
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 6.648936170212766,
|
| 281 |
+
"grad_norm": 3.6773109436035156,
|
| 282 |
+
"learning_rate": 5.9903999999999994e-05,
|
| 283 |
+
"loss": 0.563,
|
| 284 |
+
"step": 2500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 6.648936170212766,
|
| 288 |
+
"eval_cer": 0.22932224792040026,
|
| 289 |
+
"eval_loss": 0.5041821002960205,
|
| 290 |
+
"eval_runtime": 208.2744,
|
| 291 |
+
"eval_samples_per_second": 23.82,
|
| 292 |
+
"eval_steps_per_second": 2.982,
|
| 293 |
+
"eval_wer": 1.0,
|
| 294 |
+
"step": 2500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 6.914893617021277,
|
| 298 |
+
"eval_cer": 0.22474864722958696,
|
| 299 |
+
"eval_loss": 0.4917772710323334,
|
| 300 |
+
"eval_runtime": 207.9185,
|
| 301 |
+
"eval_samples_per_second": 23.86,
|
| 302 |
+
"eval_steps_per_second": 2.987,
|
| 303 |
+
"eval_wer": 1.0,
|
| 304 |
+
"step": 2600
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 7.180851063829787,
|
| 308 |
+
"eval_cer": 0.2207608492838019,
|
| 309 |
+
"eval_loss": 0.48810550570487976,
|
| 310 |
+
"eval_runtime": 212.2308,
|
| 311 |
+
"eval_samples_per_second": 23.375,
|
| 312 |
+
"eval_steps_per_second": 2.926,
|
| 313 |
+
"eval_wer": 1.0,
|
| 314 |
+
"step": 2700
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 7.446808510638298,
|
| 318 |
+
"eval_cer": 0.21984526128972068,
|
| 319 |
+
"eval_loss": 0.47866183519363403,
|
| 320 |
+
"eval_runtime": 208.3994,
|
| 321 |
+
"eval_samples_per_second": 23.805,
|
| 322 |
+
"eval_steps_per_second": 2.98,
|
| 323 |
+
"eval_wer": 1.0,
|
| 324 |
+
"step": 2800
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 7.712765957446808,
|
| 328 |
+
"eval_cer": 0.21810521017300707,
|
| 329 |
+
"eval_loss": 0.4691566526889801,
|
| 330 |
+
"eval_runtime": 209.2233,
|
| 331 |
+
"eval_samples_per_second": 23.712,
|
| 332 |
+
"eval_steps_per_second": 2.968,
|
| 333 |
+
"eval_wer": 1.0,
|
| 334 |
+
"step": 2900
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 7.9787234042553195,
|
| 338 |
+
"grad_norm": 4.193769454956055,
|
| 339 |
+
"learning_rate": 7.1904e-05,
|
| 340 |
+
"loss": 0.4453,
|
| 341 |
+
"step": 3000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 7.9787234042553195,
|
| 345 |
+
"eval_cer": 0.21508507157641688,
|
| 346 |
+
"eval_loss": 0.4733445346355438,
|
| 347 |
+
"eval_runtime": 209.7678,
|
| 348 |
+
"eval_samples_per_second": 23.65,
|
| 349 |
+
"eval_steps_per_second": 2.96,
|
| 350 |
+
"eval_wer": 1.0,
|
| 351 |
+
"step": 3000
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 8.24468085106383,
|
| 355 |
+
"eval_cer": 0.2146988756926575,
|
| 356 |
+
"eval_loss": 0.4585338830947876,
|
| 357 |
+
"eval_runtime": 209.6674,
|
| 358 |
+
"eval_samples_per_second": 23.661,
|
| 359 |
+
"eval_steps_per_second": 2.962,
|
| 360 |
+
"eval_wer": 1.0,
|
| 361 |
+
"step": 3100
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 8.51063829787234,
|
| 365 |
+
"eval_cer": 0.21156157654706165,
|
| 366 |
+
"eval_loss": 0.446315199136734,
|
| 367 |
+
"eval_runtime": 206.8959,
|
| 368 |
+
"eval_samples_per_second": 23.978,
|
| 369 |
+
"eval_steps_per_second": 3.002,
|
| 370 |
+
"eval_wer": 1.0,
|
| 371 |
+
"step": 3200
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 8.77659574468085,
|
| 375 |
+
"eval_cer": 0.20546054943958203,
|
| 376 |
+
"eval_loss": 0.41826608777046204,
|
| 377 |
+
"eval_runtime": 207.5757,
|
| 378 |
+
"eval_samples_per_second": 23.9,
|
| 379 |
+
"eval_steps_per_second": 2.992,
|
| 380 |
+
"eval_wer": 1.0,
|
| 381 |
+
"step": 3300
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 9.042553191489361,
|
| 385 |
+
"eval_cer": 0.2031954454921394,
|
| 386 |
+
"eval_loss": 0.43083709478378296,
|
| 387 |
+
"eval_runtime": 205.6417,
|
| 388 |
+
"eval_samples_per_second": 24.124,
|
| 389 |
+
"eval_steps_per_second": 3.02,
|
| 390 |
+
"eval_wer": 0.9997984277363435,
|
| 391 |
+
"step": 3400
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 9.308510638297872,
|
| 395 |
+
"grad_norm": 2.33461332321167,
|
| 396 |
+
"learning_rate": 8.390399999999999e-05,
|
| 397 |
+
"loss": 0.3596,
|
| 398 |
+
"step": 3500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 9.308510638297872,
|
| 402 |
+
"eval_cer": 0.20223212542253735,
|
| 403 |
+
"eval_loss": 0.4070371687412262,
|
| 404 |
+
"eval_runtime": 201.8613,
|
| 405 |
+
"eval_samples_per_second": 24.576,
|
| 406 |
+
"eval_steps_per_second": 3.076,
|
| 407 |
+
"eval_wer": 1.0,
|
| 408 |
+
"step": 3500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 9.574468085106384,
|
| 412 |
+
"eval_cer": 0.20241871444502785,
|
| 413 |
+
"eval_loss": 0.42586851119995117,
|
| 414 |
+
"eval_runtime": 200.832,
|
| 415 |
+
"eval_samples_per_second": 24.702,
|
| 416 |
+
"eval_steps_per_second": 3.092,
|
| 417 |
+
"eval_wer": 1.0,
|
| 418 |
+
"step": 3600
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 9.840425531914894,
|
| 422 |
+
"eval_cer": 0.198469970015578,
|
| 423 |
+
"eval_loss": 0.40376514196395874,
|
| 424 |
+
"eval_runtime": 200.0148,
|
| 425 |
+
"eval_samples_per_second": 24.803,
|
| 426 |
+
"eval_steps_per_second": 3.105,
|
| 427 |
+
"eval_wer": 1.0,
|
| 428 |
+
"step": 3700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 10.106382978723405,
|
| 432 |
+
"eval_cer": 0.19764116761335282,
|
| 433 |
+
"eval_loss": 0.4271981716156006,
|
| 434 |
+
"eval_runtime": 199.797,
|
| 435 |
+
"eval_samples_per_second": 24.83,
|
| 436 |
+
"eval_steps_per_second": 3.108,
|
| 437 |
+
"eval_wer": 1.0,
|
| 438 |
+
"step": 3800
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 10.372340425531915,
|
| 442 |
+
"eval_cer": 0.19692084720094769,
|
| 443 |
+
"eval_loss": 0.39613601565361023,
|
| 444 |
+
"eval_runtime": 199.8886,
|
| 445 |
+
"eval_samples_per_second": 24.819,
|
| 446 |
+
"eval_steps_per_second": 3.107,
|
| 447 |
+
"eval_wer": 0.9997984277363435,
|
| 448 |
+
"step": 3900
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 10.638297872340425,
|
| 452 |
+
"grad_norm": 3.095649480819702,
|
| 453 |
+
"learning_rate": 9.5904e-05,
|
| 454 |
+
"loss": 0.2945,
|
| 455 |
+
"step": 4000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 10.638297872340425,
|
| 459 |
+
"eval_cer": 0.1943172794452665,
|
| 460 |
+
"eval_loss": 0.4180251955986023,
|
| 461 |
+
"eval_runtime": 200.3742,
|
| 462 |
+
"eval_samples_per_second": 24.759,
|
| 463 |
+
"eval_steps_per_second": 3.099,
|
| 464 |
+
"eval_wer": 1.0,
|
| 465 |
+
"step": 4000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 10.904255319148936,
|
| 469 |
+
"eval_cer": 0.19749363210719756,
|
| 470 |
+
"eval_loss": 0.39988985657691956,
|
| 471 |
+
"eval_runtime": 199.7299,
|
| 472 |
+
"eval_samples_per_second": 24.839,
|
| 473 |
+
"eval_steps_per_second": 3.109,
|
| 474 |
+
"eval_wer": 1.0,
|
| 475 |
+
"step": 4100
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 11.170212765957446,
|
| 479 |
+
"eval_cer": 0.1930458705245755,
|
| 480 |
+
"eval_loss": 0.38789525628089905,
|
| 481 |
+
"eval_runtime": 199.5558,
|
| 482 |
+
"eval_samples_per_second": 24.86,
|
| 483 |
+
"eval_steps_per_second": 3.112,
|
| 484 |
+
"eval_wer": 1.0,
|
| 485 |
+
"step": 4200
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 11.436170212765958,
|
| 489 |
+
"eval_cer": 0.19177880088347732,
|
| 490 |
+
"eval_loss": 0.3799389600753784,
|
| 491 |
+
"eval_runtime": 199.3837,
|
| 492 |
+
"eval_samples_per_second": 24.882,
|
| 493 |
+
"eval_steps_per_second": 3.115,
|
| 494 |
+
"eval_wer": 1.0,
|
| 495 |
+
"step": 4300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 11.702127659574469,
|
| 499 |
+
"eval_cer": 0.19271608527552256,
|
| 500 |
+
"eval_loss": 0.37643763422966003,
|
| 501 |
+
"eval_runtime": 200.1267,
|
| 502 |
+
"eval_samples_per_second": 24.789,
|
| 503 |
+
"eval_steps_per_second": 3.103,
|
| 504 |
+
"eval_wer": 0.9997984277363435,
|
| 505 |
+
"step": 4400
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 11.96808510638298,
|
| 509 |
+
"grad_norm": 3.6549434661865234,
|
| 510 |
+
"learning_rate": 0.00010790399999999999,
|
| 511 |
+
"loss": 0.2605,
|
| 512 |
+
"step": 4500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 11.96808510638298,
|
| 516 |
+
"eval_cer": 0.19186992575492617,
|
| 517 |
+
"eval_loss": 0.3724757134914398,
|
| 518 |
+
"eval_runtime": 200.9023,
|
| 519 |
+
"eval_samples_per_second": 24.694,
|
| 520 |
+
"eval_steps_per_second": 3.091,
|
| 521 |
+
"eval_wer": 1.0,
|
| 522 |
+
"step": 4500
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 12.23404255319149,
|
| 526 |
+
"eval_cer": 0.19185690791614776,
|
| 527 |
+
"eval_loss": 0.39096784591674805,
|
| 528 |
+
"eval_runtime": 200.1369,
|
| 529 |
+
"eval_samples_per_second": 24.788,
|
| 530 |
+
"eval_steps_per_second": 3.103,
|
| 531 |
+
"eval_wer": 1.0,
|
| 532 |
+
"step": 4600
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 12.5,
|
| 536 |
+
"eval_cer": 0.19075473089957606,
|
| 537 |
+
"eval_loss": 0.38514742255210876,
|
| 538 |
+
"eval_runtime": 200.1929,
|
| 539 |
+
"eval_samples_per_second": 24.781,
|
| 540 |
+
"eval_steps_per_second": 3.102,
|
| 541 |
+
"eval_wer": 0.999596855472687,
|
| 542 |
+
"step": 4700
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 12.76595744680851,
|
| 546 |
+
"eval_cer": 0.19056380259749275,
|
| 547 |
+
"eval_loss": 0.41147398948669434,
|
| 548 |
+
"eval_runtime": 200.2674,
|
| 549 |
+
"eval_samples_per_second": 24.772,
|
| 550 |
+
"eval_steps_per_second": 3.101,
|
| 551 |
+
"eval_wer": 1.0,
|
| 552 |
+
"step": 4800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 13.03191489361702,
|
| 556 |
+
"eval_cer": 0.18938785782784343,
|
| 557 |
+
"eval_loss": 0.3779467046260834,
|
| 558 |
+
"eval_runtime": 200.1481,
|
| 559 |
+
"eval_samples_per_second": 24.787,
|
| 560 |
+
"eval_steps_per_second": 3.103,
|
| 561 |
+
"eval_wer": 1.0,
|
| 562 |
+
"step": 4900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 13.297872340425531,
|
| 566 |
+
"grad_norm": 1.6606886386871338,
|
| 567 |
+
"learning_rate": 0.00011990399999999998,
|
| 568 |
+
"loss": 0.2223,
|
| 569 |
+
"step": 5000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 13.297872340425531,
|
| 573 |
+
"eval_cer": 0.1904336242097087,
|
| 574 |
+
"eval_loss": 0.3956213593482971,
|
| 575 |
+
"eval_runtime": 200.0105,
|
| 576 |
+
"eval_samples_per_second": 24.804,
|
| 577 |
+
"eval_steps_per_second": 3.105,
|
| 578 |
+
"eval_wer": 1.0,
|
| 579 |
+
"step": 5000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 13.563829787234042,
|
| 583 |
+
"eval_cer": 0.1906809631464984,
|
| 584 |
+
"eval_loss": 0.4001040756702423,
|
| 585 |
+
"eval_runtime": 199.9024,
|
| 586 |
+
"eval_samples_per_second": 24.817,
|
| 587 |
+
"eval_steps_per_second": 3.107,
|
| 588 |
+
"eval_wer": 1.0,
|
| 589 |
+
"step": 5100
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 13.829787234042554,
|
| 593 |
+
"eval_cer": 0.19482931443721713,
|
| 594 |
+
"eval_loss": 0.389057993888855,
|
| 595 |
+
"eval_runtime": 199.8841,
|
| 596 |
+
"eval_samples_per_second": 24.819,
|
| 597 |
+
"eval_steps_per_second": 3.107,
|
| 598 |
+
"eval_wer": 1.0,
|
| 599 |
+
"step": 5200
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 14.095744680851064,
|
| 603 |
+
"eval_cer": 0.19015157103617658,
|
| 604 |
+
"eval_loss": 0.39397361874580383,
|
| 605 |
+
"eval_runtime": 199.9667,
|
| 606 |
+
"eval_samples_per_second": 24.809,
|
| 607 |
+
"eval_steps_per_second": 3.106,
|
| 608 |
+
"eval_wer": 1.0,
|
| 609 |
+
"step": 5300
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 14.361702127659575,
|
| 613 |
+
"eval_cer": 0.19093698064247375,
|
| 614 |
+
"eval_loss": 0.40559881925582886,
|
| 615 |
+
"eval_runtime": 199.788,
|
| 616 |
+
"eval_samples_per_second": 24.831,
|
| 617 |
+
"eval_steps_per_second": 3.108,
|
| 618 |
+
"eval_wer": 1.0,
|
| 619 |
+
"step": 5400
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 14.627659574468085,
|
| 623 |
+
"grad_norm": 2.7307333946228027,
|
| 624 |
+
"learning_rate": 0.000131904,
|
| 625 |
+
"loss": 0.211,
|
| 626 |
+
"step": 5500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 14.627659574468085,
|
| 630 |
+
"eval_cer": 0.19288965645923464,
|
| 631 |
+
"eval_loss": 0.39999279379844666,
|
| 632 |
+
"eval_runtime": 199.9369,
|
| 633 |
+
"eval_samples_per_second": 24.813,
|
| 634 |
+
"eval_steps_per_second": 3.106,
|
| 635 |
+
"eval_wer": 0.9997984277363435,
|
| 636 |
+
"step": 5500
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 14.893617021276595,
|
| 640 |
+
"eval_cer": 0.18948332197888507,
|
| 641 |
+
"eval_loss": 0.3925553262233734,
|
| 642 |
+
"eval_runtime": 199.978,
|
| 643 |
+
"eval_samples_per_second": 24.808,
|
| 644 |
+
"eval_steps_per_second": 3.105,
|
| 645 |
+
"eval_wer": 1.0,
|
| 646 |
+
"step": 5600
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 15.159574468085106,
|
| 650 |
+
"eval_cer": 0.19304153124498272,
|
| 651 |
+
"eval_loss": 0.3852066397666931,
|
| 652 |
+
"eval_runtime": 200.0971,
|
| 653 |
+
"eval_samples_per_second": 24.793,
|
| 654 |
+
"eval_steps_per_second": 3.103,
|
| 655 |
+
"eval_wer": 0.9997984277363435,
|
| 656 |
+
"step": 5700
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 15.425531914893616,
|
| 660 |
+
"eval_cer": 0.18859376966236066,
|
| 661 |
+
"eval_loss": 0.3863610327243805,
|
| 662 |
+
"eval_runtime": 200.0265,
|
| 663 |
+
"eval_samples_per_second": 24.802,
|
| 664 |
+
"eval_steps_per_second": 3.105,
|
| 665 |
+
"eval_wer": 1.0,
|
| 666 |
+
"step": 5800
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 15.691489361702128,
|
| 670 |
+
"eval_cer": 0.19094131992206653,
|
| 671 |
+
"eval_loss": 0.3950639069080353,
|
| 672 |
+
"eval_runtime": 199.4054,
|
| 673 |
+
"eval_samples_per_second": 24.879,
|
| 674 |
+
"eval_steps_per_second": 3.114,
|
| 675 |
+
"eval_wer": 0.9997984277363435,
|
| 676 |
+
"step": 5900
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 15.957446808510639,
|
| 680 |
+
"grad_norm": 1.9801617860794067,
|
| 681 |
+
"learning_rate": 0.00014390399999999998,
|
| 682 |
+
"loss": 0.1983,
|
| 683 |
+
"step": 6000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 15.957446808510639,
|
| 687 |
+
"eval_cer": 0.18818153810104446,
|
| 688 |
+
"eval_loss": 0.39505481719970703,
|
| 689 |
+
"eval_runtime": 199.6961,
|
| 690 |
+
"eval_samples_per_second": 24.843,
|
| 691 |
+
"eval_steps_per_second": 3.11,
|
| 692 |
+
"eval_wer": 1.0,
|
| 693 |
+
"step": 6000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 16.22340425531915,
|
| 697 |
+
"eval_cer": 0.19177012232429172,
|
| 698 |
+
"eval_loss": 0.4086914360523224,
|
| 699 |
+
"eval_runtime": 199.9546,
|
| 700 |
+
"eval_samples_per_second": 24.811,
|
| 701 |
+
"eval_steps_per_second": 3.106,
|
| 702 |
+
"eval_wer": 1.0,
|
| 703 |
+
"step": 6100
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 16.48936170212766,
|
| 707 |
+
"eval_cer": 0.1891188224930897,
|
| 708 |
+
"eval_loss": 0.41504043340682983,
|
| 709 |
+
"eval_runtime": 200.5207,
|
| 710 |
+
"eval_samples_per_second": 24.741,
|
| 711 |
+
"eval_steps_per_second": 3.097,
|
| 712 |
+
"eval_wer": 1.0,
|
| 713 |
+
"step": 6200
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 16.75531914893617,
|
| 717 |
+
"eval_cer": 0.19071133810364804,
|
| 718 |
+
"eval_loss": 0.40079623460769653,
|
| 719 |
+
"eval_runtime": 199.571,
|
| 720 |
+
"eval_samples_per_second": 24.858,
|
| 721 |
+
"eval_steps_per_second": 3.112,
|
| 722 |
+
"eval_wer": 0.9997984277363435,
|
| 723 |
+
"step": 6300
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 17.02127659574468,
|
| 727 |
+
"eval_cer": 0.1943129401656737,
|
| 728 |
+
"eval_loss": 0.42198607325553894,
|
| 729 |
+
"eval_runtime": 199.6397,
|
| 730 |
+
"eval_samples_per_second": 24.85,
|
| 731 |
+
"eval_steps_per_second": 3.111,
|
| 732 |
+
"eval_wer": 1.0,
|
| 733 |
+
"step": 6400
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 17.28723404255319,
|
| 737 |
+
"grad_norm": 2.3918137550354004,
|
| 738 |
+
"learning_rate": 0.000155904,
|
| 739 |
+
"loss": 0.1829,
|
| 740 |
+
"step": 6500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 17.28723404255319,
|
| 744 |
+
"eval_cer": 0.19249912129588245,
|
| 745 |
+
"eval_loss": 0.4154324531555176,
|
| 746 |
+
"eval_runtime": 207.13,
|
| 747 |
+
"eval_samples_per_second": 23.951,
|
| 748 |
+
"eval_steps_per_second": 2.998,
|
| 749 |
+
"eval_wer": 1.0,
|
| 750 |
+
"step": 6500
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 17.5531914893617,
|
| 754 |
+
"eval_cer": 0.19589677721704643,
|
| 755 |
+
"eval_loss": 0.448236882686615,
|
| 756 |
+
"eval_runtime": 199.769,
|
| 757 |
+
"eval_samples_per_second": 24.834,
|
| 758 |
+
"eval_steps_per_second": 3.109,
|
| 759 |
+
"eval_wer": 1.0,
|
| 760 |
+
"step": 6600
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 17.819148936170212,
|
| 764 |
+
"eval_cer": 0.19394844067987832,
|
| 765 |
+
"eval_loss": 0.421720415353775,
|
| 766 |
+
"eval_runtime": 199.7025,
|
| 767 |
+
"eval_samples_per_second": 24.842,
|
| 768 |
+
"eval_steps_per_second": 3.11,
|
| 769 |
+
"eval_wer": 0.9997984277363435,
|
| 770 |
+
"step": 6700
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 18.085106382978722,
|
| 774 |
+
"eval_cer": 0.19163560465691487,
|
| 775 |
+
"eval_loss": 0.43830403685569763,
|
| 776 |
+
"eval_runtime": 198.9477,
|
| 777 |
+
"eval_samples_per_second": 24.936,
|
| 778 |
+
"eval_steps_per_second": 3.121,
|
| 779 |
+
"eval_wer": 0.9997984277363435,
|
| 780 |
+
"step": 6800
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 18.351063829787233,
|
| 784 |
+
"eval_cer": 0.1925902461673313,
|
| 785 |
+
"eval_loss": 0.42257052659988403,
|
| 786 |
+
"eval_runtime": 198.9428,
|
| 787 |
+
"eval_samples_per_second": 24.937,
|
| 788 |
+
"eval_steps_per_second": 3.122,
|
| 789 |
+
"eval_wer": 1.0,
|
| 790 |
+
"step": 6900
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 18.617021276595743,
|
| 794 |
+
"grad_norm": 2.7391552925109863,
|
| 795 |
+
"learning_rate": 0.00016790399999999997,
|
| 796 |
+
"loss": 0.1757,
|
| 797 |
+
"step": 7000
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 18.617021276595743,
|
| 801 |
+
"eval_cer": 0.19164428321610047,
|
| 802 |
+
"eval_loss": 0.41704750061035156,
|
| 803 |
+
"eval_runtime": 205.8394,
|
| 804 |
+
"eval_samples_per_second": 24.101,
|
| 805 |
+
"eval_steps_per_second": 3.017,
|
| 806 |
+
"eval_wer": 0.9997984277363435,
|
| 807 |
+
"step": 7000
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 18.882978723404257,
|
| 811 |
+
"eval_cer": 0.19180483656103414,
|
| 812 |
+
"eval_loss": 0.4161529242992401,
|
| 813 |
+
"eval_runtime": 199.3453,
|
| 814 |
+
"eval_samples_per_second": 24.886,
|
| 815 |
+
"eval_steps_per_second": 3.115,
|
| 816 |
+
"eval_wer": 1.0,
|
| 817 |
+
"step": 7100
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 19.148936170212767,
|
| 821 |
+
"eval_cer": 0.19098905199758737,
|
| 822 |
+
"eval_loss": 0.43501901626586914,
|
| 823 |
+
"eval_runtime": 199.3511,
|
| 824 |
+
"eval_samples_per_second": 24.886,
|
| 825 |
+
"eval_steps_per_second": 3.115,
|
| 826 |
+
"eval_wer": 0.9997984277363435,
|
| 827 |
+
"step": 7200
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 19.414893617021278,
|
| 831 |
+
"eval_cer": 0.20221476830416615,
|
| 832 |
+
"eval_loss": 0.4403385818004608,
|
| 833 |
+
"eval_runtime": 199.9937,
|
| 834 |
+
"eval_samples_per_second": 24.806,
|
| 835 |
+
"eval_steps_per_second": 3.105,
|
| 836 |
+
"eval_wer": 1.0,
|
| 837 |
+
"step": 7300
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 19.680851063829788,
|
| 841 |
+
"eval_cer": 0.19435199368200892,
|
| 842 |
+
"eval_loss": 0.4325474500656128,
|
| 843 |
+
"eval_runtime": 199.2227,
|
| 844 |
+
"eval_samples_per_second": 24.902,
|
| 845 |
+
"eval_steps_per_second": 3.117,
|
| 846 |
+
"eval_wer": 0.9997984277363435,
|
| 847 |
+
"step": 7400
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 19.9468085106383,
|
| 851 |
+
"grad_norm": 2.526855707168579,
|
| 852 |
+
"learning_rate": 0.000179904,
|
| 853 |
+
"loss": 0.1801,
|
| 854 |
+
"step": 7500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 19.9468085106383,
|
| 858 |
+
"eval_cer": 0.19768889968887365,
|
| 859 |
+
"eval_loss": 0.5487878918647766,
|
| 860 |
+
"eval_runtime": 199.4843,
|
| 861 |
+
"eval_samples_per_second": 24.869,
|
| 862 |
+
"eval_steps_per_second": 3.113,
|
| 863 |
+
"eval_wer": 1.0,
|
| 864 |
+
"step": 7500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 20.0,
|
| 868 |
+
"step": 7520,
|
| 869 |
+
"total_flos": 1.051128494332674e+19,
|
| 870 |
+
"train_loss": 1.2840657353401184,
|
| 871 |
+
"train_runtime": 37969.1145,
|
| 872 |
+
"train_samples_per_second": 6.338,
|
| 873 |
+
"train_steps_per_second": 0.198
|
| 874 |
+
}
|
| 875 |
+
],
|
| 876 |
+
"logging_steps": 500,
|
| 877 |
+
"max_steps": 7520,
|
| 878 |
+
"num_input_tokens_seen": 0,
|
| 879 |
+
"num_train_epochs": 20,
|
| 880 |
+
"save_steps": 400,
|
| 881 |
+
"stateful_callbacks": {
|
| 882 |
+
"TrainerControl": {
|
| 883 |
+
"args": {
|
| 884 |
+
"should_epoch_stop": false,
|
| 885 |
+
"should_evaluate": false,
|
| 886 |
+
"should_log": false,
|
| 887 |
+
"should_save": true,
|
| 888 |
+
"should_training_stop": true
|
| 889 |
+
},
|
| 890 |
+
"attributes": {}
|
| 891 |
+
}
|
| 892 |
+
},
|
| 893 |
+
"total_flos": 1.051128494332674e+19,
|
| 894 |
+
"train_batch_size": 16,
|
| 895 |
+
"trial_name": null,
|
| 896 |
+
"trial_params": null
|
| 897 |
+
}
|