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 +679 -0
README.md
CHANGED
|
@@ -1,8 +1,12 @@
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
|
|
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: facebook/wav2vec2-large-xlsr-53
|
| 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 |
# wav2vec2-large-xlsr-53-common_voice-ja-demo-kana-only
|
| 33 |
|
| 34 |
-
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
- Wer: 0.9998
|
| 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: facebook/wav2vec2-large-xlsr-53
|
| 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 |
# wav2vec2-large-xlsr-53-common_voice-ja-demo-kana-only
|
| 37 |
|
| 38 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the MOZILLA-FOUNDATION/COMMON_VOICE_13_0 - JA dataset.
|
| 39 |
It achieves the following results on the evaluation set:
|
| 40 |
+
- Loss: 0.6985
|
| 41 |
- Wer: 0.9998
|
| 42 |
+
- Cer: 0.3126
|
| 43 |
|
| 44 |
## Model description
|
| 45 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_cer": 0.3125918271792305,
|
| 4 |
+
"eval_loss": 0.6985270977020264,
|
| 5 |
+
"eval_runtime": 238.2595,
|
| 6 |
+
"eval_samples": 4961,
|
| 7 |
+
"eval_samples_per_second": 20.822,
|
| 8 |
+
"eval_steps_per_second": 2.606,
|
| 9 |
+
"eval_wer": 0.9997984277363435,
|
| 10 |
+
"total_flos": 2.6348448154740388e+19,
|
| 11 |
+
"train_loss": 0.9431738045198698,
|
| 12 |
+
"train_runtime": 36296.3273,
|
| 13 |
+
"train_samples": 12032,
|
| 14 |
+
"train_samples_per_second": 4.972,
|
| 15 |
+
"train_steps_per_second": 0.155
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_cer": 0.3125918271792305,
|
| 4 |
+
"eval_loss": 0.6985270977020264,
|
| 5 |
+
"eval_runtime": 238.2595,
|
| 6 |
+
"eval_samples": 4961,
|
| 7 |
+
"eval_samples_per_second": 20.822,
|
| 8 |
+
"eval_steps_per_second": 2.606,
|
| 9 |
+
"eval_wer": 0.9997984277363435
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"total_flos": 2.6348448154740388e+19,
|
| 4 |
+
"train_loss": 0.9431738045198698,
|
| 5 |
+
"train_runtime": 36296.3273,
|
| 6 |
+
"train_samples": 12032,
|
| 7 |
+
"train_samples_per_second": 4.972,
|
| 8 |
+
"train_steps_per_second": 0.155
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,679 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6012922525405884,
|
| 3 |
+
"best_model_checkpoint": "./wav2vec2-large-xlsr-53-common_voice-ja-demo-kana-only/checkpoint-2600",
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 5640,
|
| 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": 0.9998627637760951,
|
| 14 |
+
"eval_loss": 6.830887317657471,
|
| 15 |
+
"eval_runtime": 235.9339,
|
| 16 |
+
"eval_samples_per_second": 21.027,
|
| 17 |
+
"eval_steps_per_second": 2.632,
|
| 18 |
+
"eval_wer": 1.0,
|
| 19 |
+
"step": 100
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.5319148936170213,
|
| 23 |
+
"eval_cer": 0.9998627637760951,
|
| 24 |
+
"eval_loss": 4.129940986633301,
|
| 25 |
+
"eval_runtime": 235.1597,
|
| 26 |
+
"eval_samples_per_second": 21.096,
|
| 27 |
+
"eval_steps_per_second": 2.641,
|
| 28 |
+
"eval_wer": 1.0,
|
| 29 |
+
"step": 200
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.7978723404255319,
|
| 33 |
+
"eval_cer": 0.9869302678528182,
|
| 34 |
+
"eval_loss": 3.993030309677124,
|
| 35 |
+
"eval_runtime": 234.1893,
|
| 36 |
+
"eval_samples_per_second": 21.184,
|
| 37 |
+
"eval_steps_per_second": 2.652,
|
| 38 |
+
"eval_wer": 1.0,
|
| 39 |
+
"step": 300
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0638297872340425,
|
| 43 |
+
"eval_cer": 0.5876132198847216,
|
| 44 |
+
"eval_loss": 2.0400278568267822,
|
| 45 |
+
"eval_runtime": 237.5888,
|
| 46 |
+
"eval_samples_per_second": 20.881,
|
| 47 |
+
"eval_steps_per_second": 2.614,
|
| 48 |
+
"eval_wer": 1.0,
|
| 49 |
+
"step": 400
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 1.3297872340425532,
|
| 53 |
+
"grad_norm": 3.4516775608062744,
|
| 54 |
+
"learning_rate": 0.00029759999999999997,
|
| 55 |
+
"loss": 7.1041,
|
| 56 |
+
"step": 500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.3297872340425532,
|
| 60 |
+
"eval_cer": 0.43090559762339153,
|
| 61 |
+
"eval_loss": 1.0610116720199585,
|
| 62 |
+
"eval_runtime": 241.7004,
|
| 63 |
+
"eval_samples_per_second": 20.525,
|
| 64 |
+
"eval_steps_per_second": 2.569,
|
| 65 |
+
"eval_wer": 1.0,
|
| 66 |
+
"step": 500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.5957446808510638,
|
| 70 |
+
"eval_cer": 0.39546636097970517,
|
| 71 |
+
"eval_loss": 0.8836929798126221,
|
| 72 |
+
"eval_runtime": 237.4265,
|
| 73 |
+
"eval_samples_per_second": 20.895,
|
| 74 |
+
"eval_steps_per_second": 2.616,
|
| 75 |
+
"eval_wer": 1.0,
|
| 76 |
+
"step": 600
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 1.8617021276595744,
|
| 80 |
+
"eval_cer": 0.3790545231444855,
|
| 81 |
+
"eval_loss": 0.7705639004707336,
|
| 82 |
+
"eval_runtime": 238.309,
|
| 83 |
+
"eval_samples_per_second": 20.818,
|
| 84 |
+
"eval_steps_per_second": 2.606,
|
| 85 |
+
"eval_wer": 0.9997984277363435,
|
| 86 |
+
"step": 700
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.127659574468085,
|
| 90 |
+
"eval_cer": 0.3815893569272002,
|
| 91 |
+
"eval_loss": 0.766192615032196,
|
| 92 |
+
"eval_runtime": 235.61,
|
| 93 |
+
"eval_samples_per_second": 21.056,
|
| 94 |
+
"eval_steps_per_second": 2.636,
|
| 95 |
+
"eval_wer": 1.0,
|
| 96 |
+
"step": 800
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 2.393617021276596,
|
| 100 |
+
"eval_cer": 0.37896572323489996,
|
| 101 |
+
"eval_loss": 0.7620505094528198,
|
| 102 |
+
"eval_runtime": 237.8852,
|
| 103 |
+
"eval_samples_per_second": 20.855,
|
| 104 |
+
"eval_steps_per_second": 2.611,
|
| 105 |
+
"eval_wer": 1.0,
|
| 106 |
+
"step": 900
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 2.6595744680851063,
|
| 110 |
+
"grad_norm": 2.764026165008545,
|
| 111 |
+
"learning_rate": 0.00027105058365758754,
|
| 112 |
+
"loss": 0.803,
|
| 113 |
+
"step": 1000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.6595744680851063,
|
| 117 |
+
"eval_cer": 0.36259424899494647,
|
| 118 |
+
"eval_loss": 0.6968890428543091,
|
| 119 |
+
"eval_runtime": 238.5042,
|
| 120 |
+
"eval_samples_per_second": 20.8,
|
| 121 |
+
"eval_steps_per_second": 2.604,
|
| 122 |
+
"eval_wer": 1.0,
|
| 123 |
+
"step": 1000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.925531914893617,
|
| 127 |
+
"eval_cer": 0.3572904725769734,
|
| 128 |
+
"eval_loss": 0.6735507249832153,
|
| 129 |
+
"eval_runtime": 242.6929,
|
| 130 |
+
"eval_samples_per_second": 20.441,
|
| 131 |
+
"eval_steps_per_second": 2.559,
|
| 132 |
+
"eval_wer": 1.0,
|
| 133 |
+
"step": 1100
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 3.1914893617021276,
|
| 137 |
+
"eval_cer": 0.3543762209987568,
|
| 138 |
+
"eval_loss": 0.6823433637619019,
|
| 139 |
+
"eval_runtime": 250.863,
|
| 140 |
+
"eval_samples_per_second": 19.776,
|
| 141 |
+
"eval_steps_per_second": 2.475,
|
| 142 |
+
"eval_wer": 0.9997984277363435,
|
| 143 |
+
"step": 1200
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.4574468085106385,
|
| 147 |
+
"eval_cer": 0.3460209567786622,
|
| 148 |
+
"eval_loss": 0.6359558701515198,
|
| 149 |
+
"eval_runtime": 237.6152,
|
| 150 |
+
"eval_samples_per_second": 20.878,
|
| 151 |
+
"eval_steps_per_second": 2.613,
|
| 152 |
+
"eval_wer": 1.0,
|
| 153 |
+
"step": 1300
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 3.723404255319149,
|
| 157 |
+
"eval_cer": 0.34429339490127064,
|
| 158 |
+
"eval_loss": 0.6503622531890869,
|
| 159 |
+
"eval_runtime": 235.7171,
|
| 160 |
+
"eval_samples_per_second": 21.046,
|
| 161 |
+
"eval_steps_per_second": 2.635,
|
| 162 |
+
"eval_wer": 1.0,
|
| 163 |
+
"step": 1400
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 3.9893617021276597,
|
| 167 |
+
"grad_norm": 1.9848517179489136,
|
| 168 |
+
"learning_rate": 0.00024186770428015562,
|
| 169 |
+
"loss": 0.5675,
|
| 170 |
+
"step": 1500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.9893617021276597,
|
| 174 |
+
"eval_cer": 0.34139528876116054,
|
| 175 |
+
"eval_loss": 0.6246533393859863,
|
| 176 |
+
"eval_runtime": 239.1195,
|
| 177 |
+
"eval_samples_per_second": 20.747,
|
| 178 |
+
"eval_steps_per_second": 2.597,
|
| 179 |
+
"eval_wer": 1.0,
|
| 180 |
+
"step": 1500
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 4.25531914893617,
|
| 184 |
+
"eval_cer": 0.3424851058333468,
|
| 185 |
+
"eval_loss": 0.6397080421447754,
|
| 186 |
+
"eval_runtime": 242.7125,
|
| 187 |
+
"eval_samples_per_second": 20.44,
|
| 188 |
+
"eval_steps_per_second": 2.559,
|
| 189 |
+
"eval_wer": 0.9997984277363435,
|
| 190 |
+
"step": 1600
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 4.5212765957446805,
|
| 194 |
+
"eval_cer": 0.3439462679819817,
|
| 195 |
+
"eval_loss": 0.6589427590370178,
|
| 196 |
+
"eval_runtime": 243.7764,
|
| 197 |
+
"eval_samples_per_second": 20.351,
|
| 198 |
+
"eval_steps_per_second": 2.547,
|
| 199 |
+
"eval_wer": 1.0,
|
| 200 |
+
"step": 1700
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.787234042553192,
|
| 204 |
+
"eval_cer": 0.3449472851445824,
|
| 205 |
+
"eval_loss": 0.6345330476760864,
|
| 206 |
+
"eval_runtime": 237.3086,
|
| 207 |
+
"eval_samples_per_second": 20.905,
|
| 208 |
+
"eval_steps_per_second": 2.617,
|
| 209 |
+
"eval_wer": 1.0,
|
| 210 |
+
"step": 1800
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.053191489361702,
|
| 214 |
+
"eval_cer": 0.3379805286016436,
|
| 215 |
+
"eval_loss": 0.652181088924408,
|
| 216 |
+
"eval_runtime": 240.2379,
|
| 217 |
+
"eval_samples_per_second": 20.65,
|
| 218 |
+
"eval_steps_per_second": 2.585,
|
| 219 |
+
"eval_wer": 0.999596855472687,
|
| 220 |
+
"step": 1900
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 5.319148936170213,
|
| 224 |
+
"grad_norm": 2.756148099899292,
|
| 225 |
+
"learning_rate": 0.00021268482490272372,
|
| 226 |
+
"loss": 0.4421,
|
| 227 |
+
"step": 2000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 5.319148936170213,
|
| 231 |
+
"eval_cer": 0.3371732566963205,
|
| 232 |
+
"eval_loss": 0.6292724609375,
|
| 233 |
+
"eval_runtime": 240.5588,
|
| 234 |
+
"eval_samples_per_second": 20.623,
|
| 235 |
+
"eval_steps_per_second": 2.581,
|
| 236 |
+
"eval_wer": 1.0,
|
| 237 |
+
"step": 2000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 5.585106382978723,
|
| 241 |
+
"eval_cer": 0.3342428596799974,
|
| 242 |
+
"eval_loss": 0.6095719337463379,
|
| 243 |
+
"eval_runtime": 241.1797,
|
| 244 |
+
"eval_samples_per_second": 20.57,
|
| 245 |
+
"eval_steps_per_second": 2.575,
|
| 246 |
+
"eval_wer": 1.0,
|
| 247 |
+
"step": 2100
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 5.851063829787234,
|
| 251 |
+
"eval_cer": 0.3320712982546781,
|
| 252 |
+
"eval_loss": 0.6107529401779175,
|
| 253 |
+
"eval_runtime": 238.2079,
|
| 254 |
+
"eval_samples_per_second": 20.826,
|
| 255 |
+
"eval_steps_per_second": 2.607,
|
| 256 |
+
"eval_wer": 1.0,
|
| 257 |
+
"step": 2200
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 6.117021276595745,
|
| 261 |
+
"eval_cer": 0.3354295493808224,
|
| 262 |
+
"eval_loss": 0.6200144290924072,
|
| 263 |
+
"eval_runtime": 235.9894,
|
| 264 |
+
"eval_samples_per_second": 21.022,
|
| 265 |
+
"eval_steps_per_second": 2.631,
|
| 266 |
+
"eval_wer": 1.0,
|
| 267 |
+
"step": 2300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 6.382978723404255,
|
| 271 |
+
"eval_cer": 0.33407333257987953,
|
| 272 |
+
"eval_loss": 0.6413326263427734,
|
| 273 |
+
"eval_runtime": 236.6024,
|
| 274 |
+
"eval_samples_per_second": 20.968,
|
| 275 |
+
"eval_steps_per_second": 2.625,
|
| 276 |
+
"eval_wer": 1.0,
|
| 277 |
+
"step": 2400
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 6.648936170212766,
|
| 281 |
+
"grad_norm": 4.183356761932373,
|
| 282 |
+
"learning_rate": 0.00018350194552529183,
|
| 283 |
+
"loss": 0.3699,
|
| 284 |
+
"step": 2500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 6.648936170212766,
|
| 288 |
+
"eval_cer": 0.3359462034002293,
|
| 289 |
+
"eval_loss": 0.6303403973579407,
|
| 290 |
+
"eval_runtime": 237.052,
|
| 291 |
+
"eval_samples_per_second": 20.928,
|
| 292 |
+
"eval_steps_per_second": 2.62,
|
| 293 |
+
"eval_wer": 0.999596855472687,
|
| 294 |
+
"step": 2500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 6.914893617021277,
|
| 298 |
+
"eval_cer": 0.3307796632061611,
|
| 299 |
+
"eval_loss": 0.6012922525405884,
|
| 300 |
+
"eval_runtime": 233.4251,
|
| 301 |
+
"eval_samples_per_second": 21.253,
|
| 302 |
+
"eval_steps_per_second": 2.66,
|
| 303 |
+
"eval_wer": 1.0,
|
| 304 |
+
"step": 2600
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 7.180851063829787,
|
| 308 |
+
"eval_cer": 0.32859195634273536,
|
| 309 |
+
"eval_loss": 0.634281575679779,
|
| 310 |
+
"eval_runtime": 232.2994,
|
| 311 |
+
"eval_samples_per_second": 21.356,
|
| 312 |
+
"eval_steps_per_second": 2.673,
|
| 313 |
+
"eval_wer": 1.0,
|
| 314 |
+
"step": 2700
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 7.446808510638298,
|
| 318 |
+
"eval_cer": 0.3260086862457013,
|
| 319 |
+
"eval_loss": 0.6208234429359436,
|
| 320 |
+
"eval_runtime": 235.1613,
|
| 321 |
+
"eval_samples_per_second": 21.096,
|
| 322 |
+
"eval_steps_per_second": 2.641,
|
| 323 |
+
"eval_wer": 0.9997984277363435,
|
| 324 |
+
"step": 2800
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 7.712765957446808,
|
| 328 |
+
"eval_cer": 0.3287291925666403,
|
| 329 |
+
"eval_loss": 0.6095162034034729,
|
| 330 |
+
"eval_runtime": 241.6677,
|
| 331 |
+
"eval_samples_per_second": 20.528,
|
| 332 |
+
"eval_steps_per_second": 2.57,
|
| 333 |
+
"eval_wer": 0.9997984277363435,
|
| 334 |
+
"step": 2900
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 7.9787234042553195,
|
| 338 |
+
"grad_norm": 1.701768159866333,
|
| 339 |
+
"learning_rate": 0.0001543190661478599,
|
| 340 |
+
"loss": 0.3146,
|
| 341 |
+
"step": 3000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 7.9787234042553195,
|
| 345 |
+
"eval_cer": 0.3265818492984807,
|
| 346 |
+
"eval_loss": 0.6058351397514343,
|
| 347 |
+
"eval_runtime": 235.659,
|
| 348 |
+
"eval_samples_per_second": 21.052,
|
| 349 |
+
"eval_steps_per_second": 2.635,
|
| 350 |
+
"eval_wer": 0.999596855472687,
|
| 351 |
+
"step": 3000
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 8.24468085106383,
|
| 355 |
+
"eval_cer": 0.3251368325879523,
|
| 356 |
+
"eval_loss": 0.6612707376480103,
|
| 357 |
+
"eval_runtime": 233.8878,
|
| 358 |
+
"eval_samples_per_second": 21.211,
|
| 359 |
+
"eval_steps_per_second": 2.655,
|
| 360 |
+
"eval_wer": 0.999596855472687,
|
| 361 |
+
"step": 3100
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 8.51063829787234,
|
| 365 |
+
"eval_cer": 0.3244022151541082,
|
| 366 |
+
"eval_loss": 0.6538846492767334,
|
| 367 |
+
"eval_runtime": 235.5702,
|
| 368 |
+
"eval_samples_per_second": 21.06,
|
| 369 |
+
"eval_steps_per_second": 2.636,
|
| 370 |
+
"eval_wer": 1.0,
|
| 371 |
+
"step": 3200
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 8.77659574468085,
|
| 375 |
+
"eval_cer": 0.32638810404120316,
|
| 376 |
+
"eval_loss": 0.6331284046173096,
|
| 377 |
+
"eval_runtime": 236.8675,
|
| 378 |
+
"eval_samples_per_second": 20.944,
|
| 379 |
+
"eval_steps_per_second": 2.622,
|
| 380 |
+
"eval_wer": 1.0,
|
| 381 |
+
"step": 3300
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 9.042553191489361,
|
| 385 |
+
"eval_cer": 0.32281996221967485,
|
| 386 |
+
"eval_loss": 0.6436325907707214,
|
| 387 |
+
"eval_runtime": 236.1998,
|
| 388 |
+
"eval_samples_per_second": 21.003,
|
| 389 |
+
"eval_steps_per_second": 2.629,
|
| 390 |
+
"eval_wer": 1.0,
|
| 391 |
+
"step": 3400
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 9.308510638297872,
|
| 395 |
+
"grad_norm": 1.067522406578064,
|
| 396 |
+
"learning_rate": 0.000125136186770428,
|
| 397 |
+
"loss": 0.2576,
|
| 398 |
+
"step": 3500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 9.308510638297872,
|
| 402 |
+
"eval_cer": 0.32351421605825276,
|
| 403 |
+
"eval_loss": 0.6328682899475098,
|
| 404 |
+
"eval_runtime": 234.2706,
|
| 405 |
+
"eval_samples_per_second": 21.176,
|
| 406 |
+
"eval_steps_per_second": 2.651,
|
| 407 |
+
"eval_wer": 1.0,
|
| 408 |
+
"step": 3500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 9.574468085106384,
|
| 412 |
+
"eval_cer": 0.3196958199460742,
|
| 413 |
+
"eval_loss": 0.6314510703086853,
|
| 414 |
+
"eval_runtime": 234.3665,
|
| 415 |
+
"eval_samples_per_second": 21.168,
|
| 416 |
+
"eval_steps_per_second": 2.65,
|
| 417 |
+
"eval_wer": 0.9997984277363435,
|
| 418 |
+
"step": 3600
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 9.840425531914894,
|
| 422 |
+
"eval_cer": 0.32033356475127955,
|
| 423 |
+
"eval_loss": 0.6280702352523804,
|
| 424 |
+
"eval_runtime": 236.442,
|
| 425 |
+
"eval_samples_per_second": 20.982,
|
| 426 |
+
"eval_steps_per_second": 2.626,
|
| 427 |
+
"eval_wer": 0.9997984277363435,
|
| 428 |
+
"step": 3700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 10.106382978723405,
|
| 432 |
+
"eval_cer": 0.3196473836317548,
|
| 433 |
+
"eval_loss": 0.6696433424949646,
|
| 434 |
+
"eval_runtime": 232.6849,
|
| 435 |
+
"eval_samples_per_second": 21.321,
|
| 436 |
+
"eval_steps_per_second": 2.669,
|
| 437 |
+
"eval_wer": 0.999596855472687,
|
| 438 |
+
"step": 3800
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 10.372340425531915,
|
| 442 |
+
"eval_cer": 0.3199218560795647,
|
| 443 |
+
"eval_loss": 0.6629988551139832,
|
| 444 |
+
"eval_runtime": 233.1379,
|
| 445 |
+
"eval_samples_per_second": 21.279,
|
| 446 |
+
"eval_steps_per_second": 2.664,
|
| 447 |
+
"eval_wer": 0.999596855472687,
|
| 448 |
+
"step": 3900
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 10.638297872340425,
|
| 452 |
+
"grad_norm": 1.6601381301879883,
|
| 453 |
+
"learning_rate": 9.595330739299609e-05,
|
| 454 |
+
"loss": 0.2201,
|
| 455 |
+
"step": 4000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 10.638297872340425,
|
| 459 |
+
"eval_cer": 0.3203093465941198,
|
| 460 |
+
"eval_loss": 0.6781216859817505,
|
| 461 |
+
"eval_runtime": 232.2432,
|
| 462 |
+
"eval_samples_per_second": 21.361,
|
| 463 |
+
"eval_steps_per_second": 2.674,
|
| 464 |
+
"eval_wer": 1.0,
|
| 465 |
+
"step": 4000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 10.904255319148936,
|
| 469 |
+
"eval_cer": 0.31962316547459513,
|
| 470 |
+
"eval_loss": 0.6531046628952026,
|
| 471 |
+
"eval_runtime": 239.9955,
|
| 472 |
+
"eval_samples_per_second": 20.671,
|
| 473 |
+
"eval_steps_per_second": 2.588,
|
| 474 |
+
"eval_wer": 1.0,
|
| 475 |
+
"step": 4100
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 11.170212765957446,
|
| 479 |
+
"eval_cer": 0.3192598931171997,
|
| 480 |
+
"eval_loss": 0.6762946248054504,
|
| 481 |
+
"eval_runtime": 235.4369,
|
| 482 |
+
"eval_samples_per_second": 21.071,
|
| 483 |
+
"eval_steps_per_second": 2.638,
|
| 484 |
+
"eval_wer": 0.9997984277363435,
|
| 485 |
+
"step": 4200
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 11.436170212765958,
|
| 489 |
+
"eval_cer": 0.3184364757737701,
|
| 490 |
+
"eval_loss": 0.6785323023796082,
|
| 491 |
+
"eval_runtime": 232.8132,
|
| 492 |
+
"eval_samples_per_second": 21.309,
|
| 493 |
+
"eval_steps_per_second": 2.667,
|
| 494 |
+
"eval_wer": 1.0,
|
| 495 |
+
"step": 4300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 11.702127659574469,
|
| 499 |
+
"eval_cer": 0.31793596719246975,
|
| 500 |
+
"eval_loss": 0.666408360004425,
|
| 501 |
+
"eval_runtime": 232.0593,
|
| 502 |
+
"eval_samples_per_second": 21.378,
|
| 503 |
+
"eval_steps_per_second": 2.676,
|
| 504 |
+
"eval_wer": 0.9997984277363435,
|
| 505 |
+
"step": 4400
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 11.96808510638298,
|
| 509 |
+
"grad_norm": 0.7692495584487915,
|
| 510 |
+
"learning_rate": 6.67704280155642e-05,
|
| 511 |
+
"loss": 0.1931,
|
| 512 |
+
"step": 4500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 11.96808510638298,
|
| 516 |
+
"eval_cer": 0.31842033033566364,
|
| 517 |
+
"eval_loss": 0.6682071089744568,
|
| 518 |
+
"eval_runtime": 235.6644,
|
| 519 |
+
"eval_samples_per_second": 21.051,
|
| 520 |
+
"eval_steps_per_second": 2.635,
|
| 521 |
+
"eval_wer": 0.9997984277363435,
|
| 522 |
+
"step": 4500
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 12.23404255319149,
|
| 526 |
+
"eval_cer": 0.3168219319631238,
|
| 527 |
+
"eval_loss": 0.6799555420875549,
|
| 528 |
+
"eval_runtime": 235.3483,
|
| 529 |
+
"eval_samples_per_second": 21.079,
|
| 530 |
+
"eval_steps_per_second": 2.639,
|
| 531 |
+
"eval_wer": 0.9997984277363435,
|
| 532 |
+
"step": 4600
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 12.5,
|
| 536 |
+
"eval_cer": 0.3162245507531847,
|
| 537 |
+
"eval_loss": 0.6925452351570129,
|
| 538 |
+
"eval_runtime": 235.4523,
|
| 539 |
+
"eval_samples_per_second": 21.07,
|
| 540 |
+
"eval_steps_per_second": 2.637,
|
| 541 |
+
"eval_wer": 1.0,
|
| 542 |
+
"step": 4700
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 12.76595744680851,
|
| 546 |
+
"eval_cer": 0.3144646979995802,
|
| 547 |
+
"eval_loss": 0.7046905159950256,
|
| 548 |
+
"eval_runtime": 235.1343,
|
| 549 |
+
"eval_samples_per_second": 21.099,
|
| 550 |
+
"eval_steps_per_second": 2.641,
|
| 551 |
+
"eval_wer": 1.0,
|
| 552 |
+
"step": 4800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 13.03191489361702,
|
| 556 |
+
"eval_cer": 0.3147230250092836,
|
| 557 |
+
"eval_loss": 0.6918711066246033,
|
| 558 |
+
"eval_runtime": 238.6326,
|
| 559 |
+
"eval_samples_per_second": 20.789,
|
| 560 |
+
"eval_steps_per_second": 2.602,
|
| 561 |
+
"eval_wer": 0.9997984277363435,
|
| 562 |
+
"step": 4900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 13.297872340425531,
|
| 566 |
+
"grad_norm": 1.052759051322937,
|
| 567 |
+
"learning_rate": 3.758754863813229e-05,
|
| 568 |
+
"loss": 0.1694,
|
| 569 |
+
"step": 5000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 13.297872340425531,
|
| 573 |
+
"eval_cer": 0.31423866186608973,
|
| 574 |
+
"eval_loss": 0.699897289276123,
|
| 575 |
+
"eval_runtime": 256.2343,
|
| 576 |
+
"eval_samples_per_second": 19.361,
|
| 577 |
+
"eval_steps_per_second": 2.424,
|
| 578 |
+
"eval_wer": 0.9997984277363435,
|
| 579 |
+
"step": 5000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 13.563829787234042,
|
| 583 |
+
"eval_cer": 0.3134394626798198,
|
| 584 |
+
"eval_loss": 0.6994884610176086,
|
| 585 |
+
"eval_runtime": 236.0872,
|
| 586 |
+
"eval_samples_per_second": 21.013,
|
| 587 |
+
"eval_steps_per_second": 2.63,
|
| 588 |
+
"eval_wer": 1.0,
|
| 589 |
+
"step": 5100
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 13.829787234042554,
|
| 593 |
+
"eval_cer": 0.3133829536464472,
|
| 594 |
+
"eval_loss": 0.6916852593421936,
|
| 595 |
+
"eval_runtime": 234.7745,
|
| 596 |
+
"eval_samples_per_second": 21.131,
|
| 597 |
+
"eval_steps_per_second": 2.645,
|
| 598 |
+
"eval_wer": 0.9997984277363435,
|
| 599 |
+
"step": 5200
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 14.095744680851064,
|
| 603 |
+
"eval_cer": 0.3128501541889339,
|
| 604 |
+
"eval_loss": 0.6962644457817078,
|
| 605 |
+
"eval_runtime": 235.7153,
|
| 606 |
+
"eval_samples_per_second": 21.047,
|
| 607 |
+
"eval_steps_per_second": 2.635,
|
| 608 |
+
"eval_wer": 0.9997984277363435,
|
| 609 |
+
"step": 5300
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 14.361702127659575,
|
| 613 |
+
"eval_cer": 0.3128259360317742,
|
| 614 |
+
"eval_loss": 0.6961241960525513,
|
| 615 |
+
"eval_runtime": 245.639,
|
| 616 |
+
"eval_samples_per_second": 20.196,
|
| 617 |
+
"eval_steps_per_second": 2.528,
|
| 618 |
+
"eval_wer": 0.9997984277363435,
|
| 619 |
+
"step": 5400
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 14.627659574468085,
|
| 623 |
+
"grad_norm": 2.195582389831543,
|
| 624 |
+
"learning_rate": 8.404669260700388e-06,
|
| 625 |
+
"loss": 0.1548,
|
| 626 |
+
"step": 5500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 14.627659574468085,
|
| 630 |
+
"eval_cer": 0.31290666322230654,
|
| 631 |
+
"eval_loss": 0.6963828802108765,
|
| 632 |
+
"eval_runtime": 237.6797,
|
| 633 |
+
"eval_samples_per_second": 20.873,
|
| 634 |
+
"eval_steps_per_second": 2.613,
|
| 635 |
+
"eval_wer": 1.0,
|
| 636 |
+
"step": 5500
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 14.893617021276595,
|
| 640 |
+
"eval_cer": 0.3126725543697628,
|
| 641 |
+
"eval_loss": 0.6983669400215149,
|
| 642 |
+
"eval_runtime": 245.8138,
|
| 643 |
+
"eval_samples_per_second": 20.182,
|
| 644 |
+
"eval_steps_per_second": 2.526,
|
| 645 |
+
"eval_wer": 0.9997984277363435,
|
| 646 |
+
"step": 5600
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 15.0,
|
| 650 |
+
"step": 5640,
|
| 651 |
+
"total_flos": 2.6348448154740388e+19,
|
| 652 |
+
"train_loss": 0.9431738045198698,
|
| 653 |
+
"train_runtime": 36296.3273,
|
| 654 |
+
"train_samples_per_second": 4.972,
|
| 655 |
+
"train_steps_per_second": 0.155
|
| 656 |
+
}
|
| 657 |
+
],
|
| 658 |
+
"logging_steps": 500,
|
| 659 |
+
"max_steps": 5640,
|
| 660 |
+
"num_input_tokens_seen": 0,
|
| 661 |
+
"num_train_epochs": 15,
|
| 662 |
+
"save_steps": 400,
|
| 663 |
+
"stateful_callbacks": {
|
| 664 |
+
"TrainerControl": {
|
| 665 |
+
"args": {
|
| 666 |
+
"should_epoch_stop": false,
|
| 667 |
+
"should_evaluate": false,
|
| 668 |
+
"should_log": false,
|
| 669 |
+
"should_save": true,
|
| 670 |
+
"should_training_stop": true
|
| 671 |
+
},
|
| 672 |
+
"attributes": {}
|
| 673 |
+
}
|
| 674 |
+
},
|
| 675 |
+
"total_flos": 2.6348448154740388e+19,
|
| 676 |
+
"train_batch_size": 16,
|
| 677 |
+
"trial_name": null,
|
| 678 |
+
"trial_params": null
|
| 679 |
+
}
|