End of training
Browse files- README.md +9 -6
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Jan26_04-57-46_ip-172-31-23-125/events.out.tfevents.1706285008.ip-172-31-23-125.1023.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +1449 -0
README.md
CHANGED
@@ -1,21 +1,24 @@
|
|
1 |
---
|
|
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: arun100/whisper-base-bn-3
|
4 |
tags:
|
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
-
- common_voice_16_0
|
8 |
metrics:
|
9 |
- wer
|
10 |
model-index:
|
11 |
-
- name:
|
12 |
results:
|
13 |
- task:
|
14 |
name: Automatic Speech Recognition
|
15 |
type: automatic-speech-recognition
|
16 |
dataset:
|
17 |
-
name: common_voice_16_0
|
18 |
-
type: common_voice_16_0
|
19 |
config: bn
|
20 |
split: test
|
21 |
args: bn
|
@@ -28,9 +31,9 @@ model-index:
|
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
should probably proofread and complete it, then remove this comment. -->
|
30 |
|
31 |
-
#
|
32 |
|
33 |
-
This model is a fine-tuned version of [arun100/whisper-base-bn-3](https://huggingface.co/arun100/whisper-base-bn-3) on the common_voice_16_0 dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
- Loss: 0.2078
|
36 |
- Wer: 28.8185
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- bn
|
4 |
license: apache-2.0
|
5 |
base_model: arun100/whisper-base-bn-3
|
6 |
tags:
|
7 |
+
- whisper-event
|
8 |
- generated_from_trainer
|
9 |
datasets:
|
10 |
+
- mozilla-foundation/common_voice_16_0
|
11 |
metrics:
|
12 |
- wer
|
13 |
model-index:
|
14 |
+
- name: Whisper Base Bengali
|
15 |
results:
|
16 |
- task:
|
17 |
name: Automatic Speech Recognition
|
18 |
type: automatic-speech-recognition
|
19 |
dataset:
|
20 |
+
name: mozilla-foundation/common_voice_16_0 bn
|
21 |
+
type: mozilla-foundation/common_voice_16_0
|
22 |
config: bn
|
23 |
split: test
|
24 |
args: bn
|
|
|
31 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
32 |
should probably proofread and complete it, then remove this comment. -->
|
33 |
|
34 |
+
# Whisper Base Bengali
|
35 |
|
36 |
+
This model is a fine-tuned version of [arun100/whisper-base-bn-3](https://huggingface.co/arun100/whisper-base-bn-3) on the mozilla-foundation/common_voice_16_0 bn dataset.
|
37 |
It achieves the following results on the evaluation set:
|
38 |
- Loss: 0.2078
|
39 |
- Wer: 28.8185
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 18.87,
|
3 |
+
"eval_loss": 0.20784465968608856,
|
4 |
+
"eval_runtime": 1894.4239,
|
5 |
+
"eval_samples_per_second": 4.917,
|
6 |
+
"eval_steps_per_second": 0.154,
|
7 |
+
"eval_wer": 28.818465723515253,
|
8 |
+
"train_loss": 0.14273875054446133,
|
9 |
+
"train_runtime": 37524.2555,
|
10 |
+
"train_samples_per_second": 9.381,
|
11 |
+
"train_steps_per_second": 0.147
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 18.87,
|
3 |
+
"eval_loss": 0.20784465968608856,
|
4 |
+
"eval_runtime": 1894.4239,
|
5 |
+
"eval_samples_per_second": 4.917,
|
6 |
+
"eval_steps_per_second": 0.154,
|
7 |
+
"eval_wer": 28.818465723515253
|
8 |
+
}
|
runs/Jan26_04-57-46_ip-172-31-23-125/events.out.tfevents.1706285008.ip-172-31-23-125.1023.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e8358ef197089c2bae6c459f9ae1ad2abb351d05e6ec140d6ae44ce077972a01
|
3 |
+
size 406
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 18.87,
|
3 |
+
"train_loss": 0.14273875054446133,
|
4 |
+
"train_runtime": 37524.2555,
|
5 |
+
"train_samples_per_second": 9.381,
|
6 |
+
"train_steps_per_second": 0.147
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1449 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 28.818465723515253,
|
3 |
+
"best_model_checkpoint": "./checkpoint-5500",
|
4 |
+
"epoch": 18.867924528301888,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 5500,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.09,
|
13 |
+
"learning_rate": 2.5e-08,
|
14 |
+
"loss": 0.1552,
|
15 |
+
"step": 25
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.17,
|
19 |
+
"learning_rate": 5e-08,
|
20 |
+
"loss": 0.1515,
|
21 |
+
"step": 50
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.26,
|
25 |
+
"learning_rate": 7.5e-08,
|
26 |
+
"loss": 0.1549,
|
27 |
+
"step": 75
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.34,
|
31 |
+
"learning_rate": 1e-07,
|
32 |
+
"loss": 0.1553,
|
33 |
+
"step": 100
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.43,
|
37 |
+
"learning_rate": 1.25e-07,
|
38 |
+
"loss": 0.1547,
|
39 |
+
"step": 125
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.51,
|
43 |
+
"learning_rate": 1.5e-07,
|
44 |
+
"loss": 0.1544,
|
45 |
+
"step": 150
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.6,
|
49 |
+
"learning_rate": 1.75e-07,
|
50 |
+
"loss": 0.1523,
|
51 |
+
"step": 175
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.69,
|
55 |
+
"learning_rate": 2e-07,
|
56 |
+
"loss": 0.1511,
|
57 |
+
"step": 200
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.77,
|
61 |
+
"learning_rate": 2.25e-07,
|
62 |
+
"loss": 0.1556,
|
63 |
+
"step": 225
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.86,
|
67 |
+
"learning_rate": 2.5e-07,
|
68 |
+
"loss": 0.1521,
|
69 |
+
"step": 250
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.94,
|
73 |
+
"learning_rate": 2.75e-07,
|
74 |
+
"loss": 0.1526,
|
75 |
+
"step": 275
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 1.03,
|
79 |
+
"learning_rate": 3e-07,
|
80 |
+
"loss": 0.1534,
|
81 |
+
"step": 300
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 1.11,
|
85 |
+
"learning_rate": 3.25e-07,
|
86 |
+
"loss": 0.1533,
|
87 |
+
"step": 325
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 1.2,
|
91 |
+
"learning_rate": 3.5e-07,
|
92 |
+
"loss": 0.152,
|
93 |
+
"step": 350
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 1.29,
|
97 |
+
"learning_rate": 3.75e-07,
|
98 |
+
"loss": 0.1531,
|
99 |
+
"step": 375
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 1.37,
|
103 |
+
"learning_rate": 4e-07,
|
104 |
+
"loss": 0.1516,
|
105 |
+
"step": 400
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 1.46,
|
109 |
+
"learning_rate": 4.2499999999999995e-07,
|
110 |
+
"loss": 0.1539,
|
111 |
+
"step": 425
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 1.54,
|
115 |
+
"learning_rate": 4.5e-07,
|
116 |
+
"loss": 0.1531,
|
117 |
+
"step": 450
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 1.63,
|
121 |
+
"learning_rate": 4.7499999999999995e-07,
|
122 |
+
"loss": 0.154,
|
123 |
+
"step": 475
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 1.72,
|
127 |
+
"learning_rate": 5e-07,
|
128 |
+
"loss": 0.1568,
|
129 |
+
"step": 500
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 1.72,
|
133 |
+
"eval_loss": 0.21451939642429352,
|
134 |
+
"eval_runtime": 1884.5796,
|
135 |
+
"eval_samples_per_second": 4.942,
|
136 |
+
"eval_steps_per_second": 0.155,
|
137 |
+
"eval_wer": 29.834993244029246,
|
138 |
+
"step": 500
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 1.8,
|
142 |
+
"learning_rate": 4.975e-07,
|
143 |
+
"loss": 0.1544,
|
144 |
+
"step": 525
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 1.89,
|
148 |
+
"learning_rate": 4.95e-07,
|
149 |
+
"loss": 0.1502,
|
150 |
+
"step": 550
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 1.97,
|
154 |
+
"learning_rate": 4.924999999999999e-07,
|
155 |
+
"loss": 0.1525,
|
156 |
+
"step": 575
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 2.06,
|
160 |
+
"learning_rate": 4.9e-07,
|
161 |
+
"loss": 0.1522,
|
162 |
+
"step": 600
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 2.14,
|
166 |
+
"learning_rate": 4.875e-07,
|
167 |
+
"loss": 0.1506,
|
168 |
+
"step": 625
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 2.23,
|
172 |
+
"learning_rate": 4.85e-07,
|
173 |
+
"loss": 0.1537,
|
174 |
+
"step": 650
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 2.32,
|
178 |
+
"learning_rate": 4.824999999999999e-07,
|
179 |
+
"loss": 0.1536,
|
180 |
+
"step": 675
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 2.4,
|
184 |
+
"learning_rate": 4.8e-07,
|
185 |
+
"loss": 0.1561,
|
186 |
+
"step": 700
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 2.49,
|
190 |
+
"learning_rate": 4.775e-07,
|
191 |
+
"loss": 0.1488,
|
192 |
+
"step": 725
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 2.57,
|
196 |
+
"learning_rate": 4.7499999999999995e-07,
|
197 |
+
"loss": 0.1507,
|
198 |
+
"step": 750
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 2.66,
|
202 |
+
"learning_rate": 4.725e-07,
|
203 |
+
"loss": 0.1498,
|
204 |
+
"step": 775
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 2.74,
|
208 |
+
"learning_rate": 4.6999999999999995e-07,
|
209 |
+
"loss": 0.1461,
|
210 |
+
"step": 800
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 2.83,
|
214 |
+
"learning_rate": 4.675e-07,
|
215 |
+
"loss": 0.1531,
|
216 |
+
"step": 825
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 2.92,
|
220 |
+
"learning_rate": 4.65e-07,
|
221 |
+
"loss": 0.1542,
|
222 |
+
"step": 850
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 3.0,
|
226 |
+
"learning_rate": 4.625e-07,
|
227 |
+
"loss": 0.1517,
|
228 |
+
"step": 875
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 3.09,
|
232 |
+
"learning_rate": 4.6e-07,
|
233 |
+
"loss": 0.1554,
|
234 |
+
"step": 900
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 3.17,
|
238 |
+
"learning_rate": 4.575e-07,
|
239 |
+
"loss": 0.1456,
|
240 |
+
"step": 925
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 3.26,
|
244 |
+
"learning_rate": 4.55e-07,
|
245 |
+
"loss": 0.1457,
|
246 |
+
"step": 950
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 3.34,
|
250 |
+
"learning_rate": 4.525e-07,
|
251 |
+
"loss": 0.1536,
|
252 |
+
"step": 975
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 3.43,
|
256 |
+
"learning_rate": 4.5e-07,
|
257 |
+
"loss": 0.1507,
|
258 |
+
"step": 1000
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 3.43,
|
262 |
+
"eval_loss": 0.213189035654068,
|
263 |
+
"eval_runtime": 1879.633,
|
264 |
+
"eval_samples_per_second": 4.955,
|
265 |
+
"eval_steps_per_second": 0.155,
|
266 |
+
"eval_wer": 29.559385430368764,
|
267 |
+
"step": 1000
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 3.52,
|
271 |
+
"learning_rate": 4.475e-07,
|
272 |
+
"loss": 0.1523,
|
273 |
+
"step": 1025
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 3.6,
|
277 |
+
"learning_rate": 4.45e-07,
|
278 |
+
"loss": 0.1476,
|
279 |
+
"step": 1050
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 3.69,
|
283 |
+
"learning_rate": 4.425e-07,
|
284 |
+
"loss": 0.1485,
|
285 |
+
"step": 1075
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 3.77,
|
289 |
+
"learning_rate": 4.3999999999999997e-07,
|
290 |
+
"loss": 0.1508,
|
291 |
+
"step": 1100
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 3.86,
|
295 |
+
"learning_rate": 4.375e-07,
|
296 |
+
"loss": 0.1481,
|
297 |
+
"step": 1125
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 3.95,
|
301 |
+
"learning_rate": 4.3499999999999996e-07,
|
302 |
+
"loss": 0.1485,
|
303 |
+
"step": 1150
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 4.03,
|
307 |
+
"learning_rate": 4.325e-07,
|
308 |
+
"loss": 0.1483,
|
309 |
+
"step": 1175
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 4.12,
|
313 |
+
"learning_rate": 4.2999999999999996e-07,
|
314 |
+
"loss": 0.148,
|
315 |
+
"step": 1200
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 4.2,
|
319 |
+
"learning_rate": 4.275e-07,
|
320 |
+
"loss": 0.1461,
|
321 |
+
"step": 1225
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 4.29,
|
325 |
+
"learning_rate": 4.2499999999999995e-07,
|
326 |
+
"loss": 0.1501,
|
327 |
+
"step": 1250
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 4.37,
|
331 |
+
"learning_rate": 4.225e-07,
|
332 |
+
"loss": 0.1474,
|
333 |
+
"step": 1275
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 4.46,
|
337 |
+
"learning_rate": 4.1999999999999995e-07,
|
338 |
+
"loss": 0.1479,
|
339 |
+
"step": 1300
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 4.55,
|
343 |
+
"learning_rate": 4.1749999999999997e-07,
|
344 |
+
"loss": 0.1514,
|
345 |
+
"step": 1325
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 4.63,
|
349 |
+
"learning_rate": 4.1499999999999994e-07,
|
350 |
+
"loss": 0.1464,
|
351 |
+
"step": 1350
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 4.72,
|
355 |
+
"learning_rate": 4.1249999999999997e-07,
|
356 |
+
"loss": 0.1472,
|
357 |
+
"step": 1375
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 4.8,
|
361 |
+
"learning_rate": 4.0999999999999994e-07,
|
362 |
+
"loss": 0.1481,
|
363 |
+
"step": 1400
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 4.89,
|
367 |
+
"learning_rate": 4.0749999999999996e-07,
|
368 |
+
"loss": 0.1486,
|
369 |
+
"step": 1425
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 4.97,
|
373 |
+
"learning_rate": 4.05e-07,
|
374 |
+
"loss": 0.1475,
|
375 |
+
"step": 1450
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 5.06,
|
379 |
+
"learning_rate": 4.025e-07,
|
380 |
+
"loss": 0.149,
|
381 |
+
"step": 1475
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 5.15,
|
385 |
+
"learning_rate": 4e-07,
|
386 |
+
"loss": 0.1466,
|
387 |
+
"step": 1500
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 5.15,
|
391 |
+
"eval_loss": 0.21188345551490784,
|
392 |
+
"eval_runtime": 1877.5748,
|
393 |
+
"eval_samples_per_second": 4.961,
|
394 |
+
"eval_steps_per_second": 0.156,
|
395 |
+
"eval_wer": 29.357601138224474,
|
396 |
+
"step": 1500
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 5.23,
|
400 |
+
"learning_rate": 3.975e-07,
|
401 |
+
"loss": 0.1464,
|
402 |
+
"step": 1525
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 5.32,
|
406 |
+
"learning_rate": 3.95e-07,
|
407 |
+
"loss": 0.1455,
|
408 |
+
"step": 1550
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 5.4,
|
412 |
+
"learning_rate": 3.925e-07,
|
413 |
+
"loss": 0.1457,
|
414 |
+
"step": 1575
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 5.49,
|
418 |
+
"learning_rate": 3.8999999999999997e-07,
|
419 |
+
"loss": 0.1458,
|
420 |
+
"step": 1600
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 5.57,
|
424 |
+
"learning_rate": 3.875e-07,
|
425 |
+
"loss": 0.1447,
|
426 |
+
"step": 1625
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 5.66,
|
430 |
+
"learning_rate": 3.8499999999999997e-07,
|
431 |
+
"loss": 0.1472,
|
432 |
+
"step": 1650
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 5.75,
|
436 |
+
"learning_rate": 3.825e-07,
|
437 |
+
"loss": 0.1469,
|
438 |
+
"step": 1675
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 5.83,
|
442 |
+
"learning_rate": 3.7999999999999996e-07,
|
443 |
+
"loss": 0.1446,
|
444 |
+
"step": 1700
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 5.92,
|
448 |
+
"learning_rate": 3.775e-07,
|
449 |
+
"loss": 0.1433,
|
450 |
+
"step": 1725
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 6.0,
|
454 |
+
"learning_rate": 3.75e-07,
|
455 |
+
"loss": 0.1504,
|
456 |
+
"step": 1750
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 6.09,
|
460 |
+
"learning_rate": 3.725e-07,
|
461 |
+
"loss": 0.147,
|
462 |
+
"step": 1775
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 6.17,
|
466 |
+
"learning_rate": 3.7e-07,
|
467 |
+
"loss": 0.148,
|
468 |
+
"step": 1800
|
469 |
+
},
|
470 |
+
{
|
471 |
+
"epoch": 6.26,
|
472 |
+
"learning_rate": 3.675e-07,
|
473 |
+
"loss": 0.1471,
|
474 |
+
"step": 1825
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 6.35,
|
478 |
+
"learning_rate": 3.65e-07,
|
479 |
+
"loss": 0.1412,
|
480 |
+
"step": 1850
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 6.43,
|
484 |
+
"learning_rate": 3.6249999999999997e-07,
|
485 |
+
"loss": 0.1472,
|
486 |
+
"step": 1875
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 6.52,
|
490 |
+
"learning_rate": 3.6e-07,
|
491 |
+
"loss": 0.1427,
|
492 |
+
"step": 1900
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 6.6,
|
496 |
+
"learning_rate": 3.5749999999999997e-07,
|
497 |
+
"loss": 0.1431,
|
498 |
+
"step": 1925
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 6.69,
|
502 |
+
"learning_rate": 3.55e-07,
|
503 |
+
"loss": 0.1449,
|
504 |
+
"step": 1950
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 6.78,
|
508 |
+
"learning_rate": 3.5249999999999996e-07,
|
509 |
+
"loss": 0.1404,
|
510 |
+
"step": 1975
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 6.86,
|
514 |
+
"learning_rate": 3.5e-07,
|
515 |
+
"loss": 0.1463,
|
516 |
+
"step": 2000
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 6.86,
|
520 |
+
"eval_loss": 0.21055449545383453,
|
521 |
+
"eval_runtime": 1876.6891,
|
522 |
+
"eval_samples_per_second": 4.963,
|
523 |
+
"eval_steps_per_second": 0.156,
|
524 |
+
"eval_wer": 29.2927259223466,
|
525 |
+
"step": 2000
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 6.95,
|
529 |
+
"learning_rate": 3.4749999999999996e-07,
|
530 |
+
"loss": 0.1432,
|
531 |
+
"step": 2025
|
532 |
+
},
|
533 |
+
{
|
534 |
+
"epoch": 7.03,
|
535 |
+
"learning_rate": 3.45e-07,
|
536 |
+
"loss": 0.1493,
|
537 |
+
"step": 2050
|
538 |
+
},
|
539 |
+
{
|
540 |
+
"epoch": 7.12,
|
541 |
+
"learning_rate": 3.425e-07,
|
542 |
+
"loss": 0.1445,
|
543 |
+
"step": 2075
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 7.2,
|
547 |
+
"learning_rate": 3.4000000000000003e-07,
|
548 |
+
"loss": 0.1396,
|
549 |
+
"step": 2100
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 7.29,
|
553 |
+
"learning_rate": 3.375e-07,
|
554 |
+
"loss": 0.1446,
|
555 |
+
"step": 2125
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 7.38,
|
559 |
+
"learning_rate": 3.35e-07,
|
560 |
+
"loss": 0.1429,
|
561 |
+
"step": 2150
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 7.46,
|
565 |
+
"learning_rate": 3.325e-07,
|
566 |
+
"loss": 0.1444,
|
567 |
+
"step": 2175
|
568 |
+
},
|
569 |
+
{
|
570 |
+
"epoch": 7.55,
|
571 |
+
"learning_rate": 3.3e-07,
|
572 |
+
"loss": 0.1414,
|
573 |
+
"step": 2200
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 7.63,
|
577 |
+
"learning_rate": 3.275e-07,
|
578 |
+
"loss": 0.1395,
|
579 |
+
"step": 2225
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 7.72,
|
583 |
+
"learning_rate": 3.25e-07,
|
584 |
+
"loss": 0.1447,
|
585 |
+
"step": 2250
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 7.8,
|
589 |
+
"learning_rate": 3.225e-07,
|
590 |
+
"loss": 0.1447,
|
591 |
+
"step": 2275
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 7.89,
|
595 |
+
"learning_rate": 3.2e-07,
|
596 |
+
"loss": 0.1464,
|
597 |
+
"step": 2300
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 7.98,
|
601 |
+
"learning_rate": 3.175e-07,
|
602 |
+
"loss": 0.1436,
|
603 |
+
"step": 2325
|
604 |
+
},
|
605 |
+
{
|
606 |
+
"epoch": 8.06,
|
607 |
+
"learning_rate": 3.1509999999999996e-07,
|
608 |
+
"loss": 0.1452,
|
609 |
+
"step": 2350
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 8.15,
|
613 |
+
"learning_rate": 3.126e-07,
|
614 |
+
"loss": 0.1378,
|
615 |
+
"step": 2375
|
616 |
+
},
|
617 |
+
{
|
618 |
+
"epoch": 8.23,
|
619 |
+
"learning_rate": 3.1009999999999995e-07,
|
620 |
+
"loss": 0.1423,
|
621 |
+
"step": 2400
|
622 |
+
},
|
623 |
+
{
|
624 |
+
"epoch": 8.32,
|
625 |
+
"learning_rate": 3.076e-07,
|
626 |
+
"loss": 0.1444,
|
627 |
+
"step": 2425
|
628 |
+
},
|
629 |
+
{
|
630 |
+
"epoch": 8.4,
|
631 |
+
"learning_rate": 3.0509999999999995e-07,
|
632 |
+
"loss": 0.1447,
|
633 |
+
"step": 2450
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 8.49,
|
637 |
+
"learning_rate": 3.0259999999999997e-07,
|
638 |
+
"loss": 0.1457,
|
639 |
+
"step": 2475
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 8.58,
|
643 |
+
"learning_rate": 3.0009999999999994e-07,
|
644 |
+
"loss": 0.1426,
|
645 |
+
"step": 2500
|
646 |
+
},
|
647 |
+
{
|
648 |
+
"epoch": 8.58,
|
649 |
+
"eval_loss": 0.2098465859889984,
|
650 |
+
"eval_runtime": 1886.03,
|
651 |
+
"eval_samples_per_second": 4.938,
|
652 |
+
"eval_steps_per_second": 0.155,
|
653 |
+
"eval_wer": 29.222034307803817,
|
654 |
+
"step": 2500
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 8.66,
|
658 |
+
"learning_rate": 2.9759999999999996e-07,
|
659 |
+
"loss": 0.1429,
|
660 |
+
"step": 2525
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 8.75,
|
664 |
+
"learning_rate": 2.9509999999999994e-07,
|
665 |
+
"loss": 0.1384,
|
666 |
+
"step": 2550
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 8.83,
|
670 |
+
"learning_rate": 2.926e-07,
|
671 |
+
"loss": 0.1372,
|
672 |
+
"step": 2575
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 8.92,
|
676 |
+
"learning_rate": 2.9010000000000004e-07,
|
677 |
+
"loss": 0.1446,
|
678 |
+
"step": 2600
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 9.01,
|
682 |
+
"learning_rate": 2.876e-07,
|
683 |
+
"loss": 0.1424,
|
684 |
+
"step": 2625
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 9.09,
|
688 |
+
"learning_rate": 2.8510000000000003e-07,
|
689 |
+
"loss": 0.1426,
|
690 |
+
"step": 2650
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 9.18,
|
694 |
+
"learning_rate": 2.826e-07,
|
695 |
+
"loss": 0.1433,
|
696 |
+
"step": 2675
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 9.26,
|
700 |
+
"learning_rate": 2.8010000000000003e-07,
|
701 |
+
"loss": 0.1413,
|
702 |
+
"step": 2700
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 9.35,
|
706 |
+
"learning_rate": 2.776e-07,
|
707 |
+
"loss": 0.1361,
|
708 |
+
"step": 2725
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 9.43,
|
712 |
+
"learning_rate": 2.751e-07,
|
713 |
+
"loss": 0.1418,
|
714 |
+
"step": 2750
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 9.52,
|
718 |
+
"learning_rate": 2.726e-07,
|
719 |
+
"loss": 0.1396,
|
720 |
+
"step": 2775
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 9.61,
|
724 |
+
"learning_rate": 2.701e-07,
|
725 |
+
"loss": 0.1389,
|
726 |
+
"step": 2800
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 9.69,
|
730 |
+
"learning_rate": 2.676e-07,
|
731 |
+
"loss": 0.1438,
|
732 |
+
"step": 2825
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 9.78,
|
736 |
+
"learning_rate": 2.651e-07,
|
737 |
+
"loss": 0.1424,
|
738 |
+
"step": 2850
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 9.86,
|
742 |
+
"learning_rate": 2.626e-07,
|
743 |
+
"loss": 0.1393,
|
744 |
+
"step": 2875
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 9.95,
|
748 |
+
"learning_rate": 2.601e-07,
|
749 |
+
"loss": 0.1441,
|
750 |
+
"step": 2900
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 10.03,
|
754 |
+
"learning_rate": 2.576e-07,
|
755 |
+
"loss": 0.1404,
|
756 |
+
"step": 2925
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 10.12,
|
760 |
+
"learning_rate": 2.551e-07,
|
761 |
+
"loss": 0.1349,
|
762 |
+
"step": 2950
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 10.21,
|
766 |
+
"learning_rate": 2.5259999999999997e-07,
|
767 |
+
"loss": 0.1409,
|
768 |
+
"step": 2975
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 10.29,
|
772 |
+
"learning_rate": 2.501e-07,
|
773 |
+
"loss": 0.139,
|
774 |
+
"step": 3000
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 10.29,
|
778 |
+
"eval_loss": 0.20934279263019562,
|
779 |
+
"eval_runtime": 1881.23,
|
780 |
+
"eval_samples_per_second": 4.951,
|
781 |
+
"eval_steps_per_second": 0.155,
|
782 |
+
"eval_wer": 29.107495995633226,
|
783 |
+
"step": 3000
|
784 |
+
},
|
785 |
+
{
|
786 |
+
"epoch": 10.38,
|
787 |
+
"learning_rate": 2.4759999999999997e-07,
|
788 |
+
"loss": 0.1379,
|
789 |
+
"step": 3025
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 10.46,
|
793 |
+
"learning_rate": 2.451e-07,
|
794 |
+
"loss": 0.1419,
|
795 |
+
"step": 3050
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 10.55,
|
799 |
+
"learning_rate": 2.426e-07,
|
800 |
+
"loss": 0.1382,
|
801 |
+
"step": 3075
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 10.63,
|
805 |
+
"learning_rate": 2.401e-07,
|
806 |
+
"loss": 0.1378,
|
807 |
+
"step": 3100
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 10.72,
|
811 |
+
"learning_rate": 2.3759999999999998e-07,
|
812 |
+
"loss": 0.1426,
|
813 |
+
"step": 3125
|
814 |
+
},
|
815 |
+
{
|
816 |
+
"epoch": 10.81,
|
817 |
+
"learning_rate": 2.3509999999999998e-07,
|
818 |
+
"loss": 0.1395,
|
819 |
+
"step": 3150
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 10.89,
|
823 |
+
"learning_rate": 2.3259999999999998e-07,
|
824 |
+
"loss": 0.145,
|
825 |
+
"step": 3175
|
826 |
+
},
|
827 |
+
{
|
828 |
+
"epoch": 10.98,
|
829 |
+
"learning_rate": 2.3009999999999998e-07,
|
830 |
+
"loss": 0.1437,
|
831 |
+
"step": 3200
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 11.06,
|
835 |
+
"learning_rate": 2.2759999999999997e-07,
|
836 |
+
"loss": 0.1384,
|
837 |
+
"step": 3225
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 11.15,
|
841 |
+
"learning_rate": 2.251e-07,
|
842 |
+
"loss": 0.1395,
|
843 |
+
"step": 3250
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 11.23,
|
847 |
+
"learning_rate": 2.226e-07,
|
848 |
+
"loss": 0.1383,
|
849 |
+
"step": 3275
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 11.32,
|
853 |
+
"learning_rate": 2.201e-07,
|
854 |
+
"loss": 0.1365,
|
855 |
+
"step": 3300
|
856 |
+
},
|
857 |
+
{
|
858 |
+
"epoch": 11.41,
|
859 |
+
"learning_rate": 2.176e-07,
|
860 |
+
"loss": 0.1356,
|
861 |
+
"step": 3325
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 11.49,
|
865 |
+
"learning_rate": 2.1510000000000001e-07,
|
866 |
+
"loss": 0.138,
|
867 |
+
"step": 3350
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 11.58,
|
871 |
+
"learning_rate": 2.126e-07,
|
872 |
+
"loss": 0.1432,
|
873 |
+
"step": 3375
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 11.66,
|
877 |
+
"learning_rate": 2.101e-07,
|
878 |
+
"loss": 0.1444,
|
879 |
+
"step": 3400
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 11.75,
|
883 |
+
"learning_rate": 2.076e-07,
|
884 |
+
"loss": 0.139,
|
885 |
+
"step": 3425
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 11.84,
|
889 |
+
"learning_rate": 2.051e-07,
|
890 |
+
"loss": 0.1393,
|
891 |
+
"step": 3450
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 11.92,
|
895 |
+
"learning_rate": 2.026e-07,
|
896 |
+
"loss": 0.1379,
|
897 |
+
"step": 3475
|
898 |
+
},
|
899 |
+
{
|
900 |
+
"epoch": 12.01,
|
901 |
+
"learning_rate": 2.001e-07,
|
902 |
+
"loss": 0.1373,
|
903 |
+
"step": 3500
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 12.01,
|
907 |
+
"eval_loss": 0.20872431993484497,
|
908 |
+
"eval_runtime": 1879.9357,
|
909 |
+
"eval_samples_per_second": 4.954,
|
910 |
+
"eval_steps_per_second": 0.155,
|
911 |
+
"eval_wer": 29.087809723228908,
|
912 |
+
"step": 3500
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 12.09,
|
916 |
+
"learning_rate": 1.976e-07,
|
917 |
+
"loss": 0.1383,
|
918 |
+
"step": 3525
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 12.18,
|
922 |
+
"learning_rate": 1.951e-07,
|
923 |
+
"loss": 0.1391,
|
924 |
+
"step": 3550
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 12.26,
|
928 |
+
"learning_rate": 1.926e-07,
|
929 |
+
"loss": 0.1397,
|
930 |
+
"step": 3575
|
931 |
+
},
|
932 |
+
{
|
933 |
+
"epoch": 12.35,
|
934 |
+
"learning_rate": 1.901e-07,
|
935 |
+
"loss": 0.1401,
|
936 |
+
"step": 3600
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 12.44,
|
940 |
+
"learning_rate": 1.8759999999999999e-07,
|
941 |
+
"loss": 0.1384,
|
942 |
+
"step": 3625
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 12.52,
|
946 |
+
"learning_rate": 1.8509999999999998e-07,
|
947 |
+
"loss": 0.1379,
|
948 |
+
"step": 3650
|
949 |
+
},
|
950 |
+
{
|
951 |
+
"epoch": 12.61,
|
952 |
+
"learning_rate": 1.826e-07,
|
953 |
+
"loss": 0.14,
|
954 |
+
"step": 3675
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 12.69,
|
958 |
+
"learning_rate": 1.801e-07,
|
959 |
+
"loss": 0.137,
|
960 |
+
"step": 3700
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 12.78,
|
964 |
+
"learning_rate": 1.776e-07,
|
965 |
+
"loss": 0.1368,
|
966 |
+
"step": 3725
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 12.86,
|
970 |
+
"learning_rate": 1.751e-07,
|
971 |
+
"loss": 0.1399,
|
972 |
+
"step": 3750
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 12.95,
|
976 |
+
"learning_rate": 1.726e-07,
|
977 |
+
"loss": 0.1365,
|
978 |
+
"step": 3775
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 13.04,
|
982 |
+
"learning_rate": 1.701e-07,
|
983 |
+
"loss": 0.1363,
|
984 |
+
"step": 3800
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 13.12,
|
988 |
+
"learning_rate": 1.676e-07,
|
989 |
+
"loss": 0.1401,
|
990 |
+
"step": 3825
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 13.21,
|
994 |
+
"learning_rate": 1.651e-07,
|
995 |
+
"loss": 0.1377,
|
996 |
+
"step": 3850
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 13.29,
|
1000 |
+
"learning_rate": 1.626e-07,
|
1001 |
+
"loss": 0.1388,
|
1002 |
+
"step": 3875
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 13.38,
|
1006 |
+
"learning_rate": 1.6009999999999998e-07,
|
1007 |
+
"loss": 0.1343,
|
1008 |
+
"step": 3900
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 13.46,
|
1012 |
+
"learning_rate": 1.5759999999999998e-07,
|
1013 |
+
"loss": 0.1392,
|
1014 |
+
"step": 3925
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 13.55,
|
1018 |
+
"learning_rate": 1.5509999999999998e-07,
|
1019 |
+
"loss": 0.1369,
|
1020 |
+
"step": 3950
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 13.64,
|
1024 |
+
"learning_rate": 1.526e-07,
|
1025 |
+
"loss": 0.1355,
|
1026 |
+
"step": 3975
|
1027 |
+
},
|
1028 |
+
{
|
1029 |
+
"epoch": 13.72,
|
1030 |
+
"learning_rate": 1.501e-07,
|
1031 |
+
"loss": 0.1362,
|
1032 |
+
"step": 4000
|
1033 |
+
},
|
1034 |
+
{
|
1035 |
+
"epoch": 13.72,
|
1036 |
+
"eval_loss": 0.20841579139232635,
|
1037 |
+
"eval_runtime": 1892.3942,
|
1038 |
+
"eval_samples_per_second": 4.922,
|
1039 |
+
"eval_steps_per_second": 0.154,
|
1040 |
+
"eval_wer": 28.976850733313647,
|
1041 |
+
"step": 4000
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 13.81,
|
1045 |
+
"learning_rate": 1.476e-07,
|
1046 |
+
"loss": 0.1383,
|
1047 |
+
"step": 4025
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 13.89,
|
1051 |
+
"learning_rate": 1.451e-07,
|
1052 |
+
"loss": 0.1403,
|
1053 |
+
"step": 4050
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 13.98,
|
1057 |
+
"learning_rate": 1.426e-07,
|
1058 |
+
"loss": 0.138,
|
1059 |
+
"step": 4075
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 14.07,
|
1063 |
+
"learning_rate": 1.401e-07,
|
1064 |
+
"loss": 0.1379,
|
1065 |
+
"step": 4100
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 14.15,
|
1069 |
+
"learning_rate": 1.376e-07,
|
1070 |
+
"loss": 0.1379,
|
1071 |
+
"step": 4125
|
1072 |
+
},
|
1073 |
+
{
|
1074 |
+
"epoch": 14.24,
|
1075 |
+
"learning_rate": 1.3509999999999999e-07,
|
1076 |
+
"loss": 0.1351,
|
1077 |
+
"step": 4150
|
1078 |
+
},
|
1079 |
+
{
|
1080 |
+
"epoch": 14.32,
|
1081 |
+
"learning_rate": 1.3259999999999998e-07,
|
1082 |
+
"loss": 0.1402,
|
1083 |
+
"step": 4175
|
1084 |
+
},
|
1085 |
+
{
|
1086 |
+
"epoch": 14.41,
|
1087 |
+
"learning_rate": 1.3009999999999998e-07,
|
1088 |
+
"loss": 0.1384,
|
1089 |
+
"step": 4200
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 14.49,
|
1093 |
+
"learning_rate": 1.2759999999999998e-07,
|
1094 |
+
"loss": 0.1344,
|
1095 |
+
"step": 4225
|
1096 |
+
},
|
1097 |
+
{
|
1098 |
+
"epoch": 14.58,
|
1099 |
+
"learning_rate": 1.2509999999999998e-07,
|
1100 |
+
"loss": 0.1415,
|
1101 |
+
"step": 4250
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 14.67,
|
1105 |
+
"learning_rate": 1.226e-07,
|
1106 |
+
"loss": 0.136,
|
1107 |
+
"step": 4275
|
1108 |
+
},
|
1109 |
+
{
|
1110 |
+
"epoch": 14.75,
|
1111 |
+
"learning_rate": 1.201e-07,
|
1112 |
+
"loss": 0.1365,
|
1113 |
+
"step": 4300
|
1114 |
+
},
|
1115 |
+
{
|
1116 |
+
"epoch": 14.84,
|
1117 |
+
"learning_rate": 1.176e-07,
|
1118 |
+
"loss": 0.1359,
|
1119 |
+
"step": 4325
|
1120 |
+
},
|
1121 |
+
{
|
1122 |
+
"epoch": 14.92,
|
1123 |
+
"learning_rate": 1.1509999999999999e-07,
|
1124 |
+
"loss": 0.1352,
|
1125 |
+
"step": 4350
|
1126 |
+
},
|
1127 |
+
{
|
1128 |
+
"epoch": 15.01,
|
1129 |
+
"learning_rate": 1.1269999999999998e-07,
|
1130 |
+
"loss": 0.1363,
|
1131 |
+
"step": 4375
|
1132 |
+
},
|
1133 |
+
{
|
1134 |
+
"epoch": 15.09,
|
1135 |
+
"learning_rate": 1.1020000000000001e-07,
|
1136 |
+
"loss": 0.1365,
|
1137 |
+
"step": 4400
|
1138 |
+
},
|
1139 |
+
{
|
1140 |
+
"epoch": 15.18,
|
1141 |
+
"learning_rate": 1.077e-07,
|
1142 |
+
"loss": 0.1369,
|
1143 |
+
"step": 4425
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 15.27,
|
1147 |
+
"learning_rate": 1.052e-07,
|
1148 |
+
"loss": 0.1397,
|
1149 |
+
"step": 4450
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 15.35,
|
1153 |
+
"learning_rate": 1.027e-07,
|
1154 |
+
"loss": 0.1373,
|
1155 |
+
"step": 4475
|
1156 |
+
},
|
1157 |
+
{
|
1158 |
+
"epoch": 15.44,
|
1159 |
+
"learning_rate": 1.002e-07,
|
1160 |
+
"loss": 0.1333,
|
1161 |
+
"step": 4500
|
1162 |
+
},
|
1163 |
+
{
|
1164 |
+
"epoch": 15.44,
|
1165 |
+
"eval_loss": 0.20814116299152374,
|
1166 |
+
"eval_runtime": 1898.1875,
|
1167 |
+
"eval_samples_per_second": 4.907,
|
1168 |
+
"eval_steps_per_second": 0.154,
|
1169 |
+
"eval_wer": 28.912870347999608,
|
1170 |
+
"step": 4500
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 15.52,
|
1174 |
+
"learning_rate": 9.77e-08,
|
1175 |
+
"loss": 0.139,
|
1176 |
+
"step": 4525
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 15.61,
|
1180 |
+
"learning_rate": 9.52e-08,
|
1181 |
+
"loss": 0.1354,
|
1182 |
+
"step": 4550
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 15.69,
|
1186 |
+
"learning_rate": 9.27e-08,
|
1187 |
+
"loss": 0.1356,
|
1188 |
+
"step": 4575
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 15.78,
|
1192 |
+
"learning_rate": 9.02e-08,
|
1193 |
+
"loss": 0.1392,
|
1194 |
+
"step": 4600
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 15.87,
|
1198 |
+
"learning_rate": 8.77e-08,
|
1199 |
+
"loss": 0.1369,
|
1200 |
+
"step": 4625
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 15.95,
|
1204 |
+
"learning_rate": 8.52e-08,
|
1205 |
+
"loss": 0.1352,
|
1206 |
+
"step": 4650
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 16.04,
|
1210 |
+
"learning_rate": 8.269999999999999e-08,
|
1211 |
+
"loss": 0.1337,
|
1212 |
+
"step": 4675
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 16.12,
|
1216 |
+
"learning_rate": 8.019999999999999e-08,
|
1217 |
+
"loss": 0.14,
|
1218 |
+
"step": 4700
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 16.21,
|
1222 |
+
"learning_rate": 7.77e-08,
|
1223 |
+
"loss": 0.1348,
|
1224 |
+
"step": 4725
|
1225 |
+
},
|
1226 |
+
{
|
1227 |
+
"epoch": 16.3,
|
1228 |
+
"learning_rate": 7.52e-08,
|
1229 |
+
"loss": 0.1363,
|
1230 |
+
"step": 4750
|
1231 |
+
},
|
1232 |
+
{
|
1233 |
+
"epoch": 16.38,
|
1234 |
+
"learning_rate": 7.27e-08,
|
1235 |
+
"loss": 0.1358,
|
1236 |
+
"step": 4775
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 16.47,
|
1240 |
+
"learning_rate": 7.019999999999999e-08,
|
1241 |
+
"loss": 0.136,
|
1242 |
+
"step": 4800
|
1243 |
+
},
|
1244 |
+
{
|
1245 |
+
"epoch": 16.55,
|
1246 |
+
"learning_rate": 6.769999999999999e-08,
|
1247 |
+
"loss": 0.1357,
|
1248 |
+
"step": 4825
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 16.64,
|
1252 |
+
"learning_rate": 6.519999999999999e-08,
|
1253 |
+
"loss": 0.1368,
|
1254 |
+
"step": 4850
|
1255 |
+
},
|
1256 |
+
{
|
1257 |
+
"epoch": 16.72,
|
1258 |
+
"learning_rate": 6.27e-08,
|
1259 |
+
"loss": 0.141,
|
1260 |
+
"step": 4875
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 16.81,
|
1264 |
+
"learning_rate": 6.02e-08,
|
1265 |
+
"loss": 0.1338,
|
1266 |
+
"step": 4900
|
1267 |
+
},
|
1268 |
+
{
|
1269 |
+
"epoch": 16.9,
|
1270 |
+
"learning_rate": 5.77e-08,
|
1271 |
+
"loss": 0.1371,
|
1272 |
+
"step": 4925
|
1273 |
+
},
|
1274 |
+
{
|
1275 |
+
"epoch": 16.98,
|
1276 |
+
"learning_rate": 5.52e-08,
|
1277 |
+
"loss": 0.1323,
|
1278 |
+
"step": 4950
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 17.07,
|
1282 |
+
"learning_rate": 5.2699999999999995e-08,
|
1283 |
+
"loss": 0.1344,
|
1284 |
+
"step": 4975
|
1285 |
+
},
|
1286 |
+
{
|
1287 |
+
"epoch": 17.15,
|
1288 |
+
"learning_rate": 5.02e-08,
|
1289 |
+
"loss": 0.1332,
|
1290 |
+
"step": 5000
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 17.15,
|
1294 |
+
"eval_loss": 0.207936629652977,
|
1295 |
+
"eval_runtime": 1896.8056,
|
1296 |
+
"eval_samples_per_second": 4.91,
|
1297 |
+
"eval_steps_per_second": 0.154,
|
1298 |
+
"eval_wer": 28.894526321441035,
|
1299 |
+
"step": 5000
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 17.24,
|
1303 |
+
"learning_rate": 4.77e-08,
|
1304 |
+
"loss": 0.1378,
|
1305 |
+
"step": 5025
|
1306 |
+
},
|
1307 |
+
{
|
1308 |
+
"epoch": 17.32,
|
1309 |
+
"learning_rate": 4.5199999999999994e-08,
|
1310 |
+
"loss": 0.1385,
|
1311 |
+
"step": 5050
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 17.41,
|
1315 |
+
"learning_rate": 4.27e-08,
|
1316 |
+
"loss": 0.1367,
|
1317 |
+
"step": 5075
|
1318 |
+
},
|
1319 |
+
{
|
1320 |
+
"epoch": 17.5,
|
1321 |
+
"learning_rate": 4.0199999999999996e-08,
|
1322 |
+
"loss": 0.135,
|
1323 |
+
"step": 5100
|
1324 |
+
},
|
1325 |
+
{
|
1326 |
+
"epoch": 17.58,
|
1327 |
+
"learning_rate": 3.7699999999999993e-08,
|
1328 |
+
"loss": 0.1326,
|
1329 |
+
"step": 5125
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 17.67,
|
1333 |
+
"learning_rate": 3.52e-08,
|
1334 |
+
"loss": 0.1342,
|
1335 |
+
"step": 5150
|
1336 |
+
},
|
1337 |
+
{
|
1338 |
+
"epoch": 17.75,
|
1339 |
+
"learning_rate": 3.27e-08,
|
1340 |
+
"loss": 0.137,
|
1341 |
+
"step": 5175
|
1342 |
+
},
|
1343 |
+
{
|
1344 |
+
"epoch": 17.84,
|
1345 |
+
"learning_rate": 3.02e-08,
|
1346 |
+
"loss": 0.1393,
|
1347 |
+
"step": 5200
|
1348 |
+
},
|
1349 |
+
{
|
1350 |
+
"epoch": 17.92,
|
1351 |
+
"learning_rate": 2.7699999999999997e-08,
|
1352 |
+
"loss": 0.1349,
|
1353 |
+
"step": 5225
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 18.01,
|
1357 |
+
"learning_rate": 2.52e-08,
|
1358 |
+
"loss": 0.1363,
|
1359 |
+
"step": 5250
|
1360 |
+
},
|
1361 |
+
{
|
1362 |
+
"epoch": 18.1,
|
1363 |
+
"learning_rate": 2.27e-08,
|
1364 |
+
"loss": 0.1393,
|
1365 |
+
"step": 5275
|
1366 |
+
},
|
1367 |
+
{
|
1368 |
+
"epoch": 18.18,
|
1369 |
+
"learning_rate": 2.02e-08,
|
1370 |
+
"loss": 0.1352,
|
1371 |
+
"step": 5300
|
1372 |
+
},
|
1373 |
+
{
|
1374 |
+
"epoch": 18.27,
|
1375 |
+
"learning_rate": 1.77e-08,
|
1376 |
+
"loss": 0.1348,
|
1377 |
+
"step": 5325
|
1378 |
+
},
|
1379 |
+
{
|
1380 |
+
"epoch": 18.35,
|
1381 |
+
"learning_rate": 1.52e-08,
|
1382 |
+
"loss": 0.1352,
|
1383 |
+
"step": 5350
|
1384 |
+
},
|
1385 |
+
{
|
1386 |
+
"epoch": 18.44,
|
1387 |
+
"learning_rate": 1.27e-08,
|
1388 |
+
"loss": 0.1335,
|
1389 |
+
"step": 5375
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 18.52,
|
1393 |
+
"learning_rate": 1.02e-08,
|
1394 |
+
"loss": 0.132,
|
1395 |
+
"step": 5400
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 18.61,
|
1399 |
+
"learning_rate": 7.7e-09,
|
1400 |
+
"loss": 0.1379,
|
1401 |
+
"step": 5425
|
1402 |
+
},
|
1403 |
+
{
|
1404 |
+
"epoch": 18.7,
|
1405 |
+
"learning_rate": 5.1999999999999994e-09,
|
1406 |
+
"loss": 0.1374,
|
1407 |
+
"step": 5450
|
1408 |
+
},
|
1409 |
+
{
|
1410 |
+
"epoch": 18.78,
|
1411 |
+
"learning_rate": 2.7e-09,
|
1412 |
+
"loss": 0.1368,
|
1413 |
+
"step": 5475
|
1414 |
+
},
|
1415 |
+
{
|
1416 |
+
"epoch": 18.87,
|
1417 |
+
"learning_rate": 2e-10,
|
1418 |
+
"loss": 0.1363,
|
1419 |
+
"step": 5500
|
1420 |
+
},
|
1421 |
+
{
|
1422 |
+
"epoch": 18.87,
|
1423 |
+
"eval_loss": 0.20784465968608856,
|
1424 |
+
"eval_runtime": 1895.0895,
|
1425 |
+
"eval_samples_per_second": 4.915,
|
1426 |
+
"eval_steps_per_second": 0.154,
|
1427 |
+
"eval_wer": 28.818465723515253,
|
1428 |
+
"step": 5500
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 18.87,
|
1432 |
+
"step": 5500,
|
1433 |
+
"total_flos": 2.279802921025536e+19,
|
1434 |
+
"train_loss": 0.14273875054446133,
|
1435 |
+
"train_runtime": 37524.2555,
|
1436 |
+
"train_samples_per_second": 9.381,
|
1437 |
+
"train_steps_per_second": 0.147
|
1438 |
+
}
|
1439 |
+
],
|
1440 |
+
"logging_steps": 25,
|
1441 |
+
"max_steps": 5500,
|
1442 |
+
"num_input_tokens_seen": 0,
|
1443 |
+
"num_train_epochs": 19,
|
1444 |
+
"save_steps": 500,
|
1445 |
+
"total_flos": 2.279802921025536e+19,
|
1446 |
+
"train_batch_size": 32,
|
1447 |
+
"trial_name": null,
|
1448 |
+
"trial_params": null
|
1449 |
+
}
|