rishabhjain16
commited on
Commit
•
1955041
1
Parent(s):
596eada
End of training
Browse files- README.md +17 -4
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Feb20_19-32-51_6abce58e1145/events.out.tfevents.1708508512.6abce58e1145.1924.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +1320 -0
README.md
CHANGED
@@ -2,20 +2,33 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: openai/whisper-large-v3
|
4 |
tags:
|
|
|
5 |
- generated_from_trainer
|
|
|
|
|
6 |
metrics:
|
7 |
- wer
|
8 |
model-index:
|
9 |
-
- name:
|
10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
---
|
12 |
|
13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
should probably proofread and complete it, then remove this comment. -->
|
15 |
|
16 |
-
#
|
17 |
|
18 |
-
This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on
|
19 |
It achieves the following results on the evaluation set:
|
20 |
- Loss: 0.3189
|
21 |
- Wer: 23.8109
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: openai/whisper-large-v3
|
4 |
tags:
|
5 |
+
- whisper-event
|
6 |
- generated_from_trainer
|
7 |
+
datasets:
|
8 |
+
- rishabhjain16/kaggle_albanian
|
9 |
metrics:
|
10 |
- wer
|
11 |
model-index:
|
12 |
+
- name: Whisper large V3 to Kaggle Albanian
|
13 |
+
results:
|
14 |
+
- task:
|
15 |
+
name: Automatic Speech Recognition
|
16 |
+
type: automatic-speech-recognition
|
17 |
+
dataset:
|
18 |
+
name: rishabhjain16/kaggle_albanian default
|
19 |
+
type: rishabhjain16/kaggle_albanian
|
20 |
+
metrics:
|
21 |
+
- name: Wer
|
22 |
+
type: wer
|
23 |
+
value: 23.81094676029683
|
24 |
---
|
25 |
|
26 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
27 |
should probably proofread and complete it, then remove this comment. -->
|
28 |
|
29 |
+
# Whisper large V3 to Kaggle Albanian
|
30 |
|
31 |
+
This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the rishabhjain16/kaggle_albanian default dataset.
|
32 |
It achieves the following results on the evaluation set:
|
33 |
- Loss: 0.3189
|
34 |
- Wer: 23.8109
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.18,
|
3 |
+
"eval_loss": 0.31885361671447754,
|
4 |
+
"eval_runtime": 2125.9841,
|
5 |
+
"eval_samples_per_second": 3.535,
|
6 |
+
"eval_steps_per_second": 0.221,
|
7 |
+
"eval_wer": 23.81094676029683,
|
8 |
+
"train_loss": 0.2540892813682556,
|
9 |
+
"train_runtime": 48626.5038,
|
10 |
+
"train_samples_per_second": 1.645,
|
11 |
+
"train_steps_per_second": 0.103
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.18,
|
3 |
+
"eval_loss": 0.31885361671447754,
|
4 |
+
"eval_runtime": 2125.9841,
|
5 |
+
"eval_samples_per_second": 3.535,
|
6 |
+
"eval_steps_per_second": 0.221,
|
7 |
+
"eval_wer": 23.81094676029683
|
8 |
+
}
|
runs/Feb20_19-32-51_6abce58e1145/events.out.tfevents.1708508512.6abce58e1145.1924.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7d3b853ccc8a656a7094821673eadc9523847e6f83efbc1dab7472c1c4488955
|
3 |
+
size 358
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.18,
|
3 |
+
"train_loss": 0.2540892813682556,
|
4 |
+
"train_runtime": 48626.5038,
|
5 |
+
"train_samples_per_second": 1.645,
|
6 |
+
"train_steps_per_second": 0.103
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1320 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 23.81094676029683,
|
3 |
+
"best_model_checkpoint": "./checkpoint-5000",
|
4 |
+
"epoch": 2.1788,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 5000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01,
|
13 |
+
"learning_rate": 4.6000000000000004e-07,
|
14 |
+
"loss": 1.7155,
|
15 |
+
"step": 25
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.01,
|
19 |
+
"learning_rate": 9.600000000000001e-07,
|
20 |
+
"loss": 1.2833,
|
21 |
+
"step": 50
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.01,
|
25 |
+
"learning_rate": 1.46e-06,
|
26 |
+
"loss": 0.6549,
|
27 |
+
"step": 75
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.02,
|
31 |
+
"learning_rate": 1.9600000000000003e-06,
|
32 |
+
"loss": 0.4167,
|
33 |
+
"step": 100
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.03,
|
37 |
+
"learning_rate": 2.46e-06,
|
38 |
+
"loss": 0.3397,
|
39 |
+
"step": 125
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.03,
|
43 |
+
"learning_rate": 2.96e-06,
|
44 |
+
"loss": 0.3208,
|
45 |
+
"step": 150
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.04,
|
49 |
+
"learning_rate": 3.46e-06,
|
50 |
+
"loss": 0.2839,
|
51 |
+
"step": 175
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.04,
|
55 |
+
"learning_rate": 3.96e-06,
|
56 |
+
"loss": 0.2953,
|
57 |
+
"step": 200
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.04,
|
61 |
+
"learning_rate": 4.4600000000000005e-06,
|
62 |
+
"loss": 0.3841,
|
63 |
+
"step": 225
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.05,
|
67 |
+
"learning_rate": 4.960000000000001e-06,
|
68 |
+
"loss": 0.416,
|
69 |
+
"step": 250
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.06,
|
73 |
+
"learning_rate": 5.460000000000001e-06,
|
74 |
+
"loss": 0.3926,
|
75 |
+
"step": 275
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.06,
|
79 |
+
"learning_rate": 5.9600000000000005e-06,
|
80 |
+
"loss": 0.3496,
|
81 |
+
"step": 300
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.07,
|
85 |
+
"learning_rate": 6.460000000000001e-06,
|
86 |
+
"loss": 0.4149,
|
87 |
+
"step": 325
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.07,
|
91 |
+
"learning_rate": 6.96e-06,
|
92 |
+
"loss": 0.482,
|
93 |
+
"step": 350
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.07,
|
97 |
+
"learning_rate": 7.4600000000000006e-06,
|
98 |
+
"loss": 0.4613,
|
99 |
+
"step": 375
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.08,
|
103 |
+
"learning_rate": 7.960000000000002e-06,
|
104 |
+
"loss": 0.4522,
|
105 |
+
"step": 400
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.09,
|
109 |
+
"learning_rate": 8.46e-06,
|
110 |
+
"loss": 0.4463,
|
111 |
+
"step": 425
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.09,
|
115 |
+
"learning_rate": 8.96e-06,
|
116 |
+
"loss": 0.3219,
|
117 |
+
"step": 450
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 0.1,
|
121 |
+
"learning_rate": 9.460000000000001e-06,
|
122 |
+
"loss": 0.3347,
|
123 |
+
"step": 475
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.1,
|
127 |
+
"learning_rate": 9.960000000000001e-06,
|
128 |
+
"loss": 0.2778,
|
129 |
+
"step": 500
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.1,
|
133 |
+
"eval_loss": 0.49452078342437744,
|
134 |
+
"eval_runtime": 2170.0908,
|
135 |
+
"eval_samples_per_second": 3.463,
|
136 |
+
"eval_steps_per_second": 0.217,
|
137 |
+
"eval_wer": 36.98391354151103,
|
138 |
+
"step": 500
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 0.1,
|
142 |
+
"learning_rate": 9.94888888888889e-06,
|
143 |
+
"loss": 0.2984,
|
144 |
+
"step": 525
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.11,
|
148 |
+
"learning_rate": 9.893333333333334e-06,
|
149 |
+
"loss": 0.3115,
|
150 |
+
"step": 550
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.12,
|
154 |
+
"learning_rate": 9.837777777777778e-06,
|
155 |
+
"loss": 0.2502,
|
156 |
+
"step": 575
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.12,
|
160 |
+
"learning_rate": 9.782222222222222e-06,
|
161 |
+
"loss": 0.2514,
|
162 |
+
"step": 600
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 0.12,
|
166 |
+
"learning_rate": 9.726666666666668e-06,
|
167 |
+
"loss": 0.3575,
|
168 |
+
"step": 625
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 0.13,
|
172 |
+
"learning_rate": 9.671111111111112e-06,
|
173 |
+
"loss": 0.4231,
|
174 |
+
"step": 650
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.14,
|
178 |
+
"learning_rate": 9.615555555555558e-06,
|
179 |
+
"loss": 0.4043,
|
180 |
+
"step": 675
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 0.14,
|
184 |
+
"learning_rate": 9.56e-06,
|
185 |
+
"loss": 0.4514,
|
186 |
+
"step": 700
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 0.14,
|
190 |
+
"learning_rate": 9.504444444444446e-06,
|
191 |
+
"loss": 0.4511,
|
192 |
+
"step": 725
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 0.15,
|
196 |
+
"learning_rate": 9.44888888888889e-06,
|
197 |
+
"loss": 0.4803,
|
198 |
+
"step": 750
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.15,
|
202 |
+
"learning_rate": 9.393333333333334e-06,
|
203 |
+
"loss": 0.4437,
|
204 |
+
"step": 775
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.16,
|
208 |
+
"learning_rate": 9.33777777777778e-06,
|
209 |
+
"loss": 0.4146,
|
210 |
+
"step": 800
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 0.17,
|
214 |
+
"learning_rate": 9.282222222222222e-06,
|
215 |
+
"loss": 0.4221,
|
216 |
+
"step": 825
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 0.17,
|
220 |
+
"learning_rate": 9.226666666666668e-06,
|
221 |
+
"loss": 0.424,
|
222 |
+
"step": 850
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 0.17,
|
226 |
+
"learning_rate": 9.171111111111112e-06,
|
227 |
+
"loss": 0.4471,
|
228 |
+
"step": 875
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 0.18,
|
232 |
+
"learning_rate": 9.115555555555556e-06,
|
233 |
+
"loss": 0.4358,
|
234 |
+
"step": 900
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 0.18,
|
238 |
+
"learning_rate": 9.060000000000001e-06,
|
239 |
+
"loss": 0.4,
|
240 |
+
"step": 925
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.19,
|
244 |
+
"learning_rate": 9.004444444444445e-06,
|
245 |
+
"loss": 0.4205,
|
246 |
+
"step": 950
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 0.2,
|
250 |
+
"learning_rate": 8.94888888888889e-06,
|
251 |
+
"loss": 0.4146,
|
252 |
+
"step": 975
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 0.2,
|
256 |
+
"learning_rate": 8.893333333333333e-06,
|
257 |
+
"loss": 0.4343,
|
258 |
+
"step": 1000
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 0.2,
|
262 |
+
"eval_loss": 0.42902088165283203,
|
263 |
+
"eval_runtime": 2158.1828,
|
264 |
+
"eval_samples_per_second": 3.483,
|
265 |
+
"eval_steps_per_second": 0.218,
|
266 |
+
"eval_wer": 35.183525075949994,
|
267 |
+
"step": 1000
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 0.2,
|
271 |
+
"learning_rate": 8.83777777777778e-06,
|
272 |
+
"loss": 0.418,
|
273 |
+
"step": 1025
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 0.21,
|
277 |
+
"learning_rate": 8.782222222222223e-06,
|
278 |
+
"loss": 0.4076,
|
279 |
+
"step": 1050
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 0.21,
|
283 |
+
"learning_rate": 8.726666666666667e-06,
|
284 |
+
"loss": 0.4059,
|
285 |
+
"step": 1075
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 0.22,
|
289 |
+
"learning_rate": 8.671111111111113e-06,
|
290 |
+
"loss": 0.3988,
|
291 |
+
"step": 1100
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 0.23,
|
295 |
+
"learning_rate": 8.615555555555555e-06,
|
296 |
+
"loss": 0.3707,
|
297 |
+
"step": 1125
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 0.23,
|
301 |
+
"learning_rate": 8.560000000000001e-06,
|
302 |
+
"loss": 0.3797,
|
303 |
+
"step": 1150
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.23,
|
307 |
+
"learning_rate": 8.504444444444445e-06,
|
308 |
+
"loss": 0.3647,
|
309 |
+
"step": 1175
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 0.24,
|
313 |
+
"learning_rate": 8.448888888888889e-06,
|
314 |
+
"loss": 0.3669,
|
315 |
+
"step": 1200
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 0.24,
|
319 |
+
"learning_rate": 8.393333333333335e-06,
|
320 |
+
"loss": 0.3469,
|
321 |
+
"step": 1225
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 0.25,
|
325 |
+
"learning_rate": 8.337777777777777e-06,
|
326 |
+
"loss": 0.336,
|
327 |
+
"step": 1250
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 0.26,
|
331 |
+
"learning_rate": 8.282222222222223e-06,
|
332 |
+
"loss": 0.34,
|
333 |
+
"step": 1275
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 0.26,
|
337 |
+
"learning_rate": 8.226666666666667e-06,
|
338 |
+
"loss": 0.3328,
|
339 |
+
"step": 1300
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 0.27,
|
343 |
+
"learning_rate": 8.171111111111113e-06,
|
344 |
+
"loss": 0.3497,
|
345 |
+
"step": 1325
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.27,
|
349 |
+
"learning_rate": 8.115555555555557e-06,
|
350 |
+
"loss": 0.3747,
|
351 |
+
"step": 1350
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 0.28,
|
355 |
+
"learning_rate": 8.06e-06,
|
356 |
+
"loss": 0.3966,
|
357 |
+
"step": 1375
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 0.28,
|
361 |
+
"learning_rate": 8.004444444444445e-06,
|
362 |
+
"loss": 0.3475,
|
363 |
+
"step": 1400
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 0.28,
|
367 |
+
"learning_rate": 7.948888888888889e-06,
|
368 |
+
"loss": 0.3441,
|
369 |
+
"step": 1425
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 0.29,
|
373 |
+
"learning_rate": 7.893333333333335e-06,
|
374 |
+
"loss": 0.3357,
|
375 |
+
"step": 1450
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 0.29,
|
379 |
+
"learning_rate": 7.837777777777779e-06,
|
380 |
+
"loss": 0.378,
|
381 |
+
"step": 1475
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 0.3,
|
385 |
+
"learning_rate": 7.782222222222223e-06,
|
386 |
+
"loss": 0.3523,
|
387 |
+
"step": 1500
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.3,
|
391 |
+
"eval_loss": 0.3780156672000885,
|
392 |
+
"eval_runtime": 2149.1405,
|
393 |
+
"eval_samples_per_second": 3.497,
|
394 |
+
"eval_steps_per_second": 0.219,
|
395 |
+
"eval_wer": 30.651426863887643,
|
396 |
+
"step": 1500
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 0.3,
|
400 |
+
"learning_rate": 7.726666666666667e-06,
|
401 |
+
"loss": 0.3923,
|
402 |
+
"step": 1525
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 0.31,
|
406 |
+
"learning_rate": 7.67111111111111e-06,
|
407 |
+
"loss": 0.3449,
|
408 |
+
"step": 1550
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.32,
|
412 |
+
"learning_rate": 7.6155555555555564e-06,
|
413 |
+
"loss": 0.3714,
|
414 |
+
"step": 1575
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 0.32,
|
418 |
+
"learning_rate": 7.5600000000000005e-06,
|
419 |
+
"loss": 0.4006,
|
420 |
+
"step": 1600
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 0.33,
|
424 |
+
"learning_rate": 7.504444444444445e-06,
|
425 |
+
"loss": 0.3612,
|
426 |
+
"step": 1625
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 0.33,
|
430 |
+
"learning_rate": 7.44888888888889e-06,
|
431 |
+
"loss": 0.3639,
|
432 |
+
"step": 1650
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 0.34,
|
436 |
+
"learning_rate": 7.393333333333333e-06,
|
437 |
+
"loss": 0.4153,
|
438 |
+
"step": 1675
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 0.34,
|
442 |
+
"learning_rate": 7.337777777777778e-06,
|
443 |
+
"loss": 0.3818,
|
444 |
+
"step": 1700
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 0.34,
|
448 |
+
"learning_rate": 7.282222222222222e-06,
|
449 |
+
"loss": 0.3564,
|
450 |
+
"step": 1725
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.35,
|
454 |
+
"learning_rate": 7.226666666666667e-06,
|
455 |
+
"loss": 0.3629,
|
456 |
+
"step": 1750
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 0.35,
|
460 |
+
"learning_rate": 7.171111111111112e-06,
|
461 |
+
"loss": 0.2898,
|
462 |
+
"step": 1775
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 0.36,
|
466 |
+
"learning_rate": 7.115555555555557e-06,
|
467 |
+
"loss": 0.3372,
|
468 |
+
"step": 1800
|
469 |
+
},
|
470 |
+
{
|
471 |
+
"epoch": 0.36,
|
472 |
+
"learning_rate": 7.06e-06,
|
473 |
+
"loss": 0.3539,
|
474 |
+
"step": 1825
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 0.37,
|
478 |
+
"learning_rate": 7.004444444444445e-06,
|
479 |
+
"loss": 0.3791,
|
480 |
+
"step": 1850
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 0.38,
|
484 |
+
"learning_rate": 6.948888888888889e-06,
|
485 |
+
"loss": 0.3461,
|
486 |
+
"step": 1875
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 0.38,
|
490 |
+
"learning_rate": 6.893333333333334e-06,
|
491 |
+
"loss": 0.3498,
|
492 |
+
"step": 1900
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.39,
|
496 |
+
"learning_rate": 6.837777777777779e-06,
|
497 |
+
"loss": 0.3502,
|
498 |
+
"step": 1925
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 0.39,
|
502 |
+
"learning_rate": 6.782222222222222e-06,
|
503 |
+
"loss": 0.3696,
|
504 |
+
"step": 1950
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 0.4,
|
508 |
+
"learning_rate": 6.726666666666667e-06,
|
509 |
+
"loss": 0.3524,
|
510 |
+
"step": 1975
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 0.4,
|
514 |
+
"learning_rate": 6.671111111111112e-06,
|
515 |
+
"loss": 0.3833,
|
516 |
+
"step": 2000
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 0.4,
|
520 |
+
"eval_loss": 0.343627005815506,
|
521 |
+
"eval_runtime": 2165.5496,
|
522 |
+
"eval_samples_per_second": 3.471,
|
523 |
+
"eval_steps_per_second": 0.217,
|
524 |
+
"eval_wer": 28.031774490761492,
|
525 |
+
"step": 2000
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 0.41,
|
529 |
+
"learning_rate": 6.615555555555556e-06,
|
530 |
+
"loss": 0.3653,
|
531 |
+
"step": 2025
|
532 |
+
},
|
533 |
+
{
|
534 |
+
"epoch": 0.41,
|
535 |
+
"learning_rate": 6.560000000000001e-06,
|
536 |
+
"loss": 0.3649,
|
537 |
+
"step": 2050
|
538 |
+
},
|
539 |
+
{
|
540 |
+
"epoch": 1.0,
|
541 |
+
"learning_rate": 6.504444444444446e-06,
|
542 |
+
"loss": 0.3044,
|
543 |
+
"step": 2075
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 1.01,
|
547 |
+
"learning_rate": 6.448888888888889e-06,
|
548 |
+
"loss": 0.2725,
|
549 |
+
"step": 2100
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 1.01,
|
553 |
+
"learning_rate": 6.393333333333334e-06,
|
554 |
+
"loss": 0.2116,
|
555 |
+
"step": 2125
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 1.02,
|
559 |
+
"learning_rate": 6.3377777777777786e-06,
|
560 |
+
"loss": 0.1644,
|
561 |
+
"step": 2150
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 1.02,
|
565 |
+
"learning_rate": 6.282222222222223e-06,
|
566 |
+
"loss": 0.1579,
|
567 |
+
"step": 2175
|
568 |
+
},
|
569 |
+
{
|
570 |
+
"epoch": 1.03,
|
571 |
+
"learning_rate": 6.2266666666666675e-06,
|
572 |
+
"loss": 0.1648,
|
573 |
+
"step": 2200
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 1.03,
|
577 |
+
"learning_rate": 6.171111111111112e-06,
|
578 |
+
"loss": 0.1634,
|
579 |
+
"step": 2225
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 1.04,
|
583 |
+
"learning_rate": 6.1155555555555555e-06,
|
584 |
+
"loss": 0.1412,
|
585 |
+
"step": 2250
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 1.04,
|
589 |
+
"learning_rate": 6.0600000000000004e-06,
|
590 |
+
"loss": 0.2006,
|
591 |
+
"step": 2275
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 1.05,
|
595 |
+
"learning_rate": 6.004444444444445e-06,
|
596 |
+
"loss": 0.2172,
|
597 |
+
"step": 2300
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 1.05,
|
601 |
+
"learning_rate": 5.948888888888889e-06,
|
602 |
+
"loss": 0.1989,
|
603 |
+
"step": 2325
|
604 |
+
},
|
605 |
+
{
|
606 |
+
"epoch": 1.06,
|
607 |
+
"learning_rate": 5.893333333333334e-06,
|
608 |
+
"loss": 0.2036,
|
609 |
+
"step": 2350
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 1.06,
|
613 |
+
"learning_rate": 5.837777777777777e-06,
|
614 |
+
"loss": 0.1612,
|
615 |
+
"step": 2375
|
616 |
+
},
|
617 |
+
{
|
618 |
+
"epoch": 1.07,
|
619 |
+
"learning_rate": 5.782222222222222e-06,
|
620 |
+
"loss": 0.192,
|
621 |
+
"step": 2400
|
622 |
+
},
|
623 |
+
{
|
624 |
+
"epoch": 1.07,
|
625 |
+
"learning_rate": 5.726666666666667e-06,
|
626 |
+
"loss": 0.1787,
|
627 |
+
"step": 2425
|
628 |
+
},
|
629 |
+
{
|
630 |
+
"epoch": 1.08,
|
631 |
+
"learning_rate": 5.671111111111112e-06,
|
632 |
+
"loss": 0.1886,
|
633 |
+
"step": 2450
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 1.08,
|
637 |
+
"learning_rate": 5.615555555555556e-06,
|
638 |
+
"loss": 0.1996,
|
639 |
+
"step": 2475
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 1.09,
|
643 |
+
"learning_rate": 5.560000000000001e-06,
|
644 |
+
"loss": 0.2029,
|
645 |
+
"step": 2500
|
646 |
+
},
|
647 |
+
{
|
648 |
+
"epoch": 1.09,
|
649 |
+
"eval_loss": 0.33604511618614197,
|
650 |
+
"eval_runtime": 2142.8094,
|
651 |
+
"eval_samples_per_second": 3.508,
|
652 |
+
"eval_steps_per_second": 0.219,
|
653 |
+
"eval_wer": 26.51277454056477,
|
654 |
+
"step": 2500
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 1.09,
|
658 |
+
"learning_rate": 5.504444444444444e-06,
|
659 |
+
"loss": 0.2064,
|
660 |
+
"step": 2525
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 1.1,
|
664 |
+
"learning_rate": 5.448888888888889e-06,
|
665 |
+
"loss": 0.1988,
|
666 |
+
"step": 2550
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 1.1,
|
670 |
+
"learning_rate": 5.393333333333334e-06,
|
671 |
+
"loss": 0.1774,
|
672 |
+
"step": 2575
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 1.11,
|
676 |
+
"learning_rate": 5.337777777777779e-06,
|
677 |
+
"loss": 0.2277,
|
678 |
+
"step": 2600
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 1.11,
|
682 |
+
"learning_rate": 5.282222222222223e-06,
|
683 |
+
"loss": 0.1804,
|
684 |
+
"step": 2625
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 1.12,
|
688 |
+
"learning_rate": 5.226666666666667e-06,
|
689 |
+
"loss": 0.2029,
|
690 |
+
"step": 2650
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 1.12,
|
694 |
+
"learning_rate": 5.171111111111111e-06,
|
695 |
+
"loss": 0.169,
|
696 |
+
"step": 2675
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 1.13,
|
700 |
+
"learning_rate": 5.115555555555556e-06,
|
701 |
+
"loss": 0.1985,
|
702 |
+
"step": 2700
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 1.13,
|
706 |
+
"learning_rate": 5.060000000000001e-06,
|
707 |
+
"loss": 0.2063,
|
708 |
+
"step": 2725
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 1.14,
|
712 |
+
"learning_rate": 5.004444444444445e-06,
|
713 |
+
"loss": 0.2117,
|
714 |
+
"step": 2750
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 1.14,
|
718 |
+
"learning_rate": 4.94888888888889e-06,
|
719 |
+
"loss": 0.2109,
|
720 |
+
"step": 2775
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 1.15,
|
724 |
+
"learning_rate": 4.893333333333334e-06,
|
725 |
+
"loss": 0.163,
|
726 |
+
"step": 2800
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 1.15,
|
730 |
+
"learning_rate": 4.837777777777778e-06,
|
731 |
+
"loss": 0.1834,
|
732 |
+
"step": 2825
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 1.16,
|
736 |
+
"learning_rate": 4.7822222222222226e-06,
|
737 |
+
"loss": 0.196,
|
738 |
+
"step": 2850
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 1.16,
|
742 |
+
"learning_rate": 4.7266666666666674e-06,
|
743 |
+
"loss": 0.1746,
|
744 |
+
"step": 2875
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 1.17,
|
748 |
+
"learning_rate": 4.6711111111111115e-06,
|
749 |
+
"loss": 0.1844,
|
750 |
+
"step": 2900
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 1.17,
|
754 |
+
"learning_rate": 4.6155555555555555e-06,
|
755 |
+
"loss": 0.1548,
|
756 |
+
"step": 2925
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 1.18,
|
760 |
+
"learning_rate": 4.56e-06,
|
761 |
+
"loss": 0.1546,
|
762 |
+
"step": 2950
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 1.18,
|
766 |
+
"learning_rate": 4.504444444444444e-06,
|
767 |
+
"loss": 0.1539,
|
768 |
+
"step": 2975
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 1.19,
|
772 |
+
"learning_rate": 4.448888888888889e-06,
|
773 |
+
"loss": 0.1797,
|
774 |
+
"step": 3000
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 1.19,
|
778 |
+
"eval_loss": 0.3350614607334137,
|
779 |
+
"eval_runtime": 2157.9989,
|
780 |
+
"eval_samples_per_second": 3.483,
|
781 |
+
"eval_steps_per_second": 0.218,
|
782 |
+
"eval_wer": 25.892723741222174,
|
783 |
+
"step": 3000
|
784 |
+
},
|
785 |
+
{
|
786 |
+
"epoch": 1.19,
|
787 |
+
"learning_rate": 4.393333333333334e-06,
|
788 |
+
"loss": 0.1737,
|
789 |
+
"step": 3025
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 1.2,
|
793 |
+
"learning_rate": 4.337777777777778e-06,
|
794 |
+
"loss": 0.1736,
|
795 |
+
"step": 3050
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 1.2,
|
799 |
+
"learning_rate": 4.282222222222222e-06,
|
800 |
+
"loss": 0.2304,
|
801 |
+
"step": 3075
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 1.21,
|
805 |
+
"learning_rate": 4.226666666666667e-06,
|
806 |
+
"loss": 0.2147,
|
807 |
+
"step": 3100
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 1.21,
|
811 |
+
"learning_rate": 4.171111111111111e-06,
|
812 |
+
"loss": 0.2228,
|
813 |
+
"step": 3125
|
814 |
+
},
|
815 |
+
{
|
816 |
+
"epoch": 1.22,
|
817 |
+
"learning_rate": 4.115555555555556e-06,
|
818 |
+
"loss": 0.238,
|
819 |
+
"step": 3150
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 1.22,
|
823 |
+
"learning_rate": 4.060000000000001e-06,
|
824 |
+
"loss": 0.1769,
|
825 |
+
"step": 3175
|
826 |
+
},
|
827 |
+
{
|
828 |
+
"epoch": 1.23,
|
829 |
+
"learning_rate": 4.004444444444445e-06,
|
830 |
+
"loss": 0.1293,
|
831 |
+
"step": 3200
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 1.23,
|
835 |
+
"learning_rate": 3.948888888888889e-06,
|
836 |
+
"loss": 0.1206,
|
837 |
+
"step": 3225
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 1.24,
|
841 |
+
"learning_rate": 3.893333333333333e-06,
|
842 |
+
"loss": 0.1162,
|
843 |
+
"step": 3250
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 1.24,
|
847 |
+
"learning_rate": 3.837777777777778e-06,
|
848 |
+
"loss": 0.1242,
|
849 |
+
"step": 3275
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 1.25,
|
853 |
+
"learning_rate": 3.782222222222223e-06,
|
854 |
+
"loss": 0.1193,
|
855 |
+
"step": 3300
|
856 |
+
},
|
857 |
+
{
|
858 |
+
"epoch": 1.25,
|
859 |
+
"learning_rate": 3.726666666666667e-06,
|
860 |
+
"loss": 0.0961,
|
861 |
+
"step": 3325
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 1.26,
|
865 |
+
"learning_rate": 3.6711111111111113e-06,
|
866 |
+
"loss": 0.1158,
|
867 |
+
"step": 3350
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 1.26,
|
871 |
+
"learning_rate": 3.615555555555556e-06,
|
872 |
+
"loss": 0.1642,
|
873 |
+
"step": 3375
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 1.27,
|
877 |
+
"learning_rate": 3.5600000000000002e-06,
|
878 |
+
"loss": 0.1731,
|
879 |
+
"step": 3400
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 1.27,
|
883 |
+
"learning_rate": 3.5044444444444447e-06,
|
884 |
+
"loss": 0.1629,
|
885 |
+
"step": 3425
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 1.28,
|
889 |
+
"learning_rate": 3.4488888888888896e-06,
|
890 |
+
"loss": 0.1607,
|
891 |
+
"step": 3450
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 1.28,
|
895 |
+
"learning_rate": 3.3933333333333336e-06,
|
896 |
+
"loss": 0.1553,
|
897 |
+
"step": 3475
|
898 |
+
},
|
899 |
+
{
|
900 |
+
"epoch": 1.29,
|
901 |
+
"learning_rate": 3.337777777777778e-06,
|
902 |
+
"loss": 0.1687,
|
903 |
+
"step": 3500
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 1.29,
|
907 |
+
"eval_loss": 0.32494643330574036,
|
908 |
+
"eval_runtime": 2166.1529,
|
909 |
+
"eval_samples_per_second": 3.47,
|
910 |
+
"eval_steps_per_second": 0.217,
|
911 |
+
"eval_wer": 25.862841774988794,
|
912 |
+
"step": 3500
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 1.29,
|
916 |
+
"learning_rate": 3.282222222222223e-06,
|
917 |
+
"loss": 0.1774,
|
918 |
+
"step": 3525
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 1.3,
|
922 |
+
"learning_rate": 3.226666666666667e-06,
|
923 |
+
"loss": 0.1959,
|
924 |
+
"step": 3550
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 1.3,
|
928 |
+
"learning_rate": 3.1711111111111114e-06,
|
929 |
+
"loss": 0.163,
|
930 |
+
"step": 3575
|
931 |
+
},
|
932 |
+
{
|
933 |
+
"epoch": 1.31,
|
934 |
+
"learning_rate": 3.1155555555555555e-06,
|
935 |
+
"loss": 0.1777,
|
936 |
+
"step": 3600
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 1.31,
|
940 |
+
"learning_rate": 3.0600000000000003e-06,
|
941 |
+
"loss": 0.1728,
|
942 |
+
"step": 3625
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 1.32,
|
946 |
+
"learning_rate": 3.004444444444445e-06,
|
947 |
+
"loss": 0.1662,
|
948 |
+
"step": 3650
|
949 |
+
},
|
950 |
+
{
|
951 |
+
"epoch": 1.32,
|
952 |
+
"learning_rate": 2.948888888888889e-06,
|
953 |
+
"loss": 0.1561,
|
954 |
+
"step": 3675
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 1.33,
|
958 |
+
"learning_rate": 2.8933333333333337e-06,
|
959 |
+
"loss": 0.1787,
|
960 |
+
"step": 3700
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 1.33,
|
964 |
+
"learning_rate": 2.837777777777778e-06,
|
965 |
+
"loss": 0.1754,
|
966 |
+
"step": 3725
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 1.34,
|
970 |
+
"learning_rate": 2.7822222222222222e-06,
|
971 |
+
"loss": 0.1773,
|
972 |
+
"step": 3750
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 1.34,
|
976 |
+
"learning_rate": 2.726666666666667e-06,
|
977 |
+
"loss": 0.1892,
|
978 |
+
"step": 3775
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 1.35,
|
982 |
+
"learning_rate": 2.6711111111111116e-06,
|
983 |
+
"loss": 0.1488,
|
984 |
+
"step": 3800
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 1.35,
|
988 |
+
"learning_rate": 2.6155555555555556e-06,
|
989 |
+
"loss": 0.1608,
|
990 |
+
"step": 3825
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 1.36,
|
994 |
+
"learning_rate": 2.56e-06,
|
995 |
+
"loss": 0.1294,
|
996 |
+
"step": 3850
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 1.36,
|
1000 |
+
"learning_rate": 2.504444444444445e-06,
|
1001 |
+
"loss": 0.1756,
|
1002 |
+
"step": 3875
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 1.37,
|
1006 |
+
"learning_rate": 2.448888888888889e-06,
|
1007 |
+
"loss": 0.1413,
|
1008 |
+
"step": 3900
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 1.37,
|
1012 |
+
"learning_rate": 2.3933333333333334e-06,
|
1013 |
+
"loss": 0.1489,
|
1014 |
+
"step": 3925
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 1.38,
|
1018 |
+
"learning_rate": 2.337777777777778e-06,
|
1019 |
+
"loss": 0.1581,
|
1020 |
+
"step": 3950
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 1.38,
|
1024 |
+
"learning_rate": 2.2822222222222223e-06,
|
1025 |
+
"loss": 0.1448,
|
1026 |
+
"step": 3975
|
1027 |
+
},
|
1028 |
+
{
|
1029 |
+
"epoch": 1.39,
|
1030 |
+
"learning_rate": 2.226666666666667e-06,
|
1031 |
+
"loss": 0.1603,
|
1032 |
+
"step": 4000
|
1033 |
+
},
|
1034 |
+
{
|
1035 |
+
"epoch": 1.39,
|
1036 |
+
"eval_loss": 0.3088971972465515,
|
1037 |
+
"eval_runtime": 2143.1525,
|
1038 |
+
"eval_samples_per_second": 3.507,
|
1039 |
+
"eval_steps_per_second": 0.219,
|
1040 |
+
"eval_wer": 23.987748393844317,
|
1041 |
+
"step": 4000
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 1.39,
|
1045 |
+
"learning_rate": 2.1711111111111113e-06,
|
1046 |
+
"loss": 0.1934,
|
1047 |
+
"step": 4025
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 1.4,
|
1051 |
+
"learning_rate": 2.1155555555555557e-06,
|
1052 |
+
"loss": 0.1749,
|
1053 |
+
"step": 4050
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 1.4,
|
1057 |
+
"learning_rate": 2.06e-06,
|
1058 |
+
"loss": 0.1772,
|
1059 |
+
"step": 4075
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 1.41,
|
1063 |
+
"learning_rate": 2.0044444444444446e-06,
|
1064 |
+
"loss": 0.2196,
|
1065 |
+
"step": 4100
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 2.0,
|
1069 |
+
"learning_rate": 1.948888888888889e-06,
|
1070 |
+
"loss": 0.1208,
|
1071 |
+
"step": 4125
|
1072 |
+
},
|
1073 |
+
{
|
1074 |
+
"epoch": 2.01,
|
1075 |
+
"learning_rate": 1.8933333333333333e-06,
|
1076 |
+
"loss": 0.0793,
|
1077 |
+
"step": 4150
|
1078 |
+
},
|
1079 |
+
{
|
1080 |
+
"epoch": 2.01,
|
1081 |
+
"learning_rate": 1.837777777777778e-06,
|
1082 |
+
"loss": 0.0843,
|
1083 |
+
"step": 4175
|
1084 |
+
},
|
1085 |
+
{
|
1086 |
+
"epoch": 2.02,
|
1087 |
+
"learning_rate": 1.7822222222222225e-06,
|
1088 |
+
"loss": 0.0769,
|
1089 |
+
"step": 4200
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 2.02,
|
1093 |
+
"learning_rate": 1.7266666666666667e-06,
|
1094 |
+
"loss": 0.0777,
|
1095 |
+
"step": 4225
|
1096 |
+
},
|
1097 |
+
{
|
1098 |
+
"epoch": 2.03,
|
1099 |
+
"learning_rate": 1.6711111111111112e-06,
|
1100 |
+
"loss": 0.086,
|
1101 |
+
"step": 4250
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 2.03,
|
1105 |
+
"learning_rate": 1.6155555555555559e-06,
|
1106 |
+
"loss": 0.0766,
|
1107 |
+
"step": 4275
|
1108 |
+
},
|
1109 |
+
{
|
1110 |
+
"epoch": 2.04,
|
1111 |
+
"learning_rate": 1.56e-06,
|
1112 |
+
"loss": 0.0885,
|
1113 |
+
"step": 4300
|
1114 |
+
},
|
1115 |
+
{
|
1116 |
+
"epoch": 2.04,
|
1117 |
+
"learning_rate": 1.5044444444444446e-06,
|
1118 |
+
"loss": 0.0778,
|
1119 |
+
"step": 4325
|
1120 |
+
},
|
1121 |
+
{
|
1122 |
+
"epoch": 2.05,
|
1123 |
+
"learning_rate": 1.4488888888888892e-06,
|
1124 |
+
"loss": 0.088,
|
1125 |
+
"step": 4350
|
1126 |
+
},
|
1127 |
+
{
|
1128 |
+
"epoch": 2.05,
|
1129 |
+
"learning_rate": 1.3933333333333335e-06,
|
1130 |
+
"loss": 0.0715,
|
1131 |
+
"step": 4375
|
1132 |
+
},
|
1133 |
+
{
|
1134 |
+
"epoch": 2.06,
|
1135 |
+
"learning_rate": 1.337777777777778e-06,
|
1136 |
+
"loss": 0.0683,
|
1137 |
+
"step": 4400
|
1138 |
+
},
|
1139 |
+
{
|
1140 |
+
"epoch": 2.06,
|
1141 |
+
"learning_rate": 1.2822222222222222e-06,
|
1142 |
+
"loss": 0.0714,
|
1143 |
+
"step": 4425
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 2.07,
|
1147 |
+
"learning_rate": 1.2266666666666666e-06,
|
1148 |
+
"loss": 0.0788,
|
1149 |
+
"step": 4450
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 2.07,
|
1153 |
+
"learning_rate": 1.171111111111111e-06,
|
1154 |
+
"loss": 0.0792,
|
1155 |
+
"step": 4475
|
1156 |
+
},
|
1157 |
+
{
|
1158 |
+
"epoch": 2.08,
|
1159 |
+
"learning_rate": 1.1155555555555558e-06,
|
1160 |
+
"loss": 0.0762,
|
1161 |
+
"step": 4500
|
1162 |
+
},
|
1163 |
+
{
|
1164 |
+
"epoch": 2.08,
|
1165 |
+
"eval_loss": 0.3345198631286621,
|
1166 |
+
"eval_runtime": 2127.0754,
|
1167 |
+
"eval_samples_per_second": 3.533,
|
1168 |
+
"eval_steps_per_second": 0.221,
|
1169 |
+
"eval_wer": 24.580407390806318,
|
1170 |
+
"step": 4500
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 2.08,
|
1174 |
+
"learning_rate": 1.06e-06,
|
1175 |
+
"loss": 0.0786,
|
1176 |
+
"step": 4525
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 2.09,
|
1180 |
+
"learning_rate": 1.0044444444444445e-06,
|
1181 |
+
"loss": 0.0932,
|
1182 |
+
"step": 4550
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 2.09,
|
1186 |
+
"learning_rate": 9.488888888888889e-07,
|
1187 |
+
"loss": 0.102,
|
1188 |
+
"step": 4575
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 2.1,
|
1192 |
+
"learning_rate": 8.933333333333334e-07,
|
1193 |
+
"loss": 0.0689,
|
1194 |
+
"step": 4600
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 2.1,
|
1198 |
+
"learning_rate": 8.37777777777778e-07,
|
1199 |
+
"loss": 0.0841,
|
1200 |
+
"step": 4625
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 2.11,
|
1204 |
+
"learning_rate": 7.822222222222223e-07,
|
1205 |
+
"loss": 0.077,
|
1206 |
+
"step": 4650
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 2.11,
|
1210 |
+
"learning_rate": 7.266666666666668e-07,
|
1211 |
+
"loss": 0.0685,
|
1212 |
+
"step": 4675
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 2.12,
|
1216 |
+
"learning_rate": 6.711111111111111e-07,
|
1217 |
+
"loss": 0.0774,
|
1218 |
+
"step": 4700
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 2.12,
|
1222 |
+
"learning_rate": 6.155555555555556e-07,
|
1223 |
+
"loss": 0.0784,
|
1224 |
+
"step": 4725
|
1225 |
+
},
|
1226 |
+
{
|
1227 |
+
"epoch": 2.13,
|
1228 |
+
"learning_rate": 5.6e-07,
|
1229 |
+
"loss": 0.0741,
|
1230 |
+
"step": 4750
|
1231 |
+
},
|
1232 |
+
{
|
1233 |
+
"epoch": 2.13,
|
1234 |
+
"learning_rate": 5.044444444444445e-07,
|
1235 |
+
"loss": 0.0779,
|
1236 |
+
"step": 4775
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 2.14,
|
1240 |
+
"learning_rate": 4.488888888888889e-07,
|
1241 |
+
"loss": 0.1124,
|
1242 |
+
"step": 4800
|
1243 |
+
},
|
1244 |
+
{
|
1245 |
+
"epoch": 2.14,
|
1246 |
+
"learning_rate": 3.9333333333333336e-07,
|
1247 |
+
"loss": 0.1224,
|
1248 |
+
"step": 4825
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 2.15,
|
1252 |
+
"learning_rate": 3.3777777777777777e-07,
|
1253 |
+
"loss": 0.1299,
|
1254 |
+
"step": 4850
|
1255 |
+
},
|
1256 |
+
{
|
1257 |
+
"epoch": 2.15,
|
1258 |
+
"learning_rate": 2.822222222222222e-07,
|
1259 |
+
"loss": 0.0879,
|
1260 |
+
"step": 4875
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 2.16,
|
1264 |
+
"learning_rate": 2.266666666666667e-07,
|
1265 |
+
"loss": 0.0581,
|
1266 |
+
"step": 4900
|
1267 |
+
},
|
1268 |
+
{
|
1269 |
+
"epoch": 2.16,
|
1270 |
+
"learning_rate": 1.7111111111111114e-07,
|
1271 |
+
"loss": 0.0656,
|
1272 |
+
"step": 4925
|
1273 |
+
},
|
1274 |
+
{
|
1275 |
+
"epoch": 2.17,
|
1276 |
+
"learning_rate": 1.1555555555555556e-07,
|
1277 |
+
"loss": 0.0622,
|
1278 |
+
"step": 4950
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 2.17,
|
1282 |
+
"learning_rate": 6.000000000000001e-08,
|
1283 |
+
"loss": 0.0549,
|
1284 |
+
"step": 4975
|
1285 |
+
},
|
1286 |
+
{
|
1287 |
+
"epoch": 2.18,
|
1288 |
+
"learning_rate": 4.444444444444445e-09,
|
1289 |
+
"loss": 0.0676,
|
1290 |
+
"step": 5000
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 2.18,
|
1294 |
+
"eval_loss": 0.31885361671447754,
|
1295 |
+
"eval_runtime": 2126.4706,
|
1296 |
+
"eval_samples_per_second": 3.534,
|
1297 |
+
"eval_steps_per_second": 0.221,
|
1298 |
+
"eval_wer": 23.81094676029683,
|
1299 |
+
"step": 5000
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 2.18,
|
1303 |
+
"step": 5000,
|
1304 |
+
"total_flos": 2.717998645248e+20,
|
1305 |
+
"train_loss": 0.2540892813682556,
|
1306 |
+
"train_runtime": 48626.5038,
|
1307 |
+
"train_samples_per_second": 1.645,
|
1308 |
+
"train_steps_per_second": 0.103
|
1309 |
+
}
|
1310 |
+
],
|
1311 |
+
"logging_steps": 25,
|
1312 |
+
"max_steps": 5000,
|
1313 |
+
"num_input_tokens_seen": 0,
|
1314 |
+
"num_train_epochs": 9223372036854775807,
|
1315 |
+
"save_steps": 1000,
|
1316 |
+
"total_flos": 2.717998645248e+20,
|
1317 |
+
"train_batch_size": 16,
|
1318 |
+
"trial_name": null,
|
1319 |
+
"trial_params": null
|
1320 |
+
}
|