End of training
Browse files- README.md +12 -9
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Apr15_16-30-25_ip-172-31-23-125/events.out.tfevents.1713220995.ip-172-31-23-125.1113.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +675 -0
README.md
CHANGED
|
@@ -1,39 +1,42 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: openai/whisper-small
|
| 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: vi
|
| 20 |
split: test
|
| 21 |
args: vi
|
| 22 |
metrics:
|
| 23 |
- name: Wer
|
| 24 |
type: wer
|
| 25 |
-
value:
|
| 26 |
---
|
| 27 |
|
| 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 [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the common_voice_16_0 dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Wer:
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 1 |
---
|
| 2 |
+
language:
|
| 3 |
+
- vi
|
| 4 |
license: apache-2.0
|
| 5 |
base_model: openai/whisper-small
|
| 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 Small Vienamese
|
| 15 |
results:
|
| 16 |
- task:
|
| 17 |
name: Automatic Speech Recognition
|
| 18 |
type: automatic-speech-recognition
|
| 19 |
dataset:
|
| 20 |
+
name: mozilla-foundation/common_voice_16_0 vi
|
| 21 |
+
type: mozilla-foundation/common_voice_16_0
|
| 22 |
config: vi
|
| 23 |
split: test
|
| 24 |
args: vi
|
| 25 |
metrics:
|
| 26 |
- name: Wer
|
| 27 |
type: wer
|
| 28 |
+
value: 24.56800162634682
|
| 29 |
---
|
| 30 |
|
| 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 Small Vienamese
|
| 35 |
|
| 36 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the mozilla-foundation/common_voice_16_0 vi dataset.
|
| 37 |
It achieves the following results on the evaluation set:
|
| 38 |
+
- Loss: 0.6705
|
| 39 |
+
- Wer: 24.5680
|
| 40 |
|
| 41 |
## Model description
|
| 42 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 166.0,
|
| 3 |
+
"eval_loss": 0.6704666614532471,
|
| 4 |
+
"eval_runtime": 250.7091,
|
| 5 |
+
"eval_samples_per_second": 5.289,
|
| 6 |
+
"eval_steps_per_second": 0.168,
|
| 7 |
+
"eval_wer": 24.56800162634682,
|
| 8 |
+
"train_loss": 0.05099657071977854,
|
| 9 |
+
"train_runtime": 22043.5536,
|
| 10 |
+
"train_samples_per_second": 7.258,
|
| 11 |
+
"train_steps_per_second": 0.113
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 166.0,
|
| 3 |
+
"eval_loss": 0.6704666614532471,
|
| 4 |
+
"eval_runtime": 250.7091,
|
| 5 |
+
"eval_samples_per_second": 5.289,
|
| 6 |
+
"eval_steps_per_second": 0.168,
|
| 7 |
+
"eval_wer": 24.56800162634682
|
| 8 |
+
}
|
runs/Apr15_16-30-25_ip-172-31-23-125/events.out.tfevents.1713220995.ip-172-31-23-125.1113.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b04ecf7868397b6c0a6c5d7a8964c310f03e6bd6e445f6a3af83083275b708d3
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 166.0,
|
| 3 |
+
"train_loss": 0.05099657071977854,
|
| 4 |
+
"train_runtime": 22043.5536,
|
| 5 |
+
"train_samples_per_second": 7.258,
|
| 6 |
+
"train_steps_per_second": 0.113
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,675 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 24.56800162634682,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-1000",
|
| 4 |
+
"epoch": 166.004,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"learning_rate": 4.6e-07,
|
| 14 |
+
"loss": 1.2564,
|
| 15 |
+
"step": 25
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 3.0,
|
| 19 |
+
"learning_rate": 9.6e-07,
|
| 20 |
+
"loss": 1.0031,
|
| 21 |
+
"step": 50
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 4.01,
|
| 25 |
+
"learning_rate": 9.906122448979592e-07,
|
| 26 |
+
"loss": 0.5375,
|
| 27 |
+
"step": 75
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 6.0,
|
| 31 |
+
"learning_rate": 9.80408163265306e-07,
|
| 32 |
+
"loss": 0.3396,
|
| 33 |
+
"step": 100
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 8.0,
|
| 37 |
+
"learning_rate": 9.70204081632653e-07,
|
| 38 |
+
"loss": 0.2754,
|
| 39 |
+
"step": 125
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 9.01,
|
| 43 |
+
"learning_rate": 9.6e-07,
|
| 44 |
+
"loss": 0.2336,
|
| 45 |
+
"step": 150
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 11.0,
|
| 49 |
+
"learning_rate": 9.497959183673469e-07,
|
| 50 |
+
"loss": 0.197,
|
| 51 |
+
"step": 175
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 13.0,
|
| 55 |
+
"learning_rate": 9.395918367346939e-07,
|
| 56 |
+
"loss": 0.1695,
|
| 57 |
+
"step": 200
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 14.01,
|
| 61 |
+
"learning_rate": 9.293877551020408e-07,
|
| 62 |
+
"loss": 0.1478,
|
| 63 |
+
"step": 225
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 16.0,
|
| 67 |
+
"learning_rate": 9.191836734693877e-07,
|
| 68 |
+
"loss": 0.1242,
|
| 69 |
+
"step": 250
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 18.0,
|
| 73 |
+
"learning_rate": 9.089795918367347e-07,
|
| 74 |
+
"loss": 0.1056,
|
| 75 |
+
"step": 275
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 19.01,
|
| 79 |
+
"learning_rate": 8.987755102040815e-07,
|
| 80 |
+
"loss": 0.0889,
|
| 81 |
+
"step": 300
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 21.0,
|
| 85 |
+
"learning_rate": 8.885714285714285e-07,
|
| 86 |
+
"loss": 0.0733,
|
| 87 |
+
"step": 325
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 23.0,
|
| 91 |
+
"learning_rate": 8.783673469387755e-07,
|
| 92 |
+
"loss": 0.0592,
|
| 93 |
+
"step": 350
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 24.01,
|
| 97 |
+
"learning_rate": 8.681632653061224e-07,
|
| 98 |
+
"loss": 0.0502,
|
| 99 |
+
"step": 375
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 26.0,
|
| 103 |
+
"learning_rate": 8.579591836734693e-07,
|
| 104 |
+
"loss": 0.0383,
|
| 105 |
+
"step": 400
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 28.0,
|
| 109 |
+
"learning_rate": 8.477551020408162e-07,
|
| 110 |
+
"loss": 0.0313,
|
| 111 |
+
"step": 425
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 29.01,
|
| 115 |
+
"learning_rate": 8.375510204081633e-07,
|
| 116 |
+
"loss": 0.0244,
|
| 117 |
+
"step": 450
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 31.0,
|
| 121 |
+
"learning_rate": 8.273469387755102e-07,
|
| 122 |
+
"loss": 0.0199,
|
| 123 |
+
"step": 475
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 33.0,
|
| 127 |
+
"learning_rate": 8.171428571428571e-07,
|
| 128 |
+
"loss": 0.0174,
|
| 129 |
+
"step": 500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 33.0,
|
| 133 |
+
"eval_loss": 0.6206670999526978,
|
| 134 |
+
"eval_runtime": 252.4938,
|
| 135 |
+
"eval_samples_per_second": 5.252,
|
| 136 |
+
"eval_steps_per_second": 0.166,
|
| 137 |
+
"eval_wer": 24.66964830250051,
|
| 138 |
+
"step": 500
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 34.01,
|
| 142 |
+
"learning_rate": 8.06938775510204e-07,
|
| 143 |
+
"loss": 0.015,
|
| 144 |
+
"step": 525
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 36.0,
|
| 148 |
+
"learning_rate": 7.96734693877551e-07,
|
| 149 |
+
"loss": 0.0134,
|
| 150 |
+
"step": 550
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 38.0,
|
| 154 |
+
"learning_rate": 7.86530612244898e-07,
|
| 155 |
+
"loss": 0.012,
|
| 156 |
+
"step": 575
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 39.01,
|
| 160 |
+
"learning_rate": 7.763265306122448e-07,
|
| 161 |
+
"loss": 0.011,
|
| 162 |
+
"step": 600
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 41.0,
|
| 166 |
+
"learning_rate": 7.661224489795918e-07,
|
| 167 |
+
"loss": 0.01,
|
| 168 |
+
"step": 625
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 43.0,
|
| 172 |
+
"learning_rate": 7.559183673469387e-07,
|
| 173 |
+
"loss": 0.0092,
|
| 174 |
+
"step": 650
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 44.01,
|
| 178 |
+
"learning_rate": 7.457142857142856e-07,
|
| 179 |
+
"loss": 0.0087,
|
| 180 |
+
"step": 675
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 46.0,
|
| 184 |
+
"learning_rate": 7.355102040816326e-07,
|
| 185 |
+
"loss": 0.0081,
|
| 186 |
+
"step": 700
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 48.0,
|
| 190 |
+
"learning_rate": 7.253061224489796e-07,
|
| 191 |
+
"loss": 0.0075,
|
| 192 |
+
"step": 725
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 49.01,
|
| 196 |
+
"learning_rate": 7.151020408163266e-07,
|
| 197 |
+
"loss": 0.0072,
|
| 198 |
+
"step": 750
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 51.0,
|
| 202 |
+
"learning_rate": 7.048979591836734e-07,
|
| 203 |
+
"loss": 0.0068,
|
| 204 |
+
"step": 775
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 53.0,
|
| 208 |
+
"learning_rate": 6.946938775510204e-07,
|
| 209 |
+
"loss": 0.0064,
|
| 210 |
+
"step": 800
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 54.01,
|
| 214 |
+
"learning_rate": 6.844897959183673e-07,
|
| 215 |
+
"loss": 0.0061,
|
| 216 |
+
"step": 825
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 56.0,
|
| 220 |
+
"learning_rate": 6.742857142857143e-07,
|
| 221 |
+
"loss": 0.0058,
|
| 222 |
+
"step": 850
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 58.0,
|
| 226 |
+
"learning_rate": 6.640816326530611e-07,
|
| 227 |
+
"loss": 0.0055,
|
| 228 |
+
"step": 875
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 59.01,
|
| 232 |
+
"learning_rate": 6.538775510204081e-07,
|
| 233 |
+
"loss": 0.0054,
|
| 234 |
+
"step": 900
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 61.0,
|
| 238 |
+
"learning_rate": 6.43673469387755e-07,
|
| 239 |
+
"loss": 0.0051,
|
| 240 |
+
"step": 925
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 63.0,
|
| 244 |
+
"learning_rate": 6.334693877551021e-07,
|
| 245 |
+
"loss": 0.0049,
|
| 246 |
+
"step": 950
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 64.01,
|
| 250 |
+
"learning_rate": 6.23265306122449e-07,
|
| 251 |
+
"loss": 0.0047,
|
| 252 |
+
"step": 975
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 66.0,
|
| 256 |
+
"learning_rate": 6.130612244897959e-07,
|
| 257 |
+
"loss": 0.0045,
|
| 258 |
+
"step": 1000
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 66.0,
|
| 262 |
+
"eval_loss": 0.6704666614532471,
|
| 263 |
+
"eval_runtime": 250.1153,
|
| 264 |
+
"eval_samples_per_second": 5.302,
|
| 265 |
+
"eval_steps_per_second": 0.168,
|
| 266 |
+
"eval_wer": 24.56800162634682,
|
| 267 |
+
"step": 1000
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 68.0,
|
| 271 |
+
"learning_rate": 6.028571428571429e-07,
|
| 272 |
+
"loss": 0.0044,
|
| 273 |
+
"step": 1025
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 69.01,
|
| 277 |
+
"learning_rate": 5.926530612244898e-07,
|
| 278 |
+
"loss": 0.0042,
|
| 279 |
+
"step": 1050
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 71.0,
|
| 283 |
+
"learning_rate": 5.824489795918367e-07,
|
| 284 |
+
"loss": 0.0041,
|
| 285 |
+
"step": 1075
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 73.0,
|
| 289 |
+
"learning_rate": 5.722448979591836e-07,
|
| 290 |
+
"loss": 0.004,
|
| 291 |
+
"step": 1100
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 74.01,
|
| 295 |
+
"learning_rate": 5.620408163265306e-07,
|
| 296 |
+
"loss": 0.0038,
|
| 297 |
+
"step": 1125
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 76.0,
|
| 301 |
+
"learning_rate": 5.518367346938774e-07,
|
| 302 |
+
"loss": 0.0037,
|
| 303 |
+
"step": 1150
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 78.0,
|
| 307 |
+
"learning_rate": 5.416326530612245e-07,
|
| 308 |
+
"loss": 0.0036,
|
| 309 |
+
"step": 1175
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 79.01,
|
| 313 |
+
"learning_rate": 5.314285714285715e-07,
|
| 314 |
+
"loss": 0.0035,
|
| 315 |
+
"step": 1200
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 81.0,
|
| 319 |
+
"learning_rate": 5.212244897959184e-07,
|
| 320 |
+
"loss": 0.0034,
|
| 321 |
+
"step": 1225
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 83.0,
|
| 325 |
+
"learning_rate": 5.110204081632653e-07,
|
| 326 |
+
"loss": 0.0033,
|
| 327 |
+
"step": 1250
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 84.01,
|
| 331 |
+
"learning_rate": 5.008163265306122e-07,
|
| 332 |
+
"loss": 0.0033,
|
| 333 |
+
"step": 1275
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 86.0,
|
| 337 |
+
"learning_rate": 4.906122448979592e-07,
|
| 338 |
+
"loss": 0.0031,
|
| 339 |
+
"step": 1300
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 88.0,
|
| 343 |
+
"learning_rate": 4.804081632653061e-07,
|
| 344 |
+
"loss": 0.0031,
|
| 345 |
+
"step": 1325
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 89.01,
|
| 349 |
+
"learning_rate": 4.7020408163265303e-07,
|
| 350 |
+
"loss": 0.0031,
|
| 351 |
+
"step": 1350
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 91.0,
|
| 355 |
+
"learning_rate": 4.6e-07,
|
| 356 |
+
"loss": 0.003,
|
| 357 |
+
"step": 1375
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 93.0,
|
| 361 |
+
"learning_rate": 4.4979591836734695e-07,
|
| 362 |
+
"loss": 0.0029,
|
| 363 |
+
"step": 1400
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 94.01,
|
| 367 |
+
"learning_rate": 4.3959183673469385e-07,
|
| 368 |
+
"loss": 0.0029,
|
| 369 |
+
"step": 1425
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 96.0,
|
| 373 |
+
"learning_rate": 4.293877551020408e-07,
|
| 374 |
+
"loss": 0.0028,
|
| 375 |
+
"step": 1450
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 98.0,
|
| 379 |
+
"learning_rate": 4.191836734693877e-07,
|
| 380 |
+
"loss": 0.0027,
|
| 381 |
+
"step": 1475
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 99.01,
|
| 385 |
+
"learning_rate": 4.089795918367347e-07,
|
| 386 |
+
"loss": 0.0027,
|
| 387 |
+
"step": 1500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 99.01,
|
| 391 |
+
"eval_loss": 0.6944558024406433,
|
| 392 |
+
"eval_runtime": 263.7584,
|
| 393 |
+
"eval_samples_per_second": 5.027,
|
| 394 |
+
"eval_steps_per_second": 0.159,
|
| 395 |
+
"eval_wer": 25.27952835942265,
|
| 396 |
+
"step": 1500
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 101.0,
|
| 400 |
+
"learning_rate": 3.987755102040816e-07,
|
| 401 |
+
"loss": 0.0026,
|
| 402 |
+
"step": 1525
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 103.0,
|
| 406 |
+
"learning_rate": 3.8857142857142853e-07,
|
| 407 |
+
"loss": 0.0026,
|
| 408 |
+
"step": 1550
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 104.01,
|
| 412 |
+
"learning_rate": 3.783673469387755e-07,
|
| 413 |
+
"loss": 0.0025,
|
| 414 |
+
"step": 1575
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 106.0,
|
| 418 |
+
"learning_rate": 3.681632653061224e-07,
|
| 419 |
+
"loss": 0.0025,
|
| 420 |
+
"step": 1600
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 108.0,
|
| 424 |
+
"learning_rate": 3.579591836734694e-07,
|
| 425 |
+
"loss": 0.0024,
|
| 426 |
+
"step": 1625
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 109.01,
|
| 430 |
+
"learning_rate": 3.477551020408163e-07,
|
| 431 |
+
"loss": 0.0024,
|
| 432 |
+
"step": 1650
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 111.0,
|
| 436 |
+
"learning_rate": 3.3755102040816326e-07,
|
| 437 |
+
"loss": 0.0023,
|
| 438 |
+
"step": 1675
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 113.0,
|
| 442 |
+
"learning_rate": 3.2734693877551016e-07,
|
| 443 |
+
"loss": 0.0023,
|
| 444 |
+
"step": 1700
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 114.01,
|
| 448 |
+
"learning_rate": 3.171428571428571e-07,
|
| 449 |
+
"loss": 0.0023,
|
| 450 |
+
"step": 1725
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 116.0,
|
| 454 |
+
"learning_rate": 3.0693877551020407e-07,
|
| 455 |
+
"loss": 0.0023,
|
| 456 |
+
"step": 1750
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 118.0,
|
| 460 |
+
"learning_rate": 2.9673469387755103e-07,
|
| 461 |
+
"loss": 0.0022,
|
| 462 |
+
"step": 1775
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 119.01,
|
| 466 |
+
"learning_rate": 2.8653061224489793e-07,
|
| 467 |
+
"loss": 0.0022,
|
| 468 |
+
"step": 1800
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 121.0,
|
| 472 |
+
"learning_rate": 2.763265306122449e-07,
|
| 473 |
+
"loss": 0.0022,
|
| 474 |
+
"step": 1825
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 123.0,
|
| 478 |
+
"learning_rate": 2.6612244897959185e-07,
|
| 479 |
+
"loss": 0.0021,
|
| 480 |
+
"step": 1850
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 124.01,
|
| 484 |
+
"learning_rate": 2.559183673469388e-07,
|
| 485 |
+
"loss": 0.0021,
|
| 486 |
+
"step": 1875
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 126.0,
|
| 490 |
+
"learning_rate": 2.457142857142857e-07,
|
| 491 |
+
"loss": 0.0021,
|
| 492 |
+
"step": 1900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 128.0,
|
| 496 |
+
"learning_rate": 2.3551020408163264e-07,
|
| 497 |
+
"loss": 0.0021,
|
| 498 |
+
"step": 1925
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 129.01,
|
| 502 |
+
"learning_rate": 2.253061224489796e-07,
|
| 503 |
+
"loss": 0.002,
|
| 504 |
+
"step": 1950
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 131.0,
|
| 508 |
+
"learning_rate": 2.1510204081632652e-07,
|
| 509 |
+
"loss": 0.002,
|
| 510 |
+
"step": 1975
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 133.0,
|
| 514 |
+
"learning_rate": 2.0489795918367345e-07,
|
| 515 |
+
"loss": 0.002,
|
| 516 |
+
"step": 2000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 133.0,
|
| 520 |
+
"eval_loss": 0.7078843116760254,
|
| 521 |
+
"eval_runtime": 279.582,
|
| 522 |
+
"eval_samples_per_second": 4.743,
|
| 523 |
+
"eval_steps_per_second": 0.15,
|
| 524 |
+
"eval_wer": 26.478959138036185,
|
| 525 |
+
"step": 2000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 134.01,
|
| 529 |
+
"learning_rate": 1.946938775510204e-07,
|
| 530 |
+
"loss": 0.002,
|
| 531 |
+
"step": 2025
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 136.0,
|
| 535 |
+
"learning_rate": 1.8448979591836734e-07,
|
| 536 |
+
"loss": 0.002,
|
| 537 |
+
"step": 2050
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 138.0,
|
| 541 |
+
"learning_rate": 1.742857142857143e-07,
|
| 542 |
+
"loss": 0.0019,
|
| 543 |
+
"step": 2075
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 139.01,
|
| 547 |
+
"learning_rate": 1.6408163265306123e-07,
|
| 548 |
+
"loss": 0.0019,
|
| 549 |
+
"step": 2100
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 141.0,
|
| 553 |
+
"learning_rate": 1.5387755102040818e-07,
|
| 554 |
+
"loss": 0.0019,
|
| 555 |
+
"step": 2125
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 143.0,
|
| 559 |
+
"learning_rate": 1.4367346938775509e-07,
|
| 560 |
+
"loss": 0.0019,
|
| 561 |
+
"step": 2150
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 144.01,
|
| 565 |
+
"learning_rate": 1.3346938775510202e-07,
|
| 566 |
+
"loss": 0.0019,
|
| 567 |
+
"step": 2175
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 146.0,
|
| 571 |
+
"learning_rate": 1.2326530612244897e-07,
|
| 572 |
+
"loss": 0.0019,
|
| 573 |
+
"step": 2200
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 148.0,
|
| 577 |
+
"learning_rate": 1.130612244897959e-07,
|
| 578 |
+
"loss": 0.0019,
|
| 579 |
+
"step": 2225
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 149.01,
|
| 583 |
+
"learning_rate": 1.0285714285714285e-07,
|
| 584 |
+
"loss": 0.0018,
|
| 585 |
+
"step": 2250
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 151.0,
|
| 589 |
+
"learning_rate": 9.265306122448979e-08,
|
| 590 |
+
"loss": 0.0018,
|
| 591 |
+
"step": 2275
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 153.0,
|
| 595 |
+
"learning_rate": 8.244897959183673e-08,
|
| 596 |
+
"loss": 0.0018,
|
| 597 |
+
"step": 2300
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 154.01,
|
| 601 |
+
"learning_rate": 7.224489795918368e-08,
|
| 602 |
+
"loss": 0.0018,
|
| 603 |
+
"step": 2325
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 156.0,
|
| 607 |
+
"learning_rate": 6.20408163265306e-08,
|
| 608 |
+
"loss": 0.0018,
|
| 609 |
+
"step": 2350
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 158.0,
|
| 613 |
+
"learning_rate": 5.183673469387754e-08,
|
| 614 |
+
"loss": 0.0018,
|
| 615 |
+
"step": 2375
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 159.01,
|
| 619 |
+
"learning_rate": 4.1632653061224486e-08,
|
| 620 |
+
"loss": 0.0018,
|
| 621 |
+
"step": 2400
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 161.0,
|
| 625 |
+
"learning_rate": 3.142857142857143e-08,
|
| 626 |
+
"loss": 0.0018,
|
| 627 |
+
"step": 2425
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 163.0,
|
| 631 |
+
"learning_rate": 2.122448979591837e-08,
|
| 632 |
+
"loss": 0.0018,
|
| 633 |
+
"step": 2450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 164.01,
|
| 637 |
+
"learning_rate": 1.1020408163265306e-08,
|
| 638 |
+
"loss": 0.0018,
|
| 639 |
+
"step": 2475
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 166.0,
|
| 643 |
+
"learning_rate": 8.163265306122449e-10,
|
| 644 |
+
"loss": 0.0018,
|
| 645 |
+
"step": 2500
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 166.0,
|
| 649 |
+
"eval_loss": 0.7126978635787964,
|
| 650 |
+
"eval_runtime": 279.4326,
|
| 651 |
+
"eval_samples_per_second": 4.745,
|
| 652 |
+
"eval_steps_per_second": 0.15,
|
| 653 |
+
"eval_wer": 26.397641797113238,
|
| 654 |
+
"step": 2500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 166.0,
|
| 658 |
+
"step": 2500,
|
| 659 |
+
"total_flos": 4.579042263957504e+19,
|
| 660 |
+
"train_loss": 0.05099657071977854,
|
| 661 |
+
"train_runtime": 22043.5536,
|
| 662 |
+
"train_samples_per_second": 7.258,
|
| 663 |
+
"train_steps_per_second": 0.113
|
| 664 |
+
}
|
| 665 |
+
],
|
| 666 |
+
"logging_steps": 25,
|
| 667 |
+
"max_steps": 2500,
|
| 668 |
+
"num_input_tokens_seen": 0,
|
| 669 |
+
"num_train_epochs": 9223372036854775807,
|
| 670 |
+
"save_steps": 500,
|
| 671 |
+
"total_flos": 4.579042263957504e+19,
|
| 672 |
+
"train_batch_size": 32,
|
| 673 |
+
"trial_name": null,
|
| 674 |
+
"trial_params": null
|
| 675 |
+
}
|