End of training
Browse files- README.md +14 -2
- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +720 -0
README.md
CHANGED
|
@@ -3,11 +3,23 @@ license: apache-2.0
|
|
| 3 |
base_model: openai/whisper-base
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
metrics:
|
| 7 |
- wer
|
| 8 |
model-index:
|
| 9 |
- name: whisper-base-ckb
|
| 10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -15,7 +27,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# whisper-base-ckb
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.1420
|
| 21 |
- Wer: 0.2918
|
|
|
|
| 3 |
base_model: openai/whisper-base
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- razhan/common_voice_ckb_16
|
| 8 |
metrics:
|
| 9 |
- wer
|
| 10 |
model-index:
|
| 11 |
- name: whisper-base-ckb
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Automatic Speech Recognition
|
| 15 |
+
type: automatic-speech-recognition
|
| 16 |
+
dataset:
|
| 17 |
+
name: razhan/common_voice_ckb_16
|
| 18 |
+
type: razhan/common_voice_ckb_16
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Wer
|
| 21 |
+
type: wer
|
| 22 |
+
value: 0.2917510463075469
|
| 23 |
---
|
| 24 |
|
| 25 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 27 |
|
| 28 |
# whisper-base-ckb
|
| 29 |
|
| 30 |
+
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the razhan/common_voice_ckb_16 dataset.
|
| 31 |
It achieves the following results on the evaluation set:
|
| 32 |
- Loss: 0.1420
|
| 33 |
- Wer: 0.2918
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.87,
|
| 3 |
+
"eval_loss": 0.1419677734375,
|
| 4 |
+
"eval_runtime": 117.7598,
|
| 5 |
+
"eval_samples": 4940,
|
| 6 |
+
"eval_samples_per_second": 41.95,
|
| 7 |
+
"eval_steps_per_second": 0.059,
|
| 8 |
+
"eval_wer": 0.2917510463075469,
|
| 9 |
+
"train_loss": 0.25973586964607237,
|
| 10 |
+
"train_runtime": 25779.7961,
|
| 11 |
+
"train_samples": 105929,
|
| 12 |
+
"train_samples_per_second": 44.686,
|
| 13 |
+
"train_steps_per_second": 0.039
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.87,
|
| 3 |
+
"eval_loss": 0.1419677734375,
|
| 4 |
+
"eval_runtime": 117.7598,
|
| 5 |
+
"eval_samples": 4940,
|
| 6 |
+
"eval_samples_per_second": 41.95,
|
| 7 |
+
"eval_steps_per_second": 0.059,
|
| 8 |
+
"eval_wer": 0.2917510463075469
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.87,
|
| 3 |
+
"train_loss": 0.25973586964607237,
|
| 4 |
+
"train_runtime": 25779.7961,
|
| 5 |
+
"train_samples": 105929,
|
| 6 |
+
"train_samples_per_second": 44.686,
|
| 7 |
+
"train_steps_per_second": 0.039
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,720 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.2917510463075469,
|
| 3 |
+
"best_model_checkpoint": "./whisper-base-ckb/checkpoint-1000",
|
| 4 |
+
"epoch": 10.869565217391305,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11,
|
| 13 |
+
"learning_rate": 2.0735116692035353e-06,
|
| 14 |
+
"loss": 3.7642,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.22,
|
| 19 |
+
"learning_rate": 4.841064020920768e-06,
|
| 20 |
+
"loss": 2.8375,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.33,
|
| 25 |
+
"learning_rate": 5.9179056274086315e-06,
|
| 26 |
+
"loss": 1.8942,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.43,
|
| 31 |
+
"learning_rate": 6.599279204266558e-06,
|
| 32 |
+
"loss": 1.2854,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.54,
|
| 37 |
+
"learning_rate": 7.098857723096461e-06,
|
| 38 |
+
"loss": 0.9078,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.65,
|
| 43 |
+
"learning_rate": 7.493495838168052e-06,
|
| 44 |
+
"loss": 0.6876,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.76,
|
| 49 |
+
"learning_rate": 7.819717921297098e-06,
|
| 50 |
+
"loss": 0.5329,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.87,
|
| 55 |
+
"learning_rate": 8.097777358972581e-06,
|
| 56 |
+
"loss": 0.4405,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.98,
|
| 61 |
+
"learning_rate": 8.340082901971508e-06,
|
| 62 |
+
"loss": 0.3864,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.09,
|
| 67 |
+
"learning_rate": 8.55479047323724e-06,
|
| 68 |
+
"loss": 0.3434,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.09,
|
| 73 |
+
"eval_loss": 0.384033203125,
|
| 74 |
+
"eval_runtime": 114.2378,
|
| 75 |
+
"eval_samples_per_second": 43.243,
|
| 76 |
+
"eval_steps_per_second": 0.061,
|
| 77 |
+
"eval_wer": 0.6053732955312542,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.2,
|
| 82 |
+
"learning_rate": 8.747548830298285e-06,
|
| 83 |
+
"loss": 0.3216,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.3,
|
| 88 |
+
"learning_rate": 8.92243233400783e-06,
|
| 89 |
+
"loss": 0.3038,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.41,
|
| 94 |
+
"learning_rate": 9.082476609942404e-06,
|
| 95 |
+
"loss": 0.2841,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.52,
|
| 100 |
+
"learning_rate": 9.230004150181585e-06,
|
| 101 |
+
"loss": 0.2701,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.63,
|
| 106 |
+
"learning_rate": 9.36683155598379e-06,
|
| 107 |
+
"loss": 0.255,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.74,
|
| 112 |
+
"learning_rate": 9.494406569816089e-06,
|
| 113 |
+
"loss": 0.2549,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.85,
|
| 118 |
+
"learning_rate": 9.613901637842896e-06,
|
| 119 |
+
"loss": 0.2433,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.96,
|
| 124 |
+
"learning_rate": 9.726279567611587e-06,
|
| 125 |
+
"loss": 0.2373,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 2.07,
|
| 130 |
+
"learning_rate": 9.832340708264348e-06,
|
| 131 |
+
"loss": 0.2232,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.17,
|
| 136 |
+
"learning_rate": 9.93275756211191e-06,
|
| 137 |
+
"loss": 0.2089,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.17,
|
| 142 |
+
"eval_loss": 0.265380859375,
|
| 143 |
+
"eval_runtime": 112.176,
|
| 144 |
+
"eval_samples_per_second": 44.038,
|
| 145 |
+
"eval_steps_per_second": 0.062,
|
| 146 |
+
"eval_wer": 0.4739773187525314,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.28,
|
| 151 |
+
"learning_rate": 9.975000000000002e-06,
|
| 152 |
+
"loss": 0.2122,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.39,
|
| 157 |
+
"learning_rate": 9.85e-06,
|
| 158 |
+
"loss": 0.2123,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.5,
|
| 163 |
+
"learning_rate": 9.725000000000001e-06,
|
| 164 |
+
"loss": 0.2044,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.61,
|
| 169 |
+
"learning_rate": 9.600000000000001e-06,
|
| 170 |
+
"loss": 0.1994,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.72,
|
| 175 |
+
"learning_rate": 9.475000000000002e-06,
|
| 176 |
+
"loss": 0.1911,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.83,
|
| 181 |
+
"learning_rate": 9.350000000000002e-06,
|
| 182 |
+
"loss": 0.1927,
|
| 183 |
+
"step": 260
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.93,
|
| 187 |
+
"learning_rate": 9.225e-06,
|
| 188 |
+
"loss": 0.1913,
|
| 189 |
+
"step": 270
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 3.04,
|
| 193 |
+
"learning_rate": 9.100000000000001e-06,
|
| 194 |
+
"loss": 0.1812,
|
| 195 |
+
"step": 280
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.15,
|
| 199 |
+
"learning_rate": 8.975e-06,
|
| 200 |
+
"loss": 0.1715,
|
| 201 |
+
"step": 290
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.26,
|
| 205 |
+
"learning_rate": 8.85e-06,
|
| 206 |
+
"loss": 0.167,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.26,
|
| 211 |
+
"eval_loss": 0.224609375,
|
| 212 |
+
"eval_runtime": 112.6918,
|
| 213 |
+
"eval_samples_per_second": 43.836,
|
| 214 |
+
"eval_steps_per_second": 0.062,
|
| 215 |
+
"eval_wer": 0.41899554475496154,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.37,
|
| 220 |
+
"learning_rate": 8.725000000000002e-06,
|
| 221 |
+
"loss": 0.1683,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.48,
|
| 226 |
+
"learning_rate": 8.6e-06,
|
| 227 |
+
"loss": 0.1728,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.59,
|
| 232 |
+
"learning_rate": 8.475000000000001e-06,
|
| 233 |
+
"loss": 0.1692,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 3.7,
|
| 238 |
+
"learning_rate": 8.35e-06,
|
| 239 |
+
"loss": 0.1654,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.8,
|
| 244 |
+
"learning_rate": 8.225e-06,
|
| 245 |
+
"loss": 0.1622,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.91,
|
| 250 |
+
"learning_rate": 8.1e-06,
|
| 251 |
+
"loss": 0.1597,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 4.02,
|
| 256 |
+
"learning_rate": 7.975e-06,
|
| 257 |
+
"loss": 0.1626,
|
| 258 |
+
"step": 370
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 4.13,
|
| 262 |
+
"learning_rate": 7.850000000000001e-06,
|
| 263 |
+
"loss": 0.151,
|
| 264 |
+
"step": 380
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 4.24,
|
| 268 |
+
"learning_rate": 7.725e-06,
|
| 269 |
+
"loss": 0.1481,
|
| 270 |
+
"step": 390
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 4.35,
|
| 274 |
+
"learning_rate": 7.600000000000001e-06,
|
| 275 |
+
"loss": 0.1452,
|
| 276 |
+
"step": 400
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 4.35,
|
| 280 |
+
"eval_loss": 0.1964111328125,
|
| 281 |
+
"eval_runtime": 112.4796,
|
| 282 |
+
"eval_samples_per_second": 43.919,
|
| 283 |
+
"eval_steps_per_second": 0.062,
|
| 284 |
+
"eval_wer": 0.38034966923180774,
|
| 285 |
+
"step": 400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.46,
|
| 289 |
+
"learning_rate": 7.475000000000001e-06,
|
| 290 |
+
"loss": 0.1437,
|
| 291 |
+
"step": 410
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 4.57,
|
| 295 |
+
"learning_rate": 7.350000000000001e-06,
|
| 296 |
+
"loss": 0.1477,
|
| 297 |
+
"step": 420
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.67,
|
| 301 |
+
"learning_rate": 7.225000000000001e-06,
|
| 302 |
+
"loss": 0.146,
|
| 303 |
+
"step": 430
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 4.78,
|
| 307 |
+
"learning_rate": 7.100000000000001e-06,
|
| 308 |
+
"loss": 0.1432,
|
| 309 |
+
"step": 440
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 4.89,
|
| 313 |
+
"learning_rate": 6.975000000000001e-06,
|
| 314 |
+
"loss": 0.1383,
|
| 315 |
+
"step": 450
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 5.0,
|
| 319 |
+
"learning_rate": 6.850000000000001e-06,
|
| 320 |
+
"loss": 0.1389,
|
| 321 |
+
"step": 460
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 5.11,
|
| 325 |
+
"learning_rate": 6.725000000000001e-06,
|
| 326 |
+
"loss": 0.1416,
|
| 327 |
+
"step": 470
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 5.22,
|
| 331 |
+
"learning_rate": 6.600000000000001e-06,
|
| 332 |
+
"loss": 0.1356,
|
| 333 |
+
"step": 480
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 5.33,
|
| 337 |
+
"learning_rate": 6.475e-06,
|
| 338 |
+
"loss": 0.1302,
|
| 339 |
+
"step": 490
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 5.43,
|
| 343 |
+
"learning_rate": 6.35e-06,
|
| 344 |
+
"loss": 0.1287,
|
| 345 |
+
"step": 500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 5.43,
|
| 349 |
+
"eval_loss": 0.1788330078125,
|
| 350 |
+
"eval_runtime": 133.6018,
|
| 351 |
+
"eval_samples_per_second": 36.976,
|
| 352 |
+
"eval_steps_per_second": 0.052,
|
| 353 |
+
"eval_wer": 0.35419198055893075,
|
| 354 |
+
"step": 500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 5.54,
|
| 358 |
+
"learning_rate": 6.225000000000001e-06,
|
| 359 |
+
"loss": 0.1217,
|
| 360 |
+
"step": 510
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 5.65,
|
| 364 |
+
"learning_rate": 6.1e-06,
|
| 365 |
+
"loss": 0.1332,
|
| 366 |
+
"step": 520
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 5.76,
|
| 370 |
+
"learning_rate": 5.975e-06,
|
| 371 |
+
"loss": 0.1307,
|
| 372 |
+
"step": 530
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 5.87,
|
| 376 |
+
"learning_rate": 5.85e-06,
|
| 377 |
+
"loss": 0.1281,
|
| 378 |
+
"step": 540
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 5.98,
|
| 382 |
+
"learning_rate": 5.725e-06,
|
| 383 |
+
"loss": 0.1257,
|
| 384 |
+
"step": 550
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 6.09,
|
| 388 |
+
"learning_rate": 5.600000000000001e-06,
|
| 389 |
+
"loss": 0.1176,
|
| 390 |
+
"step": 560
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 6.2,
|
| 394 |
+
"learning_rate": 5.475e-06,
|
| 395 |
+
"loss": 0.1249,
|
| 396 |
+
"step": 570
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 6.3,
|
| 400 |
+
"learning_rate": 5.3500000000000004e-06,
|
| 401 |
+
"loss": 0.1214,
|
| 402 |
+
"step": 580
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 6.41,
|
| 406 |
+
"learning_rate": 5.225e-06,
|
| 407 |
+
"loss": 0.1166,
|
| 408 |
+
"step": 590
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 6.52,
|
| 412 |
+
"learning_rate": 5.1e-06,
|
| 413 |
+
"loss": 0.1163,
|
| 414 |
+
"step": 600
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 6.52,
|
| 418 |
+
"eval_loss": 0.1650390625,
|
| 419 |
+
"eval_runtime": 120.7458,
|
| 420 |
+
"eval_samples_per_second": 40.912,
|
| 421 |
+
"eval_steps_per_second": 0.058,
|
| 422 |
+
"eval_wer": 0.33255704063723507,
|
| 423 |
+
"step": 600
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 6.63,
|
| 427 |
+
"learning_rate": 4.975000000000001e-06,
|
| 428 |
+
"loss": 0.1139,
|
| 429 |
+
"step": 610
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 6.74,
|
| 433 |
+
"learning_rate": 4.85e-06,
|
| 434 |
+
"loss": 0.1216,
|
| 435 |
+
"step": 620
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 6.85,
|
| 439 |
+
"learning_rate": 4.7250000000000005e-06,
|
| 440 |
+
"loss": 0.1189,
|
| 441 |
+
"step": 630
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.96,
|
| 445 |
+
"learning_rate": 4.600000000000001e-06,
|
| 446 |
+
"loss": 0.1167,
|
| 447 |
+
"step": 640
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 7.07,
|
| 451 |
+
"learning_rate": 4.475e-06,
|
| 452 |
+
"loss": 0.1132,
|
| 453 |
+
"step": 650
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 7.17,
|
| 457 |
+
"learning_rate": 4.350000000000001e-06,
|
| 458 |
+
"loss": 0.1046,
|
| 459 |
+
"step": 660
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 7.28,
|
| 463 |
+
"learning_rate": 4.225e-06,
|
| 464 |
+
"loss": 0.111,
|
| 465 |
+
"step": 670
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 7.39,
|
| 469 |
+
"learning_rate": 4.1e-06,
|
| 470 |
+
"loss": 0.115,
|
| 471 |
+
"step": 680
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 7.5,
|
| 475 |
+
"learning_rate": 3.975000000000001e-06,
|
| 476 |
+
"loss": 0.1097,
|
| 477 |
+
"step": 690
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 7.61,
|
| 481 |
+
"learning_rate": 3.85e-06,
|
| 482 |
+
"loss": 0.1068,
|
| 483 |
+
"step": 700
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 7.61,
|
| 487 |
+
"eval_loss": 0.156005859375,
|
| 488 |
+
"eval_runtime": 128.5945,
|
| 489 |
+
"eval_samples_per_second": 38.415,
|
| 490 |
+
"eval_steps_per_second": 0.054,
|
| 491 |
+
"eval_wer": 0.31554610503577696,
|
| 492 |
+
"step": 700
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.72,
|
| 496 |
+
"learning_rate": 3.7250000000000003e-06,
|
| 497 |
+
"loss": 0.1061,
|
| 498 |
+
"step": 710
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 7.83,
|
| 502 |
+
"learning_rate": 3.6000000000000003e-06,
|
| 503 |
+
"loss": 0.1112,
|
| 504 |
+
"step": 720
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 7.93,
|
| 508 |
+
"learning_rate": 3.475e-06,
|
| 509 |
+
"loss": 0.1128,
|
| 510 |
+
"step": 730
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 8.04,
|
| 514 |
+
"learning_rate": 3.3500000000000005e-06,
|
| 515 |
+
"loss": 0.107,
|
| 516 |
+
"step": 740
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 8.15,
|
| 520 |
+
"learning_rate": 3.2250000000000005e-06,
|
| 521 |
+
"loss": 0.1022,
|
| 522 |
+
"step": 750
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 8.26,
|
| 526 |
+
"learning_rate": 3.1000000000000004e-06,
|
| 527 |
+
"loss": 0.0977,
|
| 528 |
+
"step": 760
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 8.37,
|
| 532 |
+
"learning_rate": 2.9750000000000003e-06,
|
| 533 |
+
"loss": 0.1016,
|
| 534 |
+
"step": 770
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 8.48,
|
| 538 |
+
"learning_rate": 2.85e-06,
|
| 539 |
+
"loss": 0.1089,
|
| 540 |
+
"step": 780
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 8.59,
|
| 544 |
+
"learning_rate": 2.7250000000000006e-06,
|
| 545 |
+
"loss": 0.105,
|
| 546 |
+
"step": 790
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 8.7,
|
| 550 |
+
"learning_rate": 2.6e-06,
|
| 551 |
+
"loss": 0.1015,
|
| 552 |
+
"step": 800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 8.7,
|
| 556 |
+
"eval_loss": 0.14892578125,
|
| 557 |
+
"eval_runtime": 117.3243,
|
| 558 |
+
"eval_samples_per_second": 42.106,
|
| 559 |
+
"eval_steps_per_second": 0.06,
|
| 560 |
+
"eval_wer": 0.30592682597542864,
|
| 561 |
+
"step": 800
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 8.8,
|
| 565 |
+
"learning_rate": 2.475e-06,
|
| 566 |
+
"loss": 0.0977,
|
| 567 |
+
"step": 810
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 8.91,
|
| 571 |
+
"learning_rate": 2.35e-06,
|
| 572 |
+
"loss": 0.1023,
|
| 573 |
+
"step": 820
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 9.02,
|
| 577 |
+
"learning_rate": 2.2250000000000003e-06,
|
| 578 |
+
"loss": 0.1064,
|
| 579 |
+
"step": 830
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 9.13,
|
| 583 |
+
"learning_rate": 2.1000000000000002e-06,
|
| 584 |
+
"loss": 0.0964,
|
| 585 |
+
"step": 840
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 9.24,
|
| 589 |
+
"learning_rate": 1.975e-06,
|
| 590 |
+
"loss": 0.0973,
|
| 591 |
+
"step": 850
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 9.35,
|
| 595 |
+
"learning_rate": 1.85e-06,
|
| 596 |
+
"loss": 0.0946,
|
| 597 |
+
"step": 860
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 9.46,
|
| 601 |
+
"learning_rate": 1.725e-06,
|
| 602 |
+
"loss": 0.0965,
|
| 603 |
+
"step": 870
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 9.57,
|
| 607 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 608 |
+
"loss": 0.1048,
|
| 609 |
+
"step": 880
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 9.67,
|
| 613 |
+
"learning_rate": 1.475e-06,
|
| 614 |
+
"loss": 0.0982,
|
| 615 |
+
"step": 890
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 9.78,
|
| 619 |
+
"learning_rate": 1.3500000000000002e-06,
|
| 620 |
+
"loss": 0.0968,
|
| 621 |
+
"step": 900
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 9.78,
|
| 625 |
+
"eval_loss": 0.14404296875,
|
| 626 |
+
"eval_runtime": 113.4503,
|
| 627 |
+
"eval_samples_per_second": 43.543,
|
| 628 |
+
"eval_steps_per_second": 0.062,
|
| 629 |
+
"eval_wer": 0.2953962467935736,
|
| 630 |
+
"step": 900
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 9.89,
|
| 634 |
+
"learning_rate": 1.2250000000000001e-06,
|
| 635 |
+
"loss": 0.0951,
|
| 636 |
+
"step": 910
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 10.0,
|
| 640 |
+
"learning_rate": 1.1e-06,
|
| 641 |
+
"loss": 0.0937,
|
| 642 |
+
"step": 920
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 10.11,
|
| 646 |
+
"learning_rate": 9.750000000000002e-07,
|
| 647 |
+
"loss": 0.1012,
|
| 648 |
+
"step": 930
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 10.22,
|
| 652 |
+
"learning_rate": 8.500000000000001e-07,
|
| 653 |
+
"loss": 0.0963,
|
| 654 |
+
"step": 940
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 10.33,
|
| 658 |
+
"learning_rate": 7.25e-07,
|
| 659 |
+
"loss": 0.0941,
|
| 660 |
+
"step": 950
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 10.43,
|
| 664 |
+
"learning_rate": 6.000000000000001e-07,
|
| 665 |
+
"loss": 0.0922,
|
| 666 |
+
"step": 960
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 10.54,
|
| 670 |
+
"learning_rate": 4.7500000000000006e-07,
|
| 671 |
+
"loss": 0.0897,
|
| 672 |
+
"step": 970
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 10.65,
|
| 676 |
+
"learning_rate": 3.5000000000000004e-07,
|
| 677 |
+
"loss": 0.1039,
|
| 678 |
+
"step": 980
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 10.76,
|
| 682 |
+
"learning_rate": 2.2500000000000002e-07,
|
| 683 |
+
"loss": 0.0978,
|
| 684 |
+
"step": 990
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 10.87,
|
| 688 |
+
"learning_rate": 1.0000000000000001e-07,
|
| 689 |
+
"loss": 0.0939,
|
| 690 |
+
"step": 1000
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 10.87,
|
| 694 |
+
"eval_loss": 0.1419677734375,
|
| 695 |
+
"eval_runtime": 116.6648,
|
| 696 |
+
"eval_samples_per_second": 42.344,
|
| 697 |
+
"eval_steps_per_second": 0.06,
|
| 698 |
+
"eval_wer": 0.2917510463075469,
|
| 699 |
+
"step": 1000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 10.87,
|
| 703 |
+
"step": 1000,
|
| 704 |
+
"total_flos": 7.47187145428435e+19,
|
| 705 |
+
"train_loss": 0.25973586964607237,
|
| 706 |
+
"train_runtime": 25779.7961,
|
| 707 |
+
"train_samples_per_second": 44.686,
|
| 708 |
+
"train_steps_per_second": 0.039
|
| 709 |
+
}
|
| 710 |
+
],
|
| 711 |
+
"logging_steps": 10,
|
| 712 |
+
"max_steps": 1000,
|
| 713 |
+
"num_input_tokens_seen": 0,
|
| 714 |
+
"num_train_epochs": 11,
|
| 715 |
+
"save_steps": 100,
|
| 716 |
+
"total_flos": 7.47187145428435e+19,
|
| 717 |
+
"train_batch_size": 192,
|
| 718 |
+
"trial_name": null,
|
| 719 |
+
"trial_params": null
|
| 720 |
+
}
|