End of training
Browse files- README.md +9 -5
- all_results.json +15 -0
- eval_results.json +10 -0
- train_results.json +8 -0
- trainer_state.json +509 -0
README.md
CHANGED
|
@@ -1,6 +1,10 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
tags:
|
|
|
|
|
|
|
| 4 |
- generated_from_trainer
|
| 5 |
datasets:
|
| 6 |
- common_voice_13_0
|
|
@@ -13,15 +17,15 @@ model-index:
|
|
| 13 |
name: Automatic Speech Recognition
|
| 14 |
type: automatic-speech-recognition
|
| 15 |
dataset:
|
| 16 |
-
name:
|
| 17 |
type: common_voice_13_0
|
| 18 |
config: eo
|
| 19 |
split: validation
|
| 20 |
-
args: eo
|
| 21 |
metrics:
|
| 22 |
- name: Wer
|
| 23 |
type: wer
|
| 24 |
-
value: 0.
|
| 25 |
---
|
| 26 |
|
| 27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -29,11 +33,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 29 |
|
| 30 |
# wav2vec2-common_voice_13_0-eo-10_1
|
| 31 |
|
| 32 |
-
This model is a fine-tuned version of [xekri/wav2vec2-common_voice_13_0-eo-10](https://huggingface.co/xekri/wav2vec2-common_voice_13_0-eo-10) on the
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
- Loss: 0.0391
|
| 35 |
- Cer: 0.0098
|
| 36 |
-
- Wer: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 1 |
---
|
| 2 |
+
language:
|
| 3 |
+
- eo
|
| 4 |
license: apache-2.0
|
| 5 |
tags:
|
| 6 |
+
- automatic-speech-recognition
|
| 7 |
+
- mozilla-foundation/common_voice_13_0
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
- common_voice_13_0
|
|
|
|
| 17 |
name: Automatic Speech Recognition
|
| 18 |
type: automatic-speech-recognition
|
| 19 |
dataset:
|
| 20 |
+
name: MOZILLA-FOUNDATION/COMMON_VOICE_13_0 - EO
|
| 21 |
type: common_voice_13_0
|
| 22 |
config: eo
|
| 23 |
split: validation
|
| 24 |
+
args: 'Config: eo, Training split: train, Eval split: validation'
|
| 25 |
metrics:
|
| 26 |
- name: Wer
|
| 27 |
type: wer
|
| 28 |
+
value: 0.05342994850125446
|
| 29 |
---
|
| 30 |
|
| 31 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
# wav2vec2-common_voice_13_0-eo-10_1
|
| 35 |
|
| 36 |
+
This model is a fine-tuned version of [xekri/wav2vec2-common_voice_13_0-eo-10](https://huggingface.co/xekri/wav2vec2-common_voice_13_0-eo-10) on the MOZILLA-FOUNDATION/COMMON_VOICE_13_0 - EO dataset.
|
| 37 |
It achieves the following results on the evaluation set:
|
| 38 |
- Loss: 0.0391
|
| 39 |
- Cer: 0.0098
|
| 40 |
+
- Wer: 0.0534
|
| 41 |
|
| 42 |
## Model description
|
| 43 |
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_cer": 0.009762087662537918,
|
| 4 |
+
"eval_loss": 0.03910906985402107,
|
| 5 |
+
"eval_runtime": 1348.6492,
|
| 6 |
+
"eval_samples": 14892,
|
| 7 |
+
"eval_samples_per_second": 11.042,
|
| 8 |
+
"eval_steps_per_second": 1.381,
|
| 9 |
+
"eval_wer": 0.05342994850125446,
|
| 10 |
+
"train_loss": 0.08836701542702612,
|
| 11 |
+
"train_runtime": 131691.1828,
|
| 12 |
+
"train_samples": 143849,
|
| 13 |
+
"train_samples_per_second": 5.462,
|
| 14 |
+
"train_steps_per_second": 0.171
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_cer": 0.009762087662537918,
|
| 4 |
+
"eval_loss": 0.03910906985402107,
|
| 5 |
+
"eval_runtime": 1348.6492,
|
| 6 |
+
"eval_samples": 14892,
|
| 7 |
+
"eval_samples_per_second": 11.042,
|
| 8 |
+
"eval_steps_per_second": 1.381,
|
| 9 |
+
"eval_wer": 0.05342994850125446
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.08836701542702612,
|
| 4 |
+
"train_runtime": 131691.1828,
|
| 5 |
+
"train_samples": 143849,
|
| 6 |
+
"train_samples_per_second": 5.462,
|
| 7 |
+
"train_steps_per_second": 0.171
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,509 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.999443888332777,
|
| 5 |
+
"global_step": 22475,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.11,
|
| 12 |
+
"learning_rate": 3e-05,
|
| 13 |
+
"loss": 0.1124,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.22,
|
| 18 |
+
"learning_rate": 2.9317406143344712e-05,
|
| 19 |
+
"loss": 0.1142,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.22,
|
| 24 |
+
"eval_cer": 0.012641370841580549,
|
| 25 |
+
"eval_loss": 0.04826715216040611,
|
| 26 |
+
"eval_runtime": 1346.7575,
|
| 27 |
+
"eval_samples_per_second": 11.058,
|
| 28 |
+
"eval_steps_per_second": 1.383,
|
| 29 |
+
"eval_wer": 0.07066222104846163,
|
| 30 |
+
"step": 1000
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.33,
|
| 34 |
+
"learning_rate": 2.863481228668942e-05,
|
| 35 |
+
"loss": 0.113,
|
| 36 |
+
"step": 1500
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.44,
|
| 40 |
+
"learning_rate": 2.7953583617747444e-05,
|
| 41 |
+
"loss": 0.1049,
|
| 42 |
+
"step": 2000
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.44,
|
| 46 |
+
"eval_cer": 0.01225587771877786,
|
| 47 |
+
"eval_loss": 0.04738128185272217,
|
| 48 |
+
"eval_runtime": 1341.1997,
|
| 49 |
+
"eval_samples_per_second": 11.103,
|
| 50 |
+
"eval_steps_per_second": 1.388,
|
| 51 |
+
"eval_wer": 0.06750957348474845,
|
| 52 |
+
"step": 2000
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.56,
|
| 56 |
+
"learning_rate": 2.727098976109215e-05,
|
| 57 |
+
"loss": 0.0991,
|
| 58 |
+
"step": 2500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.67,
|
| 62 |
+
"learning_rate": 2.6588395904436862e-05,
|
| 63 |
+
"loss": 0.0982,
|
| 64 |
+
"step": 3000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.67,
|
| 68 |
+
"eval_cer": 0.011982528049881409,
|
| 69 |
+
"eval_loss": 0.04708929359912872,
|
| 70 |
+
"eval_runtime": 1340.0171,
|
| 71 |
+
"eval_samples_per_second": 11.113,
|
| 72 |
+
"eval_steps_per_second": 1.39,
|
| 73 |
+
"eval_wer": 0.0663871649280338,
|
| 74 |
+
"step": 3000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.78,
|
| 78 |
+
"learning_rate": 2.5905802047781573e-05,
|
| 79 |
+
"loss": 0.0996,
|
| 80 |
+
"step": 3500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.89,
|
| 84 |
+
"learning_rate": 2.522457337883959e-05,
|
| 85 |
+
"loss": 0.092,
|
| 86 |
+
"step": 4000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.89,
|
| 90 |
+
"eval_cer": 0.01171338376050644,
|
| 91 |
+
"eval_loss": 0.04587830230593681,
|
| 92 |
+
"eval_runtime": 1347.3232,
|
| 93 |
+
"eval_samples_per_second": 11.053,
|
| 94 |
+
"eval_steps_per_second": 1.382,
|
| 95 |
+
"eval_wer": 0.0639855407368282,
|
| 96 |
+
"step": 4000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.0,
|
| 100 |
+
"learning_rate": 2.45419795221843e-05,
|
| 101 |
+
"loss": 0.0896,
|
| 102 |
+
"step": 4500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.11,
|
| 106 |
+
"learning_rate": 2.385938566552901e-05,
|
| 107 |
+
"loss": 0.0847,
|
| 108 |
+
"step": 5000
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.11,
|
| 112 |
+
"eval_cer": 0.011517132716170525,
|
| 113 |
+
"eval_loss": 0.04586043953895569,
|
| 114 |
+
"eval_runtime": 1342.7891,
|
| 115 |
+
"eval_samples_per_second": 11.09,
|
| 116 |
+
"eval_steps_per_second": 1.387,
|
| 117 |
+
"eval_wer": 0.06308596329063779,
|
| 118 |
+
"step": 5000
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 1.22,
|
| 122 |
+
"learning_rate": 2.317679180887372e-05,
|
| 123 |
+
"loss": 0.0838,
|
| 124 |
+
"step": 5500
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.33,
|
| 128 |
+
"learning_rate": 2.249419795221843e-05,
|
| 129 |
+
"loss": 0.0837,
|
| 130 |
+
"step": 6000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.33,
|
| 134 |
+
"eval_cer": 0.011336301396746719,
|
| 135 |
+
"eval_loss": 0.04530934989452362,
|
| 136 |
+
"eval_runtime": 1343.699,
|
| 137 |
+
"eval_samples_per_second": 11.083,
|
| 138 |
+
"eval_steps_per_second": 1.386,
|
| 139 |
+
"eval_wer": 0.06242572296315859,
|
| 140 |
+
"step": 6000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.45,
|
| 144 |
+
"learning_rate": 2.181160409556314e-05,
|
| 145 |
+
"loss": 0.0822,
|
| 146 |
+
"step": 6500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.56,
|
| 150 |
+
"learning_rate": 2.1130375426621163e-05,
|
| 151 |
+
"loss": 0.0803,
|
| 152 |
+
"step": 7000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.56,
|
| 156 |
+
"eval_cer": 0.010879316822078803,
|
| 157 |
+
"eval_loss": 0.04430945962667465,
|
| 158 |
+
"eval_runtime": 1343.5225,
|
| 159 |
+
"eval_samples_per_second": 11.084,
|
| 160 |
+
"eval_steps_per_second": 1.386,
|
| 161 |
+
"eval_wer": 0.0597682556450548,
|
| 162 |
+
"step": 7000
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.67,
|
| 166 |
+
"learning_rate": 2.044778156996587e-05,
|
| 167 |
+
"loss": 0.0811,
|
| 168 |
+
"step": 7500
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.78,
|
| 172 |
+
"learning_rate": 1.976518771331058e-05,
|
| 173 |
+
"loss": 0.0826,
|
| 174 |
+
"step": 8000
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.78,
|
| 178 |
+
"eval_cer": 0.011001272828201836,
|
| 179 |
+
"eval_loss": 0.044084154069423676,
|
| 180 |
+
"eval_runtime": 1339.8673,
|
| 181 |
+
"eval_samples_per_second": 11.115,
|
| 182 |
+
"eval_steps_per_second": 1.39,
|
| 183 |
+
"eval_wer": 0.06039548395616004,
|
| 184 |
+
"step": 8000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.89,
|
| 188 |
+
"learning_rate": 1.9082593856655292e-05,
|
| 189 |
+
"loss": 0.0795,
|
| 190 |
+
"step": 8500
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 2.0,
|
| 194 |
+
"learning_rate": 1.84e-05,
|
| 195 |
+
"loss": 0.0809,
|
| 196 |
+
"step": 9000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 2.0,
|
| 200 |
+
"eval_cer": 0.01099146027598504,
|
| 201 |
+
"eval_loss": 0.04365098848938942,
|
| 202 |
+
"eval_runtime": 1342.6486,
|
| 203 |
+
"eval_samples_per_second": 11.092,
|
| 204 |
+
"eval_steps_per_second": 1.387,
|
| 205 |
+
"eval_wer": 0.06050277300937541,
|
| 206 |
+
"step": 9000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 2.11,
|
| 210 |
+
"learning_rate": 1.771740614334471e-05,
|
| 211 |
+
"loss": 0.0726,
|
| 212 |
+
"step": 9500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.22,
|
| 216 |
+
"learning_rate": 1.7034812286689422e-05,
|
| 217 |
+
"loss": 0.0728,
|
| 218 |
+
"step": 10000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.22,
|
| 222 |
+
"eval_cer": 0.010905950892381534,
|
| 223 |
+
"eval_loss": 0.045100387185811996,
|
| 224 |
+
"eval_runtime": 1342.5545,
|
| 225 |
+
"eval_samples_per_second": 11.092,
|
| 226 |
+
"eval_steps_per_second": 1.387,
|
| 227 |
+
"eval_wer": 0.05974349663277433,
|
| 228 |
+
"step": 10000
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.34,
|
| 232 |
+
"learning_rate": 1.6352218430034133e-05,
|
| 233 |
+
"loss": 0.072,
|
| 234 |
+
"step": 10500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.45,
|
| 238 |
+
"learning_rate": 1.567098976109215e-05,
|
| 239 |
+
"loss": 0.0707,
|
| 240 |
+
"step": 11000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.45,
|
| 244 |
+
"eval_cer": 0.010758762609129599,
|
| 245 |
+
"eval_loss": 0.04439757764339447,
|
| 246 |
+
"eval_runtime": 1341.1302,
|
| 247 |
+
"eval_samples_per_second": 11.104,
|
| 248 |
+
"eval_steps_per_second": 1.388,
|
| 249 |
+
"eval_wer": 0.05909976231348211,
|
| 250 |
+
"step": 11000
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.56,
|
| 254 |
+
"learning_rate": 1.4988395904436861e-05,
|
| 255 |
+
"loss": 0.072,
|
| 256 |
+
"step": 11500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.67,
|
| 260 |
+
"learning_rate": 1.430580204778157e-05,
|
| 261 |
+
"loss": 0.0698,
|
| 262 |
+
"step": 12000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 2.67,
|
| 266 |
+
"eval_cer": 0.010531672114969469,
|
| 267 |
+
"eval_loss": 0.04415421932935715,
|
| 268 |
+
"eval_runtime": 1339.1928,
|
| 269 |
+
"eval_samples_per_second": 11.12,
|
| 270 |
+
"eval_steps_per_second": 1.39,
|
| 271 |
+
"eval_wer": 0.05758120956027994,
|
| 272 |
+
"step": 12000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.78,
|
| 276 |
+
"learning_rate": 1.3624573378839592e-05,
|
| 277 |
+
"loss": 0.0958,
|
| 278 |
+
"step": 12500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.89,
|
| 282 |
+
"learning_rate": 1.29419795221843e-05,
|
| 283 |
+
"loss": 0.0981,
|
| 284 |
+
"step": 13000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.89,
|
| 288 |
+
"eval_cer": 0.010388689211239017,
|
| 289 |
+
"eval_loss": 0.04108743742108345,
|
| 290 |
+
"eval_runtime": 1339.0236,
|
| 291 |
+
"eval_samples_per_second": 11.122,
|
| 292 |
+
"eval_steps_per_second": 1.391,
|
| 293 |
+
"eval_wer": 0.05717681235969893,
|
| 294 |
+
"step": 13000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 3.0,
|
| 298 |
+
"learning_rate": 1.225938566552901e-05,
|
| 299 |
+
"loss": 0.0991,
|
| 300 |
+
"step": 13500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 3.11,
|
| 304 |
+
"learning_rate": 1.1576791808873721e-05,
|
| 305 |
+
"loss": 0.0928,
|
| 306 |
+
"step": 14000
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 3.11,
|
| 310 |
+
"eval_cer": 0.010234491962117941,
|
| 311 |
+
"eval_loss": 0.04134270176291466,
|
| 312 |
+
"eval_runtime": 1343.5325,
|
| 313 |
+
"eval_samples_per_second": 11.084,
|
| 314 |
+
"eval_steps_per_second": 1.386,
|
| 315 |
+
"eval_wer": 0.056062656807077775,
|
| 316 |
+
"step": 14000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.23,
|
| 320 |
+
"learning_rate": 1.089556313993174e-05,
|
| 321 |
+
"loss": 0.0919,
|
| 322 |
+
"step": 14500
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 3.34,
|
| 326 |
+
"learning_rate": 1.0212969283276451e-05,
|
| 327 |
+
"loss": 0.0927,
|
| 328 |
+
"step": 15000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 3.34,
|
| 332 |
+
"eval_cer": 0.010227482996248802,
|
| 333 |
+
"eval_loss": 0.041016656905412674,
|
| 334 |
+
"eval_runtime": 1340.2656,
|
| 335 |
+
"eval_samples_per_second": 11.111,
|
| 336 |
+
"eval_steps_per_second": 1.389,
|
| 337 |
+
"eval_wer": 0.0564588010035653,
|
| 338 |
+
"step": 15000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.45,
|
| 342 |
+
"learning_rate": 9.53037542662116e-06,
|
| 343 |
+
"loss": 0.0896,
|
| 344 |
+
"step": 15500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.56,
|
| 348 |
+
"learning_rate": 8.84778156996587e-06,
|
| 349 |
+
"loss": 0.0886,
|
| 350 |
+
"step": 16000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.56,
|
| 354 |
+
"eval_cer": 0.01016019692390506,
|
| 355 |
+
"eval_loss": 0.040184538811445236,
|
| 356 |
+
"eval_runtime": 1343.3925,
|
| 357 |
+
"eval_samples_per_second": 11.085,
|
| 358 |
+
"eval_steps_per_second": 1.386,
|
| 359 |
+
"eval_wer": 0.055798560676086094,
|
| 360 |
+
"step": 16000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 3.67,
|
| 364 |
+
"learning_rate": 8.16518771331058e-06,
|
| 365 |
+
"loss": 0.0872,
|
| 366 |
+
"step": 16500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 3.78,
|
| 370 |
+
"learning_rate": 7.482593856655291e-06,
|
| 371 |
+
"loss": 0.091,
|
| 372 |
+
"step": 17000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.78,
|
| 376 |
+
"eval_cer": 0.010059267815389445,
|
| 377 |
+
"eval_loss": 0.03996647894382477,
|
| 378 |
+
"eval_runtime": 1343.622,
|
| 379 |
+
"eval_samples_per_second": 11.083,
|
| 380 |
+
"eval_steps_per_second": 1.386,
|
| 381 |
+
"eval_wer": 0.05525386240591575,
|
| 382 |
+
"step": 17000
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 3.89,
|
| 386 |
+
"learning_rate": 6.8e-06,
|
| 387 |
+
"loss": 0.0882,
|
| 388 |
+
"step": 17500
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.0,
|
| 392 |
+
"learning_rate": 6.11740614334471e-06,
|
| 393 |
+
"loss": 0.0888,
|
| 394 |
+
"step": 18000
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 4.0,
|
| 398 |
+
"eval_cer": 0.009962544086395317,
|
| 399 |
+
"eval_loss": 0.039790086448192596,
|
| 400 |
+
"eval_runtime": 1340.4959,
|
| 401 |
+
"eval_samples_per_second": 11.109,
|
| 402 |
+
"eval_steps_per_second": 1.389,
|
| 403 |
+
"eval_wer": 0.05461838109071702,
|
| 404 |
+
"step": 18000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.12,
|
| 408 |
+
"learning_rate": 5.43617747440273e-06,
|
| 409 |
+
"loss": 0.0872,
|
| 410 |
+
"step": 18500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 4.23,
|
| 414 |
+
"learning_rate": 4.75358361774744e-06,
|
| 415 |
+
"loss": 0.0885,
|
| 416 |
+
"step": 19000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.23,
|
| 420 |
+
"eval_cer": 0.009899463393573059,
|
| 421 |
+
"eval_loss": 0.039466630667448044,
|
| 422 |
+
"eval_runtime": 1341.4804,
|
| 423 |
+
"eval_samples_per_second": 11.101,
|
| 424 |
+
"eval_steps_per_second": 1.388,
|
| 425 |
+
"eval_wer": 0.05423048989832299,
|
| 426 |
+
"step": 19000
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 4.34,
|
| 430 |
+
"learning_rate": 4.0709897610921505e-06,
|
| 431 |
+
"loss": 0.0885,
|
| 432 |
+
"step": 19500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 4.45,
|
| 436 |
+
"learning_rate": 3.3897610921501705e-06,
|
| 437 |
+
"loss": 0.0869,
|
| 438 |
+
"step": 20000
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 4.45,
|
| 442 |
+
"eval_cer": 0.009868623943748843,
|
| 443 |
+
"eval_loss": 0.0393984392285347,
|
| 444 |
+
"eval_runtime": 1341.5672,
|
| 445 |
+
"eval_samples_per_second": 11.1,
|
| 446 |
+
"eval_steps_per_second": 1.388,
|
| 447 |
+
"eval_wer": 0.05402416479598574,
|
| 448 |
+
"step": 20000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.56,
|
| 452 |
+
"learning_rate": 2.708532423208191e-06,
|
| 453 |
+
"loss": 0.0845,
|
| 454 |
+
"step": 20500
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 4.67,
|
| 458 |
+
"learning_rate": 2.025938566552901e-06,
|
| 459 |
+
"loss": 0.0844,
|
| 460 |
+
"step": 21000
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 4.67,
|
| 464 |
+
"eval_cer": 0.009830775528055488,
|
| 465 |
+
"eval_loss": 0.03927738592028618,
|
| 466 |
+
"eval_runtime": 1339.2196,
|
| 467 |
+
"eval_samples_per_second": 11.12,
|
| 468 |
+
"eval_steps_per_second": 1.39,
|
| 469 |
+
"eval_wer": 0.05393338175095735,
|
| 470 |
+
"step": 21000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.78,
|
| 474 |
+
"learning_rate": 1.3433447098976108e-06,
|
| 475 |
+
"loss": 0.0866,
|
| 476 |
+
"step": 21500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.89,
|
| 480 |
+
"learning_rate": 6.607508532423209e-07,
|
| 481 |
+
"loss": 0.0882,
|
| 482 |
+
"step": 22000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 4.89,
|
| 486 |
+
"eval_cer": 0.009790123526014478,
|
| 487 |
+
"eval_loss": 0.039073262363672256,
|
| 488 |
+
"eval_runtime": 1341.9861,
|
| 489 |
+
"eval_samples_per_second": 11.097,
|
| 490 |
+
"eval_steps_per_second": 1.387,
|
| 491 |
+
"eval_wer": 0.053735309652713587,
|
| 492 |
+
"step": 22000
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 5.0,
|
| 496 |
+
"step": 22475,
|
| 497 |
+
"total_flos": 1.3602795268583408e+20,
|
| 498 |
+
"train_loss": 0.08836701542702612,
|
| 499 |
+
"train_runtime": 131691.1828,
|
| 500 |
+
"train_samples_per_second": 5.462,
|
| 501 |
+
"train_steps_per_second": 0.171
|
| 502 |
+
}
|
| 503 |
+
],
|
| 504 |
+
"max_steps": 22475,
|
| 505 |
+
"num_train_epochs": 5,
|
| 506 |
+
"total_flos": 1.3602795268583408e+20,
|
| 507 |
+
"trial_name": null,
|
| 508 |
+
"trial_params": null
|
| 509 |
+
}
|