End of training
Browse files
README.md
CHANGED
|
@@ -5,9 +5,22 @@ tags:
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- mozilla-foundation/common_voice_6_1
|
|
|
|
|
|
|
| 8 |
model-index:
|
| 9 |
- name: Whisper Small Frisian 10m
|
| 10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -17,13 +30,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 6.1 dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
-
|
| 21 |
-
-
|
| 22 |
-
- eval_runtime: 569.6835
|
| 23 |
-
- eval_samples_per_second: 5.301
|
| 24 |
-
- eval_steps_per_second: 0.664
|
| 25 |
-
- epoch: 20.0
|
| 26 |
-
- step: 300
|
| 27 |
|
| 28 |
## Model description
|
| 29 |
|
|
@@ -42,7 +50,7 @@ More information needed
|
|
| 42 |
### Training hyperparameters
|
| 43 |
|
| 44 |
The following hyperparameters were used during training:
|
| 45 |
-
- learning_rate:
|
| 46 |
- train_batch_size: 8
|
| 47 |
- eval_batch_size: 8
|
| 48 |
- seed: 42
|
|
@@ -52,6 +60,22 @@ The following hyperparameters were used during training:
|
|
| 52 |
- training_steps: 1000
|
| 53 |
- mixed_precision_training: Native AMP
|
| 54 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
### Framework versions
|
| 56 |
|
| 57 |
- Transformers 4.40.1
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- mozilla-foundation/common_voice_6_1
|
| 8 |
+
metrics:
|
| 9 |
+
- wer
|
| 10 |
model-index:
|
| 11 |
- name: Whisper Small Frisian 10m
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Automatic Speech Recognition
|
| 15 |
+
type: automatic-speech-recognition
|
| 16 |
+
dataset:
|
| 17 |
+
name: Common Voice 6.1
|
| 18 |
+
type: mozilla-foundation/common_voice_6_1
|
| 19 |
+
args: 'config: frisian, split: test'
|
| 20 |
+
metrics:
|
| 21 |
+
- name: Wer
|
| 22 |
+
type: wer
|
| 23 |
+
value: 64.63375512386384
|
| 24 |
---
|
| 25 |
|
| 26 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 30 |
|
| 31 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 6.1 dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
+
- Loss: 1.6644
|
| 34 |
+
- Wer: 64.6338
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
|
| 36 |
## Model description
|
| 37 |
|
|
|
|
| 50 |
### Training hyperparameters
|
| 51 |
|
| 52 |
The following hyperparameters were used during training:
|
| 53 |
+
- learning_rate: 1e-06
|
| 54 |
- train_batch_size: 8
|
| 55 |
- eval_batch_size: 8
|
| 56 |
- seed: 42
|
|
|
|
| 60 |
- training_steps: 1000
|
| 61 |
- mixed_precision_training: Native AMP
|
| 62 |
|
| 63 |
+
### Training results
|
| 64 |
+
|
| 65 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 66 |
+
|:-------------:|:-------:|:----:|:---------------:|:-------:|
|
| 67 |
+
| 2.0563 | 6.6667 | 100 | 2.2740 | 83.8603 |
|
| 68 |
+
| 0.9188 | 13.3333 | 200 | 1.7748 | 76.3393 |
|
| 69 |
+
| 0.4133 | 20.0 | 300 | 1.6575 | 69.3138 |
|
| 70 |
+
| 0.1672 | 26.6667 | 400 | 1.6334 | 67.2750 |
|
| 71 |
+
| 0.0603 | 33.3333 | 500 | 1.6320 | 66.4160 |
|
| 72 |
+
| 0.0245 | 40.0 | 600 | 1.6434 | 65.3573 |
|
| 73 |
+
| 0.0138 | 46.6667 | 700 | 1.6523 | 64.7229 |
|
| 74 |
+
| 0.0104 | 53.3333 | 800 | 1.6592 | 64.0634 |
|
| 75 |
+
| 0.0089 | 60.0 | 900 | 1.6633 | 64.6551 |
|
| 76 |
+
| 0.0083 | 66.6667 | 1000 | 1.6644 | 64.6338 |
|
| 77 |
+
|
| 78 |
+
|
| 79 |
### Framework versions
|
| 80 |
|
| 81 |
- Transformers 4.40.1
|