update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: wav2vec2-base
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# wav2vec2-base
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 3.0808
|
| 18 |
+
- Wer: 1.0
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 0.0001
|
| 38 |
+
- train_batch_size: 4
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
+
- lr_scheduler_type: linear
|
| 43 |
+
- lr_scheduler_warmup_steps: 1000
|
| 44 |
+
- num_epochs: 15
|
| 45 |
+
- mixed_precision_training: Native AMP
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 50 |
+
|:-------------:|:-----:|:-----:|:---------------:|:---:|
|
| 51 |
+
| 3.7118 | 0.5 | 500 | 3.0635 | 1.0 |
|
| 52 |
+
| 2.9533 | 1.01 | 1000 | 3.0383 | 1.0 |
|
| 53 |
+
| 2.9493 | 1.51 | 1500 | 3.0638 | 1.0 |
|
| 54 |
+
| 2.9495 | 2.01 | 2000 | 3.0554 | 1.0 |
|
| 55 |
+
| 2.9468 | 2.51 | 2500 | 3.0630 | 1.0 |
|
| 56 |
+
| 2.9493 | 3.02 | 3000 | 3.0530 | 1.0 |
|
| 57 |
+
| 2.9457 | 3.52 | 3500 | 3.0534 | 1.0 |
|
| 58 |
+
| 2.9492 | 4.02 | 4000 | 3.0357 | 1.0 |
|
| 59 |
+
| 2.9444 | 4.52 | 4500 | 3.0366 | 1.0 |
|
| 60 |
+
| 2.9495 | 5.03 | 5000 | 3.0412 | 1.0 |
|
| 61 |
+
| 2.9468 | 5.53 | 5500 | 3.0331 | 1.0 |
|
| 62 |
+
| 2.9453 | 6.03 | 6000 | 3.0847 | 1.0 |
|
| 63 |
+
| 2.9484 | 6.53 | 6500 | 3.0661 | 1.0 |
|
| 64 |
+
| 2.9457 | 7.04 | 7000 | 3.0769 | 1.0 |
|
| 65 |
+
| 2.9449 | 7.54 | 7500 | 3.0701 | 1.0 |
|
| 66 |
+
| 2.9453 | 8.04 | 8000 | 3.1072 | 1.0 |
|
| 67 |
+
| 2.9436 | 8.54 | 8500 | 3.1043 | 1.0 |
|
| 68 |
+
| 2.9474 | 9.05 | 9000 | 3.0902 | 1.0 |
|
| 69 |
+
| 2.9452 | 9.55 | 9500 | 3.0879 | 1.0 |
|
| 70 |
+
| 2.9443 | 10.05 | 10000 | 3.1112 | 1.0 |
|
| 71 |
+
| 2.9436 | 10.55 | 10500 | 3.0946 | 1.0 |
|
| 72 |
+
| 2.9469 | 11.06 | 11000 | 3.0812 | 1.0 |
|
| 73 |
+
| 2.9434 | 11.56 | 11500 | 3.1112 | 1.0 |
|
| 74 |
+
| 2.9442 | 12.06 | 12000 | 3.0855 | 1.0 |
|
| 75 |
+
| 2.9436 | 12.56 | 12500 | 3.0786 | 1.0 |
|
| 76 |
+
| 2.9425 | 13.07 | 13000 | 3.0789 | 1.0 |
|
| 77 |
+
| 2.9418 | 13.57 | 13500 | 3.0786 | 1.0 |
|
| 78 |
+
| 2.9443 | 14.07 | 14000 | 3.0798 | 1.0 |
|
| 79 |
+
| 2.9449 | 14.57 | 14500 | 3.0808 | 1.0 |
|
| 80 |
+
|
| 81 |
+
|
| 82 |
+
### Framework versions
|
| 83 |
+
|
| 84 |
+
- Transformers 4.11.3
|
| 85 |
+
- Pytorch 1.10.2
|
| 86 |
+
- Datasets 1.18.3
|
| 87 |
+
- Tokenizers 0.10.3
|