update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
datasets:
|
| 5 |
+
- audiofolder
|
| 6 |
+
metrics:
|
| 7 |
+
- wer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: wav2vec2-large-xls-r-300m-pashto-colab-test-6
|
| 10 |
+
results:
|
| 11 |
+
- task:
|
| 12 |
+
name: Automatic Speech Recognition
|
| 13 |
+
type: automatic-speech-recognition
|
| 14 |
+
dataset:
|
| 15 |
+
name: audiofolder
|
| 16 |
+
type: audiofolder
|
| 17 |
+
config: default
|
| 18 |
+
split: train
|
| 19 |
+
args: default
|
| 20 |
+
metrics:
|
| 21 |
+
- name: Wer
|
| 22 |
+
type: wer
|
| 23 |
+
value: 1.0
|
| 24 |
+
---
|
| 25 |
+
|
| 26 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 27 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 28 |
+
|
| 29 |
+
# wav2vec2-large-xls-r-300m-pashto-colab-test-6
|
| 30 |
+
|
| 31 |
+
This model is a fine-tuned version of [rsd16/wav2vec2-large-xlsr-53-fine-tuned-farsi](https://huggingface.co/rsd16/wav2vec2-large-xlsr-53-fine-tuned-farsi) on the audiofolder dataset.
|
| 32 |
+
It achieves the following results on the evaluation set:
|
| 33 |
+
- Loss: nan
|
| 34 |
+
- Wer: 1.0
|
| 35 |
+
|
| 36 |
+
## Model description
|
| 37 |
+
|
| 38 |
+
More information needed
|
| 39 |
+
|
| 40 |
+
## Intended uses & limitations
|
| 41 |
+
|
| 42 |
+
More information needed
|
| 43 |
+
|
| 44 |
+
## Training and evaluation data
|
| 45 |
+
|
| 46 |
+
More information needed
|
| 47 |
+
|
| 48 |
+
## Training procedure
|
| 49 |
+
|
| 50 |
+
### Training hyperparameters
|
| 51 |
+
|
| 52 |
+
The following hyperparameters were used during training:
|
| 53 |
+
- learning_rate: 0.9
|
| 54 |
+
- train_batch_size: 1
|
| 55 |
+
- eval_batch_size: 8
|
| 56 |
+
- seed: 42
|
| 57 |
+
- gradient_accumulation_steps: 10
|
| 58 |
+
- total_train_batch_size: 10
|
| 59 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 60 |
+
- lr_scheduler_type: linear
|
| 61 |
+
- lr_scheduler_warmup_steps: 10
|
| 62 |
+
- num_epochs: 15
|
| 63 |
+
- mixed_precision_training: Native AMP
|
| 64 |
+
|
| 65 |
+
### Training results
|
| 66 |
+
|
| 67 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 68 |
+
|:----------------:|:-----:|:----:|:---------------:|:---:|
|
| 69 |
+
| 1860014104903.68 | 0.96 | 100 | nan | 1.0 |
|
| 70 |
+
| 0.0 | 1.91 | 200 | nan | 1.0 |
|
| 71 |
+
| 0.0 | 2.87 | 300 | nan | 1.0 |
|
| 72 |
+
| 0.0 | 3.82 | 400 | nan | 1.0 |
|
| 73 |
+
| 0.0 | 4.78 | 500 | nan | 1.0 |
|
| 74 |
+
| 0.0 | 5.73 | 600 | nan | 1.0 |
|
| 75 |
+
| 0.0 | 6.69 | 700 | nan | 1.0 |
|
| 76 |
+
| 0.0 | 7.64 | 800 | nan | 1.0 |
|
| 77 |
+
| 0.0 | 8.6 | 900 | nan | 1.0 |
|
| 78 |
+
| 0.0 | 9.55 | 1000 | nan | 1.0 |
|
| 79 |
+
| 0.0 | 10.51 | 1100 | nan | 1.0 |
|
| 80 |
+
| 0.0 | 11.46 | 1200 | nan | 1.0 |
|
| 81 |
+
| 0.0 | 12.42 | 1300 | nan | 1.0 |
|
| 82 |
+
| 0.0 | 13.37 | 1400 | nan | 1.0 |
|
| 83 |
+
| 0.0 | 14.33 | 1500 | nan | 1.0 |
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
### Framework versions
|
| 87 |
+
|
| 88 |
+
- Transformers 4.29.2
|
| 89 |
+
- Pytorch 2.0.1+cu118
|
| 90 |
+
- Datasets 2.12.0
|
| 91 |
+
- Tokenizers 0.13.3
|