azuur's picture
Training in progress, step 500
4019947
|
raw
history blame
1.61 kB
---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- common_voice
model-index:
- name: wav2vec2-base-gn-demo
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-base-gn-demo
This model is a fine-tuned version of [azuur/wav2vec2-base-gn-demo](https://huggingface.co/azuur/wav2vec2-base-gn-demo) on the common_voice dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2750
- Wer: 0.7912
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 300
- num_epochs: 55
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 0.07 | 13.16 | 500 | 1.3797 | 0.8293 |
| 0.0711 | 26.32 | 1000 | 1.2878 | 0.8277 |
| 0.0454 | 39.47 | 1500 | 1.2782 | 0.7973 |
| 0.0281 | 52.63 | 2000 | 1.2750 | 0.7912 |
### Framework versions
- Transformers 4.11.3
- Pytorch 1.10.0+cu111
- Datasets 1.18.3
- Tokenizers 0.10.3