Configuration Parsing
Warning:
In adapter_config.json: "peft.task_type" must be a string
Whisper Large Ru ORD 0.9 Peft PEFT 4-bit Q DoRA - Mizoru
This model is a fine-tuned version of openai/whisper-small on the ORD_0.9 dataset. It achieves the following results on the evaluation set:
- Loss: 0.9988
- Wer: 48.4439
- Cer: 26.5242
- Clean Wer: 40.8650
- Clean Cer: 20.9832
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: 0.001
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Cer | Clean Cer | Clean Wer | Validation Loss | Wer |
|---|---|---|---|---|---|---|---|
| 1.216 | 1.0 | 550 | 27.9352 | 22.0432 | 43.2693 | 1.0350 | 50.7505 |
| 1.1847 | 2.0 | 1100 | 26.5324 | 20.9303 | 41.2903 | 1.0187 | 49.1670 |
| 1.055 | 3.0 | 1650 | 26.7141 | 21.0494 | 41.5960 | 0.9889 | 48.8428 |
| 0.9137 | 4.0 | 2200 | 0.9988 | 48.4439 | 26.5242 | 40.8650 | 20.9832 |
Framework versions
- PEFT 0.11.2.dev0
- Transformers 4.41.0.dev0
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.19.1
- Downloads last month
- 2
Model tree for mizoru/whisper-large-ru-ORD_0.9_peft_0.2
Base model
openai/whisper-large-v2