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-large-v2 on the ORD_0.9 dataset. It achieves the following results on the evaluation set:
- Loss: 0.9589
- Wer: 40.3477
- Cer: 24.1352
- Clean Wer: 31.1530
- Clean Cer: 18.5577
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 |
|---|---|---|---|---|---|---|---|
| 0.9554 | 1.0 | 1049 | 27.0639 | 21.0162 | 36.2166 | 1.1351 | 46.3955 |
| 0.8713 | 2.0 | 2098 | 25.9010 | 20.0561 | 33.3992 | 1.0961 | 43.5685 |
| 0.8168 | 3.0 | 3147 | 0.9695 | 41.6888 | 24.5027 | 32.0230 | 18.8609 |
| 0.7647 | 4.0 | 4196 | 0.9589 | 40.3477 | 24.1352 | 31.1530 | 18.5577 |
Framework versions
- PEFT 0.12.0
- Transformers 4.41.0.dev0
- Pytorch 2.3.1
- Datasets 3.2.0
- Tokenizers 0.19.1
- Downloads last month
- 8
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for mizoru/whisper-large-ru-ORD_0.9_peft_wth_cv_0.1
Base model
openai/whisper-large-v2