Household_Robot
Collection
The final models which are deployed to FYP's Household Robot. • 4 items • Updated
MedicGrabber_WhisperSmall
This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1853
- Wer Ortho: 10.4950
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 4
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho |
|---|---|---|---|---|
| No log | 0 | 0 | 3.3948 | 99.2079 |
| 1.2079 | 1.0 | 29 | 0.1921 | 14.6535 |
| 0.1011 | 2.0 | 58 | 0.2017 | 12.0792 |
| 0.0393 | 3.0 | 87 | 0.1752 | 10.0990 |
| 0.0221 | 4.0 | 116 | 0.1907 | 9.9010 |
| 0.005 | 5.0 | 145 | 0.1853 | 10.4950 |
Framework versions
- Transformers 4.50.3
- Pytorch 2.5.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1
- Downloads last month
- 1
Model tree for borisPMC/HouseHolder_WhisperSmall
Base model
openai/whisper-small