|
--- |
|
base_model: openai/whisper-large-v2 |
|
library_name: peft |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: whisper-large-v2-ft-cv16-1__car145_ref-tms_e3n4-n4r2_owner12-copy2x-241212-v1 |
|
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. --> |
|
|
|
# whisper-large-v2-ft-cv16-1__car145_ref-tms_e3n4-n4r2_owner12-copy2x-241212-v1 |
|
|
|
This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1178 |
|
|
|
## 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: 64 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.2 |
|
- num_epochs: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 3.7208 | 1.0 | 139 | 1.4782 | |
|
| 0.6163 | 2.0 | 278 | 0.1211 | |
|
| 0.1284 | 3.0 | 417 | 0.1084 | |
|
| 0.1026 | 4.0 | 556 | 0.1070 | |
|
| 0.085 | 5.0 | 695 | 0.1069 | |
|
| 0.0714 | 6.0 | 834 | 0.1087 | |
|
| 0.0617 | 7.0 | 973 | 0.1112 | |
|
| 0.054 | 8.0 | 1112 | 0.1141 | |
|
| 0.0482 | 9.0 | 1251 | 0.1163 | |
|
| 0.0442 | 10.0 | 1390 | 0.1178 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.13.0 |
|
- Transformers 4.45.1 |
|
- Pytorch 2.5.0+cu124 |
|
- Datasets 3.0.1 |
|
- Tokenizers 0.20.0 |
