File size: 1,370 Bytes
3cb5768 726a713 3cb5768 726a713 3cb5768 726a713 3cb5768 726a713 3cb5768 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
---
license: mit
base_model: distil-whisper/distil-large-v3
tags:
- generated_from_trainer
model-index:
- name: distilwhisper_finetune
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. -->
# distilwhisper_finetune
This model is a fine-tuned version of [distil-whisper/distil-large-v3](https://huggingface.co/distil-whisper/distil-large-v3) on an unknown dataset.
It achieves the following results on the evaluation set:
- eval_loss: 0.0432
- eval_wer: 3.2604
- eval_runtime: 848.1823
- eval_samples_per_second: 0.825
- eval_steps_per_second: 0.104
- epoch: 1.7857
- step: 250
## 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: 1e-05
- train_batch_size: 20
- 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: 100
- training_steps: 1000
- mixed_precision_training: Native AMP
### Framework versions
- Transformers 4.41.0
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
|