|
|
--- |
|
|
library_name: peft |
|
|
language: |
|
|
- en |
|
|
license: apache-2.0 |
|
|
base_model: openai/whisper-small.en |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- wer |
|
|
model-index: |
|
|
- name: Whisper Small NSC part 1,2,3 (2000 steps) - Jarrett Er |
|
|
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 Small NSC part 1,2,3 (2000 steps) - Jarrett Er |
|
|
|
|
|
This model is a fine-tuned version of [openai/whisper-small.en](https://huggingface.co/openai/whisper-small.en) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.9590 |
|
|
- Wer: 5.3473 |
|
|
|
|
|
## 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: 0.0001 |
|
|
- 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: 200 |
|
|
- training_steps: 2000 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|
|:-------------:|:------:|:----:|:---------------:|:--------:| |
|
|
| 4.645 | 0.2008 | 50 | 4.1237 | 140.7387 | |
|
|
| 2.1925 | 0.4016 | 100 | 1.7341 | 28.5557 | |
|
|
| 1.611 | 0.6024 | 150 | 1.6017 | 22.2712 | |
|
|
| 1.5577 | 0.8032 | 200 | 1.5606 | 21.5546 | |
|
|
| 1.5236 | 1.0040 | 250 | 1.5283 | 24.0077 | |
|
|
| 1.4578 | 1.2048 | 300 | 1.4763 | 20.7552 | |
|
|
| 1.3667 | 1.4056 | 350 | 1.4251 | 19.3495 | |
|
|
| 1.3611 | 1.6064 | 400 | 1.3817 | 18.3572 | |
|
|
| 1.3913 | 1.8072 | 450 | 1.3487 | 18.1918 | |
|
|
| 1.3194 | 2.0080 | 500 | 1.3160 | 16.9515 | |
|
|
| 1.2678 | 2.2088 | 550 | 1.2973 | 15.9592 | |
|
|
| 1.1661 | 2.4096 | 600 | 1.2656 | 14.7740 | |
|
|
| 1.2014 | 2.6104 | 650 | 1.2339 | 14.3330 | |
|
|
| 1.1511 | 2.8112 | 700 | 1.2054 | 14.0573 | |
|
|
| 1.1239 | 3.0120 | 750 | 1.1740 | 13.0375 | |
|
|
| 1.0758 | 3.2129 | 800 | 1.1534 | 12.7343 | |
|
|
| 1.1137 | 3.4137 | 850 | 1.1401 | 11.7971 | |
|
|
| 1.0582 | 3.6145 | 900 | 1.1203 | 10.8324 | |
|
|
| 1.0749 | 3.8153 | 950 | 1.1061 | 10.5568 | |
|
|
| 1.0305 | 4.0161 | 1000 | 1.0951 | 10.6119 | |
|
|
| 1.0105 | 4.2169 | 1050 | 1.0773 | 9.7299 | |
|
|
| 1.0031 | 4.4177 | 1100 | 1.0757 | 9.9228 | |
|
|
| 1.0054 | 4.6185 | 1150 | 1.0621 | 8.8754 | |
|
|
| 0.9836 | 4.8193 | 1200 | 1.0455 | 8.4895 | |
|
|
| 0.9663 | 5.0201 | 1250 | 1.0369 | 7.8556 | |
|
|
| 0.9473 | 5.2209 | 1300 | 1.0261 | 7.8556 | |
|
|
| 0.9784 | 5.4217 | 1350 | 1.0184 | 7.7178 | |
|
|
| 0.9462 | 5.6225 | 1400 | 1.0124 | 7.6626 | |
|
|
| 0.9414 | 5.8233 | 1450 | 1.0069 | 7.2492 | |
|
|
| 0.9262 | 6.0241 | 1500 | 1.0005 | 7.0562 | |
|
|
| 0.9127 | 6.2249 | 1550 | 0.9946 | 6.8633 | |
|
|
| 0.9324 | 6.4257 | 1600 | 0.9866 | 6.3120 | |
|
|
| 0.9043 | 6.6265 | 1650 | 0.9793 | 6.3947 | |
|
|
| 0.9223 | 6.8273 | 1700 | 0.9725 | 5.5678 | |
|
|
| 0.9087 | 7.0281 | 1750 | 0.9697 | 5.4576 | |
|
|
| 0.8943 | 7.2289 | 1800 | 0.9659 | 5.6781 | |
|
|
| 0.8962 | 7.4297 | 1850 | 0.9634 | 5.5678 | |
|
|
| 0.8992 | 7.6305 | 1900 | 0.9611 | 5.4851 | |
|
|
| 0.9099 | 7.8313 | 1950 | 0.9601 | 5.4851 | |
|
|
| 0.9182 | 8.0321 | 2000 | 0.9590 | 5.3473 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.14.0 |
|
|
- Transformers 4.45.2 |
|
|
- Pytorch 2.5.1+cu124 |
|
|
- Datasets 3.2.1.dev0 |
|
|
- Tokenizers 0.20.3 |
