Downloads · 30 days
17
52% of all-time downloads
torgo-whisper-lg-3-Nov29
sqrk/torgo-whisper-lg-3-Nov29
torgo-whisper-lg-3-Nov29 is a machine learning model from sqrk. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
33
Public
Parameters
1.5B
135 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors6.2 GB · 100%
At a glance
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Nov 29, 2024
- Updated
- Nov 30, 2024
- SHA
- d3ab06df
Base models
Model card
From the Hugging Face model README
<!-- 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. -->
torgo-whisper-lg-3-Nov29
This model is a fine-tuned version of openai/whisper-large-v3 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0550
- Wer: 3.9596
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 100
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.7411 | 0.3868 | 100 | 0.2427 | 18.4499 |
| 0.2015 | 0.7737 | 200 | 0.1579 | 13.2266 |
| 0.1213 | 1.1605 | 300 | 0.1563 | 11.0362 |
| 0.0746 | 1.5474 | 400 | 0.1201 | 9.8568 |
| 0.0809 | 1.9342 | 500 | 0.0970 | 7.3294 |
| 0.037 | 2.3211 | 600 | 0.0907 | 6.8239 |
| 0.035 | 2.7079 | 700 | 0.0964 | 7.2452 |
| 0.0347 | 3.0948 | 800 | 0.0881 | 7.3294 |
| 0.0352 | 3.4816 | 900 | 0.0869 | 7.7506 |
| 0.0215 | 3.8685 | 1000 | 0.0821 | 7.4136 |
| 0.0224 | 4.2553 | 1100 | 0.0744 | 6.4027 |
| 0.0263 | 4.6422 | 1200 | 0.0783 | 7.8349 |
| 0.0173 | 5.0290 | 1300 | 0.0837 | 7.8349 |
| 0.0166 | 5.4159 | 1400 | 0.0933 | 5.8130 |
| 0.017 | 5.8027 | 1500 | 0.1006 | 7.4136 |
| 0.0157 | 6.1896 | 1600 | 0.0916 | 6.1500 |
| 0.0083 | 6.5764 | 1700 | 0.0659 | 5.3917 |
| 0.0079 | 6.9632 | 1800 | 0.0956 | 6.9924 |
| 0.0115 | 7.3501 | 1900 | 0.0703 | 6.2342 |
| 0.0073 | 7.7369 | 2000 | 0.0696 | 5.8130 |
| 0.0104 | 8.1238 | 2100 | 0.0787 | 6.3184 |
| 0.0093 | 8.5106 | 2200 | 0.0824 | 6.4027 |
| 0.0097 | 8.8975 | 2300 | 0.1042 | 7.4979 |
| 0.0094 | 9.2843 | 2400 | 0.0808 | 6.1500 |
| 0.0026 | 9.6712 | 2500 | 0.0789 | 6.6554 |
| 0.0059 | 10.0580 | 2600 | 0.0966 | 7.2452 |
| 0.0044 | 10.4449 | 2700 | 0.0882 | 8.7616 |
| 0.0059 | 10.8317 | 2800 | 0.0806 | 7.4136 |
| 0.0067 | 11.2186 | 2900 | 0.0951 | 7.6664 |
| 0.0057 | 11.6054 | 3000 | 0.0837 | 5.8972 |
| 0.0086 | 11.9923 | 3100 | 0.0774 | 5.6445 |
| 0.0096 | 12.3791 | 3200 | 0.0888 | 5.7287 |
| 0.0063 | 12.7660 | 3300 | 0.0763 | 5.0548 |
| 0.0107 | 13.1528 | 3400 | 0.0727 | 4.6335 |
| 0.0043 | 13.5397 | 3500 | 0.0886 | 6.4869 |
| 0.004 | 13.9265 | 3600 | 0.0712 | 5.3075 |
| 0.0083 | 14.3133 | 3700 | 0.0601 | 4.2123 |
| 0.002 | 14.7002 | 3800 | 0.0674 | 5.3917 |
| 0.0039 | 15.0870 | 3900 | 0.0632 | 4.2965 |
| 0.0019 | 15.4739 | 4000 | 0.0642 | 4.3808 |
| 0.0026 | 15.8607 | 4100 | 0.0699 | 4.6335 |
| 0.0031 | 16.2476 | 4200 | 0.0608 | 4.2123 |
| 0.0031 | 16.6344 | 4300 | 0.0579 | 3.8753 |
| 0.001 | 17.0213 | 4400 | 0.0608 | 4.2123 |
| 0.002 | 17.4081 | 4500 | 0.0550 | 3.9596 |
Framework versions
- Transformers 4.43.4
- Pytorch 2.4.1
- Datasets 3.0.0
- Tokenizers 0.19.1
- Type
- whisper
- License
- apache-2.0
- Created
- Nov 29, 2024
- Updated
- Nov 30, 2024