|
--- |
|
library_name: transformers |
|
license: cc-by-nc-4.0 |
|
base_model: facebook/mms-1b-all |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: mms-1b-bemgen-male-model-test |
|
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. --> |
|
|
|
# mms-1b-bemgen-male-model-test |
|
|
|
This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3060 |
|
- Wer: 0.4447 |
|
|
|
## 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: 0.0003 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 30.0 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:------:|:----:|:---------------:|:------:| |
|
| 6.9809 | 0.1034 | 100 | 1.3139 | 0.9957 | |
|
| 0.745 | 0.2068 | 200 | 0.4297 | 0.5882 | |
|
| 0.5423 | 0.3102 | 300 | 0.3886 | 0.5644 | |
|
| 0.539 | 0.4137 | 400 | 0.3683 | 0.5448 | |
|
| 0.5277 | 0.5171 | 500 | 0.3529 | 0.5083 | |
|
| 0.4708 | 0.6205 | 600 | 0.3493 | 0.4997 | |
|
| 0.4889 | 0.7239 | 700 | 0.3467 | 0.5096 | |
|
| 0.4793 | 0.8273 | 800 | 0.3407 | 0.4818 | |
|
| 0.469 | 0.9307 | 900 | 0.3455 | 0.4958 | |
|
| 0.4407 | 1.0341 | 1000 | 0.3329 | 0.4735 | |
|
| 0.4524 | 1.1375 | 1100 | 0.3289 | 0.4879 | |
|
| 0.4416 | 1.2410 | 1200 | 0.3280 | 0.4911 | |
|
| 0.4599 | 1.3444 | 1300 | 0.3285 | 0.4765 | |
|
| 0.4739 | 1.4478 | 1400 | 0.3221 | 0.4694 | |
|
| 0.4466 | 1.5512 | 1500 | 0.3196 | 0.4588 | |
|
| 0.4483 | 1.6546 | 1600 | 0.3144 | 0.4526 | |
|
| 0.4543 | 1.7580 | 1700 | 0.3170 | 0.4528 | |
|
| 0.4537 | 1.8614 | 1800 | 0.3141 | 0.4522 | |
|
| 0.4293 | 1.9648 | 1900 | 0.3106 | 0.4453 | |
|
| 0.4457 | 2.0683 | 2000 | 0.3134 | 0.4651 | |
|
| 0.4214 | 2.1717 | 2100 | 0.3119 | 0.4543 | |
|
| 0.4103 | 2.2751 | 2200 | 0.3089 | 0.4391 | |
|
| 0.407 | 2.3785 | 2300 | 0.3053 | 0.4331 | |
|
| 0.4314 | 2.4819 | 2400 | 0.3059 | 0.4337 | |
|
| 0.4144 | 2.5853 | 2500 | 0.3054 | 0.4382 | |
|
| 0.4099 | 2.6887 | 2600 | 0.3060 | 0.4447 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.47.1 |
|
- Pytorch 2.5.1+cu124 |
|
- Datasets 3.2.0 |
|
- Tokenizers 0.21.0 |
|
|
