metadata
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_Quechua_finetuned
results: []
MMS_Quechua_finetuned
This model is a fine-tuned version of facebook/mms-1b-all on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4424
- Wer: 0.3950
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: 0.001
- train_batch_size: 16
- eval_batch_size: 8
- 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: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 5.7203 | 0.1355 | 100 | 0.7301 | 0.6115 |
| 0.731 | 0.2710 | 200 | 0.5409 | 0.4553 |
| 0.5859 | 0.4065 | 300 | 0.5124 | 0.4280 |
| 0.549 | 0.5420 | 400 | 0.4815 | 0.4170 |
| 0.5341 | 0.6775 | 500 | 0.4720 | 0.4051 |
| 1.0055 | 0.8130 | 600 | 0.4601 | 0.4095 |
| 0.4943 | 0.9485 | 700 | 0.4424 | 0.3950 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.4.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3