<!-- 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. -->
mt5-text-to-ipa
This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0653
- Wer: 0.0536
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: 0.005
- train_batch_size: 8
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2000
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.8729 | 1.62 | 4000 | 0.9909 | 1.1948 |
| 0.8001 | 3.23 | 8000 | 0.2286 | 0.3110 |
| 0.3642 | 4.85 | 12000 | 0.1414 | 0.1860 |
| 0.22 | 6.46 | 16000 | 0.1037 | 0.1154 |
| 0.1289 | 8.08 | 20000 | 0.0777 | 0.0707 |
| 0.0762 | 9.7 | 24000 | 0.0653 | 0.0536 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.14.1
