metadata
language:
- es
- hch
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: byt5-base-es_hch
results: []
byt5-base-es_hch
This model is a fine-tuned version of google/byt5-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8412
- Bleu: 2.9725
- Gen Len: 98.4805
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 65
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| No log | 1.0 | 398 | 1.1778 | 0.0 | 19.0 |
| 1.6138 | 2.0 | 796 | 1.0504 | 0.0729 | 19.0 |
| 1.1972 | 3.0 | 1194 | 0.9888 | 0.0547 | 19.0 |
| 1.0888 | 4.0 | 1592 | 0.9424 | 0.0841 | 19.0 |
| 1.0888 | 5.0 | 1990 | 0.9078 | 0.0757 | 19.0 |
| 1.0299 | 6.0 | 2388 | 0.8801 | 0.0784 | 19.0 |
| 0.9817 | 7.0 | 2786 | 0.8627 | 0.0764 | 19.0 |
| 0.9466 | 8.0 | 3184 | 0.8503 | 0.0784 | 19.0 |
| 0.9286 | 9.0 | 3582 | 0.8432 | 0.0781 | 19.0 |
| 0.9286 | 10.0 | 3980 | 0.8412 | 0.0773 | 19.0 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3