metadata
language:
- zh
- yue
base_model: indiejoseph/bart-base-cantonese
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: bart-translation-zh-yue
results: []
bart-translation-zh-yue
This model is a fine-tuned version of indiejoseph/bart-base-cantonese on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1612
- Bleu: 88.1708
- Gen Len: 41.6859
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 0.2219 | 1.0 | 2336 | 0.1816 | 21.2068 | 19.2422 |
| 0.1479 | 2.0 | 4672 | 0.1609 | 21.6783 | 19.2471 |
| 0.1092 | 3.0 | 7008 | 0.1534 | 21.8526 | 19.2463 |
| 0.0856 | 4.0 | 9344 | 0.1525 | 22.0841 | 19.2482 |
| 0.0667 | 5.0 | 11680 | 0.1588 | 22.1943 | 19.2467 |
| 0.0549 | 6.0 | 14016 | 0.1612 | 22.2237 | 19.2467 |
Framework versions
- Transformers 4.35.0.dev0
- Pytorch 2.1.1+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1