mt5-zh2en
This model is a fine-tuned version of Takq/mt5-zh2en on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.0037
- Bleu: 9.0726
- Gen Len: 18.1282
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: 0.0001
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
---|---|---|---|---|---|
2.3589 | 1.0 | 258072 | 2.0607 | 8.7641 | 18.1171 |
2.2457 | 2.0 | 516144 | 2.0037 | 9.0726 | 18.1282 |
Framework versions
- Transformers 4.36.2
- Pytorch 1.11.0+cu113
- Datasets 2.1.0
- Tokenizers 0.15.1
- Downloads last month
- 1