testing_mt5-ha2zh-translation

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: 17.4027
• Bleu: 0.0
• Ter: 123.9865
• Chrf: 0.0587
• Bertscore Precision: 0.4933
• Bertscore Recall: 0.4781
• Bertscore F1: 0.4852

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• 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
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Ter	Chrf	Bertscore Precision	Bertscore Recall	Bertscore F1
26.555	1.0	63	17.4027	0.0	123.9865	0.0587	0.4933	0.4781	0.4852
23.3851	2.0	126	15.8128	0.0	125.3378	0.0587	0.4934	0.4783	0.4853
21.9346	3.0	189	15.2208	0.0	133.7838	0.0588	0.4930	0.4785	0.4853

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0