cantonese-chinese-translation-v8

This model is a fine-tuned version of fnlp/bart-base-chinese on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9569
• Bleu: 36.4877
• Chrf: 34.1701
• Gen Len: 13.2081

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Chrf	Gen Len
1.4049	1.0	3435	0.9598	35.9507	33.7741	13.1283
1.1958	2.0	6871	0.9569	36.4877	34.1701	13.2081
1.0563	3.0	10307	1.0015	34.4657	32.1094	13.2484
0.9351	4.0	13743	1.0421	33.8705	31.4738	13.2731
0.8285	5.0	17178	1.1068	32.7811	30.479	13.2466

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu117
• Datasets 2.9.0
• Tokenizers 0.12.1