lora-vit5-finetuned

This model is a fine-tuned version of VietAI/vit5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0290
• Rouge1 F1: 0.1812
• Rouge2 F1: 0.1462
• Rougel F1: 0.1688
• Bleu: 0.0

Model description

The model is trained based on vit5-base, trained to summarize Vietnamese articles

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.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: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1 F1	Rouge2 F1	Rougel F1	Bleu
0.4567	1.0	1250	0.0438	0.1776	0.1374	0.1634	0.0
0.0523	2.0	2500	0.0347	0.1798	0.1415	0.1659	0.0
0.0438	3.0	3750	0.0313	0.1806	0.1448	0.1679	0.0
0.0393	4.0	5000	0.0295	0.1812	0.1452	0.1682	0.0
0.0371	5.0	6250	0.0290	0.1812	0.1462	0.1688	0.0

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1