vit5-base-vietnews-summarization-standardized-number

This model is a fine-tuned version of VietAI/vit5-base-vietnews-summarization on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8266
• Rouge1: 80.3861
• Rouge2: 75.4112
• Rougel: 80.0172
• Rougelsum: 79.9721
• Gen Len: 7.7504

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	297	0.6401	77.9012	72.733	77.2694	77.2675	8.9022
0.7223	2.0	594	0.5931	78.4299	73.3202	78.0444	78.0744	8.7437
0.7223	3.0	891	0.6182	79.3604	74.2323	78.907	78.9878	7.5497
0.3472	4.0	1188	0.6134	79.7284	74.8063	79.3057	79.3545	8.5245
0.3472	5.0	1485	0.6625	79.7185	74.5798	79.3356	79.3615	8.2749
0.2388	6.0	1782	0.7068	79.7276	74.6552	79.2519	79.2253	8.0489
0.1551	7.0	2079	0.7646	80.3805	75.394	79.9394	79.8915	7.9983
0.1551	8.0	2376	0.7736	81.2428	76.358	80.8152	80.8272	7.6121
0.1127	9.0	2673	0.8175	80.8907	76.0278	80.3848	80.3953	7.6543
0.1127	10.0	2970	0.8266	80.3861	75.4112	80.0172	79.9721	7.7504

Framework versions

• Transformers 4.33.0
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3