pretrain_Law_model_vit5_version1

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: 0.2779
• Rouge1: 0.4859
• Rouge2: 0.3617
• Rougel: 0.4218
• Rougelsum: 0.4417

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
No log	1.0	245	0.3566	0.4739	0.3369	0.4053	0.4273
No log	2.0	490	0.3240	0.4752	0.3453	0.4095	0.4300
0.7518	3.0	735	0.3059	0.4760	0.3510	0.4112	0.4311
0.7518	4.0	980	0.2951	0.4838	0.3584	0.4164	0.4387
0.2808	5.0	1225	0.2858	0.4799	0.3582	0.4166	0.4368
0.2808	6.0	1470	0.2831	0.4839	0.3611	0.4194	0.4403
0.2351	7.0	1715	0.2814	0.4858	0.3644	0.4218	0.4423
0.2351	8.0	1960	0.2779	0.4850	0.3612	0.4206	0.4416
0.2074	9.0	2205	0.2775	0.4836	0.3590	0.4199	0.4398
0.2074	10.0	2450	0.2779	0.4859	0.3617	0.4218	0.4417

Framework versions

• Transformers 4.32.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3