pretrain_Law_model_vit5_version2

This model is a fine-tuned version of KingLTD/pretrain_Law_model_vit5_version1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1644
• Rouge1: 0.4902
• Rouge2: 0.3828
• Rougel: 0.4363
• Rougelsum: 0.4521

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	0.1692	0.4844	0.3722	0.4279	0.4439
No log	2.0	490	0.1664	0.4846	0.3719	0.4261	0.4442
0.2198	3.0	735	0.1636	0.4840	0.3752	0.4292	0.4443
0.2198	4.0	980	0.1628	0.4878	0.3754	0.4300	0.4468
0.178	5.0	1225	0.1621	0.4871	0.3791	0.4349	0.4494
0.178	6.0	1470	0.1625	0.4881	0.3782	0.4328	0.4495
0.1466	7.0	1715	0.1636	0.4910	0.3837	0.4383	0.4538
0.1466	8.0	1960	0.1633	0.4890	0.3802	0.4343	0.4506
0.1307	9.0	2205	0.1642	0.4904	0.3836	0.4364	0.4531
0.1307	10.0	2450	0.1644	0.4902	0.3828	0.4363	0.4521

Framework versions

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