pretrain_Law_model_bartpho_version3

This model is a fine-tuned version of KingLTD/pretrain_Law_model_bartpho_version2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1145
• Rouge1: 0.4814
• Rouge2: 0.3997
• Rougel: 0.4430
• Rougelsum: 0.4527

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	0.0961	0.4759	0.3999	0.4423	0.4503
No log	2.0	490	0.1000	0.4820	0.4017	0.4425	0.4522
0.0968	3.0	735	0.1065	0.4817	0.4014	0.4422	0.4527
0.0968	4.0	980	0.1072	0.4807	0.4016	0.4445	0.4542
0.0618	5.0	1225	0.1098	0.4808	0.3997	0.4435	0.4532
0.0618	6.0	1470	0.1110	0.4817	0.3990	0.4428	0.4526
0.0427	7.0	1715	0.1125	0.4796	0.3988	0.4423	0.4529
0.0427	8.0	1960	0.1136	0.4822	0.4032	0.4453	0.4539
0.0314	9.0	2205	0.1141	0.4814	0.3991	0.4430	0.4527
0.0314	10.0	2450	0.1145	0.4814	0.3997	0.4430	0.4527

Framework versions

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