final_bart

This model is a fine-tuned version of gogamza/kobart-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.6848
• Rouge1: 35.7722
• Rouge2: 12.5127
• Rougel: 23.3002
• Rdass: 0.6248
• Bleu1: 30.5261
• Bleu2: 17.6264
• Bleu3: 10.3974
• Bleu4: 5.4348
• Gen Len: 53.47

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: 3e-05
• train_batch_size: 64
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5.0

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rdass	Bleu1	Bleu2	Bleu3	Bleu4	Gen Len
2.1542	1.5	1000	2.7491	33.5554	11.2371	22.006	0.6093	27.9938	15.5354	8.2494	4.42	50.08
2.0071	2.99	2000	2.6813	35.0501	12.2759	22.6669	0.6155	29.6866	17.1396	9.7016	5.3559	54.04
1.8694	4.49	3000	2.6848	35.7722	12.5127	23.3002	0.6248	30.5261	17.6264	10.3974	5.4348	53.47

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.1+cu117
• Datasets 2.7.1
• Tokenizers 0.13.2