kobart-base-v2-2932-korean

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: 0.2633
• Rouge1: 0.36
• Rouge2: 0.1465
• Rougel: 0.3537
• Rougelsum: 0.3538
• Gen Len: 19.9992

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.2804	1.44	500	0.2805	0.3394	0.1272	0.3346	0.3344	20.0
0.2127	2.88	1000	0.2490	0.3512	0.1374	0.3461	0.3459	19.9992
0.1409	4.33	1500	0.2504	0.3594	0.1429	0.3527	0.3524	20.0
0.1208	5.77	2000	0.2506	0.3607	0.1436	0.3536	0.3538	20.0
0.0866	7.21	2500	0.2603	0.3657	0.1466	0.3589	0.3593	20.0
0.0782	8.65	3000	0.2633	0.36	0.1465	0.3537	0.3538	19.9992

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0