kobart_32_1e-4_datav2_min30_lp5.0_temperature1.0

This model is a fine-tuned version of gogamza/kobart-base-v2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.7215
• Rouge1: 36.3912
• Rouge2: 13.2376
• Rougel: 23.7632
• Bleu1: 30.6123
• Bleu2: 18.0414
• Bleu3: 10.5291
• Bleu4: 6.0123
• Gen Len: 49.5035

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: 0.0001
• train_batch_size: 32
• 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	Bleu1	Bleu2	Bleu3	Bleu4	Gen Len
1.3994	3.78	5000	2.7215	36.3912	13.2376	23.7632	30.6123	18.0414	10.5291	6.0123	49.5035

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cu117
• Datasets 2.7.1
• Tokenizers 0.13.2