kobart_32_5e-5_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: 2.6396
• Rouge1: 35.8418
• Rouge2: 12.983
• Rougel: 23.6913
• Bleu1: 29.8408
• Bleu2: 17.5438
• Bleu3: 10.2815
• Bleu4: 5.6838
• Gen Len: 50.2214

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: 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
1.6643	3.78	5000	2.6396	35.8418	12.983	23.6913	29.8408	17.5438	10.2815	5.6838	50.2214

Framework versions

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