thaisum_v2

This model is a fine-tuned version of google-t5/t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6483
• Rouge1: 0.0419
• Rouge2: 0.0194
• Rougel: 0.0415
• Rougelsum: 0.042
• Gen Len: 18.927

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.731	1.0	1250	0.6690	0.0278	0.0149	0.0274	0.0277	18.972
0.7022	2.0	2500	0.6529	0.0314	0.0177	0.0313	0.0313	18.924
0.6933	3.0	3750	0.6509	0.0396	0.0169	0.0393	0.0393	18.95
0.6821	4.0	5000	0.6483	0.0419	0.0194	0.0415	0.042	18.927

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2