flan-t5-base-jfleg

This model is a fine-tuned version of google/flan-t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7504
• Rouge1: 82.0456
• Rouge2: 71.8069
• Rougel: 81.0231
• Rougelsum: 81.0225
• Gen Len: 17.3842

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	377	0.7504	82.0456	71.8069	81.0231	81.0225	17.3842
0.547	2.0	754	0.8532	82.2122	71.9383	81.1551	81.1524	17.3909
0.4414	3.0	1131	0.9503	82.2202	71.9046	81.1744	81.1738	17.3748
0.384	4.0	1508	0.9621	82.042	71.7195	80.9726	80.9785	17.3922
0.384	5.0	1885	0.9727	82.1163	71.8526	81.0433	81.0423	17.3949

Framework versions

• Transformers 4.27.3
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2