jeongyoun/t5-FairytaleQA-AE

t5-base-finetuned-FairytaleQA-AnswerExtraction

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

• Loss: 0.0414
• Rouge1: 18.2182
• Rouge2: 7.6359
• Rougel: 17.6063
• Rougelsum: 17.5857
• F1: 1.9133
• Exact Match: 4.0
• Gen Len: 10.3756

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	F1	Exact Match	Gen Len
0.045	1.0	2137	0.0422	17.4796	6.949	17.0178	17.049	1.7117	3.3171	9.8829
0.0409	2.0	4274	0.0417	17.2926	6.6376	16.9168	16.9168	2.0107	4.3902	9.9171
0.0395	3.0	6411	0.0414	17.4948	6.7766	17.0438	17.0267	2.1929	4.4878	9.8585
0.0382	4.0	8548	0.0414	18.6484	7.8989	18.039	18.0181	1.8017	3.6098	10.7366
0.0374	5.0	10685	0.0414	18.2182	7.6359	17.6063	17.5857	1.9133	4.0	10.3756

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1