t5-finetuned-amazon-english

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

• Loss: 3.1713
• Rouge1: 19.1814
• Rouge2: 9.8673
• Rougel: 18.1982
• Rougelsum: 18.2963

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: 5.6e-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: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
3.3583	1.0	771	3.2513	16.6865	9.0598	15.8299	15.8472
3.1022	2.0	1542	3.2147	16.8499	9.4849	16.1568	16.2437
3.0067	3.0	2313	3.1718	16.9516	8.762	16.104	16.2186
2.9482	4.0	3084	3.1854	18.9582	9.5416	18.0846	18.2938
2.8934	5.0	3855	3.1669	18.857	9.934	17.9027	18.0272
2.8389	6.0	4626	3.1782	18.6736	9.326	17.6943	17.8852
2.8174	7.0	5397	3.1709	18.4342	9.6936	17.5714	17.6516
2.8	8.0	6168	3.1713	19.1814	9.8673	18.1982	18.2963

Framework versions

• Transformers 4.22.0
• Pytorch 1.12.1+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1