t5-small-finetuned-cnn-news

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

• Loss: 2.7209
• Rouge1: 23.5402
• Rouge2: 10.8834
• Rougel: 19.3936
• Rougelsum: 22.1513

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: 0.00056
• 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
2.2531	1.0	718	2.6722	23.3437	10.5433	19.2183	21.8989
2.1518	2.0	1436	2.7024	23.4068	10.716	19.0751	21.9328
2.0925	3.0	2154	2.7235	23.232	10.5236	19.2254	21.8598
2.0808	4.0	2872	2.7309	23.7401	10.7664	19.4651	22.2479
2.1114	5.0	3590	2.7209	23.5402	10.8834	19.3936	22.1513

Framework versions

• Transformers 4.27.2
• Pytorch 1.13.1+cu117
• Datasets 2.11.0
• Tokenizers 0.13.3