t5-small-finetuned-multi-news

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

• Loss: 2.8049
• Rouge1: 15.1241
• Rouge2: 4.9514
• Rougel: 11.5019
• Rougelsum: 13.3079

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: 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: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
3.2411	1.0	1250	2.8772	14.6774	4.7697	11.335	13.0082
3.079	2.0	2500	2.8438	14.9558	4.8748	11.4023	13.2198
3.0257	3.0	3750	2.8240	15.133	4.9814	11.572	13.3607
2.9903	4.0	5000	2.8153	15.1339	4.9123	11.5038	13.3464
2.9659	5.0	6250	2.8085	15.1134	5.0057	11.5478	13.3483
2.9461	6.0	7500	2.8066	15.154	4.9641	11.5276	13.3523
2.936	7.0	8750	2.8049	15.1241	4.9514	11.5019	13.3079

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3