t5-small-finetuned-samsum-en

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

• Loss: 1.9335
• Rouge1: 44.3313
• Rouge2: 20.71
• Rougel: 37.221
• Rougelsum: 40.9603

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: 10
• eval_batch_size: 10
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
1.4912	1.0	300	1.9043	44.1517	20.0186	36.6053	40.5164
1.5055	2.0	600	1.8912	44.1473	20.4456	37.069	40.6714
1.4852	3.0	900	1.8986	44.7536	20.8646	37.525	41.2189
1.4539	4.0	1200	1.9136	44.2144	20.3446	37.1088	40.7581
1.4262	5.0	1500	1.9215	44.2656	20.6044	37.3267	40.9469
1.4118	6.0	1800	1.9247	43.8793	20.4663	37.0614	40.6065
1.3987	7.0	2100	1.9256	43.9981	20.2703	36.7856	40.6354
1.3822	8.0	2400	1.9316	43.9732	20.4559	36.8039	40.5784
1.3773	9.0	2700	1.9314	44.3075	20.5435	37.0457	40.832
1.3795	10.0	3000	1.9335	44.3313	20.71	37.221	40.9603

Framework versions

• Transformers 4.19.2
• Pytorch 1.11.0+cu113
• Datasets 2.2.2
• Tokenizers 0.12.1