caramel-t0-samsum

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

• Loss: 1.3721
• Rouge1: 47.6161
• Rouge2: 23.8204
• Rougel: 40.1348
• Rougelsum: 43.758
• Gen Len: 17.2759

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: 5e-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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.4403	1.0	1842	1.3822	47.2645	23.8282	39.7516	43.5543	17.0256
1.3572	2.0	3684	1.3747	47.5267	23.6332	39.8462	43.6527	17.4347
1.2822	3.0	5526	1.3721	47.6161	23.8204	40.1348	43.758	17.2759
1.2375	4.0	7368	1.3764	47.7445	24.1348	40.1867	43.9317	17.2943
1.1935	5.0	9210	1.3781	47.5911	23.7657	39.8614	43.7314	17.3077

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.1+cu116
• Datasets 2.11.0
• Tokenizers 0.13.2