flan-t5-base-samsum

This model is a fine-tuned version of google/flan-t5-base on https://huggingface.co/datasets/samsum dataset.

It achieves the following results on the evaluation set:

• Loss: 1.3743
• Rouge1: 47.5395
• Rouge2: 24.0064
• Rougel: 40.1703
• Rougelsum: 43.8303
• Gen Len: 17.2564

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
1.4554	1.0	1842	1.3865	46.9319	23.5287	39.3501	43.0805	17.3968
1.3404	2.0	3684	1.3760	47.3057	23.7013	39.767	43.4863	16.9634
1.272	3.0	5526	1.3743	47.5395	24.0064	40.1703	43.8303	17.2564
1.2277	4.0	7368	1.3747	47.6417	23.88	40.0928	43.8293	17.2589
1.2069	5.0	9210	1.3764	47.7095	23.8971	40.0913	43.9315	17.3675

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0