flan-t5-small-samsum

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

• Loss: 1.6325
• Rouge1: 43.7829
• Rouge2: 19.4994
• Rougel: 36.5484
• Rougelsum: 39.9323
• Gen Len: 16.8730

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
1.8137	1.0	1842	1.6636	42.6155	18.9113	35.7997	39.046	16.7473
1.7502	2.0	3684	1.6408	43.3833	19.1709	36.0178	39.5973	16.8620
1.6864	3.0	5526	1.6372	43.31	19.2269	35.9239	39.6082	16.8559
1.6647	4.0	7368	1.6334	43.7043	19.462	36.4417	39.8969	16.9512
1.6391	5.0	9210	1.6325	43.7829	19.4994	36.5484	39.9323	16.8730

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0
• Datasets 2.11.0
• Tokenizers 0.13.2