flan-t5-base-samsum5

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: 1.3676
• Rouge1: 46.8382
• Rouge2: 23.107
• Rougel: 39.5293
• Rougelsum: 42.8917
• Gen Len: 17.3675

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: 6
• eval_batch_size: 6
• 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
1.4315	1.0	2456	1.3793	46.7908	22.8618	39.3679	43.0337	17.1148
1.352	2.0	4912	1.3676	46.8382	23.107	39.5293	42.8917	17.3675
1.2638	3.0	7368	1.3684	47.3491	23.5078	39.9401	43.4065	17.1832
1.2238	4.0	9824	1.3727	47.2949	23.8671	40.057	43.645	17.3260
1.1632	5.0	12280	1.3737	47.3124	23.7118	40.0067	43.6227	17.3126

Framework versions

• Transformers 4.27.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2