flan-t5-base-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: 1.3718
• Rouge1: 47.4238
• Rouge2: 23.5718
• Rougel: 39.9102
• Rougelsum: 43.5465
• Gen Len: 17.2540

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
1.4553	1.0	1842	1.3885	46.3203	22.8542	38.7501	42.6178	17.5177
1.3375	2.0	3684	1.3746	47.1463	23.6451	39.7442	43.5545	17.2418
1.2785	3.0	5526	1.3718	47.4238	23.5718	39.9102	43.5465	17.2540
1.228	4.0	7368	1.3762	47.5418	24.051	40.0543	43.7907	17.3468
1.2045	5.0	9210	1.3767	47.3198	23.7145	39.7772	43.5547	17.2894

Framework versions

• Transformers 4.43.3
• Pytorch 2.0.1
• Datasets 2.20.0
• Tokenizers 0.19.1