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: 1.3794
• Rouge1: 46.8948
• Rouge2: 23.4445
• Rougel: 39.5763
• Rougelsum: 43.209
• 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: 64
• eval_batch_size: 64
• 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
No log	1.0	231	1.3935	46.6142	23.0937	39.1018	42.8696	17.2076
No log	2.0	462	1.3848	46.5553	23.0122	39.1493	42.764	17.1465
1.4249	3.0	693	1.3813	46.8705	23.5239	39.6689	43.2545	17.2930
1.4249	4.0	924	1.3801	46.9726	23.6143	39.6028	43.3278	17.2112
1.3528	5.0	1155	1.3794	46.8948	23.4445	39.5763	43.209	17.2540

Framework versions

• Transformers 4.27.3
• Pytorch 1.10.0
• Datasets 2.10.1
• Tokenizers 0.13.2