SummarEaseV1

This model is a fine-tuned version of google-t5/t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.4382
• Rouge1: 0.2458
• Rouge2: 0.1168
• Rougel: 0.2008
• Rougelsum: 0.2001
• Gen Len: 19.0

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: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	0.7619	3	2.5665	0.2388	0.1118	0.1962	0.1959	19.0
No log	1.7778	7	2.4863	0.2439	0.1153	0.2005	0.1996	19.0
No log	2.7937	11	2.4462	0.2461	0.1169	0.2009	0.2003	19.0
No log	3.0476	12	2.4382	0.2458	0.1168	0.2008	0.2001	19.0

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1