my_longer_summarization_model

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

• Loss: 2.2700
• Rouge1: 0.4373
• Rouge2: 0.1851
• Rougel: 0.2845
• Rougelsum: 0.284
• Gen Len: 249.996

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: 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	124	2.4805	0.4073	0.1571	0.2477	0.248	225.7621
No log	2.0	248	2.4026	0.4211	0.1683	0.2619	0.2617	228.9919
No log	3.0	372	2.3562	0.4247	0.1736	0.2731	0.273	243.871
No log	4.0	496	2.3316	0.432	0.1782	0.2774	0.277	248.7419
2.688	5.0	620	2.3041	0.4264	0.1744	0.2781	0.2778	250.8065
2.688	6.0	744	2.2914	0.4289	0.1781	0.2808	0.2805	248.375
2.688	7.0	868	2.2820	0.4305	0.1797	0.2831	0.2827	249.8871
2.688	8.0	992	2.2765	0.4337	0.1824	0.2827	0.2822	249.246
2.5114	9.0	1116	2.2719	0.4338	0.1819	0.2837	0.2832	249.379
2.5114	10.0	1240	2.2700	0.4373	0.1851	0.2845	0.284	249.996

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1