T5_small_title

This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.4558
• Rouge1: 0.316
• Rouge2: 0.1498
• Rougel: 0.2735
• Rougelsum: 0.2728
• Gen Len: 16.495

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	100	2.8637	0.2464	0.093	0.207	0.2066	18.87
No log	2.0	200	2.6086	0.2702	0.1142	0.2303	0.2299	18.475
No log	3.0	300	2.5391	0.2943	0.1373	0.2572	0.2565	17.44
No log	4.0	400	2.5082	0.2997	0.1421	0.2636	0.2629	17.02
2.8756	5.0	500	2.4853	0.3111	0.145	0.271	0.2701	16.755
2.8756	6.0	600	2.4729	0.3165	0.1501	0.2753	0.2745	16.555
2.8756	7.0	700	2.4635	0.3215	0.1533	0.2771	0.2768	16.51
2.8756	8.0	800	2.4601	0.3224	0.154	0.2773	0.2776	16.38
2.8756	9.0	900	2.4569	0.3167	0.1505	0.274	0.2733	16.495
2.5758	10.0	1000	2.4558	0.316	0.1498	0.2735	0.2728	16.495

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2