t5_small_ten_epoch

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

• Loss: 2.2845
• Rouge1: 0.3621
• Rouge2: 0.2114
• Rougel: 0.2982
• Rougelsum: 0.297
• Gen Len: 17.52

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: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	275	2.8863	0.1507	0.0472	0.1346	0.1338	17.67
3.0473	2.0	550	2.6258	0.2558	0.1233	0.228	0.2293	17.49
3.0473	3.0	825	2.5073	0.318	0.1732	0.2682	0.2683	17.33
2.5236	4.0	1100	2.4166	0.3364	0.1893	0.2823	0.2822	17.25
2.5236	5.0	1375	2.3721	0.3506	0.2051	0.2927	0.2917	17.17
2.3768	6.0	1650	2.3339	0.3566	0.2101	0.2952	0.2944	17.3
2.3768	7.0	1925	2.3150	0.3546	0.2089	0.2903	0.2899	17.46
2.2982	8.0	2200	2.2954	0.3604	0.2144	0.2947	0.2941	17.53
2.2982	9.0	2475	2.2885	0.3627	0.213	0.2955	0.2945	17.52
2.2608	10.0	2750	2.2845	0.3621	0.2114	0.2982	0.297	17.52

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1