fine-tuned-biobart-14-epochs

This model is a fine-tuned version of GanjinZero/biobart-v2-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7165
• Rouge1: 0.297
• Rouge2: 0.1097
• Rougel: 0.2649
• Rougelsum: 0.2663
• Gen Len: 14.74

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: 14

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	151	0.7412	0.2238	0.0808	0.1984	0.2007	13.75
No log	2.0	302	0.6979	0.2428	0.0841	0.2132	0.2156	13.26
No log	3.0	453	0.6825	0.2488	0.0932	0.2216	0.2216	14.24
0.7256	4.0	604	0.6814	0.273	0.104	0.243	0.2442	13.94
0.7256	5.0	755	0.6845	0.2801	0.0999	0.2533	0.2542	14.2
0.7256	6.0	906	0.6837	0.2793	0.0997	0.253	0.2533	14.17
0.5118	7.0	1057	0.6941	0.2791	0.0988	0.2513	0.2517	14.62
0.5118	8.0	1208	0.6979	0.2968	0.1155	0.2677	0.2697	14.12
0.5118	9.0	1359	0.6983	0.2886	0.106	0.2633	0.2652	14.49
0.4062	10.0	1510	0.7048	0.2932	0.1117	0.2686	0.2714	14.36
0.4062	11.0	1661	0.7085	0.2773	0.1037	0.2512	0.2526	14.66
0.4062	12.0	1812	0.7140	0.2927	0.112	0.2657	0.2673	14.59
0.4062	13.0	1963	0.7162	0.2924	0.109	0.2597	0.262	14.57
0.3585	14.0	2114	0.7165	0.297	0.1097	0.2649	0.2663	14.74

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0