fine-tuned-BioBART-20-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: 0.7640
• Rouge1: 0.3185
• Rouge2: 0.1264
• Rougel: 0.2886
• Rougelsum: 0.2878
• Gen Len: 15.42

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	151	0.7532	0.2	0.0753	0.1834	0.1844	13.29
No log	2.0	302	0.7148	0.2593	0.0845	0.2302	0.2324	13.92
No log	3.0	453	0.6995	0.2446	0.0856	0.2206	0.2227	14.49
0.724	4.0	604	0.6956	0.2926	0.1063	0.2674	0.2679	14.31
0.724	5.0	755	0.7029	0.3041	0.1208	0.2828	0.2829	14.81
0.724	6.0	906	0.6965	0.2821	0.1126	0.2611	0.2596	15.0
0.5016	7.0	1057	0.7097	0.286	0.1205	0.256	0.2555	15.0
0.5016	8.0	1208	0.7140	0.2899	0.1147	0.2643	0.2636	14.3
0.5016	9.0	1359	0.7191	0.3173	0.1231	0.2897	0.2917	14.75
0.3838	10.0	1510	0.7274	0.3115	0.1273	0.2874	0.2868	14.82
0.3838	11.0	1661	0.7312	0.3132	0.1294	0.2831	0.2825	14.97
0.3838	12.0	1812	0.7419	0.2975	0.1133	0.273	0.2731	14.98
0.3838	13.0	1963	0.7441	0.2934	0.1134	0.2683	0.2693	15.1
0.3153	14.0	2114	0.7490	0.2979	0.1212	0.2751	0.275	15.29
0.3153	15.0	2265	0.7536	0.2981	0.1125	0.2691	0.2687	14.83
0.3153	16.0	2416	0.7564	0.2989	0.1094	0.2719	0.2719	15.3
0.2646	17.0	2567	0.7585	0.298	0.1128	0.2745	0.2748	15.21
0.2646	18.0	2718	0.7630	0.2941	0.1092	0.2653	0.2651	15.12
0.2646	19.0	2869	0.7632	0.298	0.11	0.2681	0.2681	15.25
0.2428	20.0	3020	0.7640	0.3185	0.1264	0.2886	0.2878	15.42

Framework versions

• Transformers 4.36.2
• Pytorch 1.12.1+cu113
• Datasets 2.15.0
• Tokenizers 0.15.0