fine-tuned-BART-20-epochs-wanglab-512-output

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

• Loss: 0.4945
• Rouge1: 0.0871
• Rouge2: 0.0196
• Rougel: 0.0787
• Rougelsum: 0.0787
• Bertscore F1: 0.837
• Bleurt Score: -1.873
• Gen Len: 20.0

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bertscore F1	Bleurt Score	Gen Len
No log	1.0	301	1.5515	0.0293	0.0	0.0286	0.0282	0.7994	-2.1159	11.68
4.8736	2.0	602	0.5364	0.0738	0.0183	0.0655	0.0653	0.8345	-1.6735	20.0
4.8736	3.0	903	0.4811	0.071	0.0191	0.0677	0.0677	0.8359	-1.7563	20.0
0.5377	4.0	1204	0.4621	0.0506	0.0125	0.0475	0.0474	0.8566	-1.8275	8.0
0.4145	5.0	1505	0.4496	0.0231	0.0036	0.0237	0.0233	0.8458	-1.4636	8.0
0.4145	6.0	1806	0.4455	0.078	0.0194	0.0714	0.071	0.8469	-1.3815	20.0
0.336	7.0	2107	0.4416	0.0871	0.0196	0.0787	0.0787	0.837	-1.873	20.0
0.336	8.0	2408	0.4440	0.0878	0.0195	0.0794	0.0791	0.8409	-1.4561	20.0
0.2698	9.0	2709	0.4505	0.0231	0.0036	0.0237	0.0233	0.8458	-1.4636	8.0
0.2225	10.0	3010	0.4546	0.0516	0.0101	0.0466	0.0463	0.8355	-1.61	20.0
0.2225	11.0	3311	0.4627	0.0877	0.0194	0.0794	0.0791	0.8388	-1.4342	20.0
0.1695	12.0	3612	0.4677	0.0704	0.0128	0.0628	0.0626	0.8218	-1.8469	20.0
0.1695	13.0	3913	0.4716	0.0615	0.0193	0.056	0.0557	0.8342	-1.5375	20.0
0.132	14.0	4214	0.4754	0.064	0.0196	0.0577	0.0576	0.839	-1.8751	20.0
0.1122	15.0	4515	0.4837	0.0712	0.0175	0.0644	0.0642	0.8373	-1.3366	20.0
0.1122	16.0	4816	0.4867	0.0817	0.01	0.0691	0.069	0.8425	-1.4584	20.0
0.0893	17.0	5117	0.4904	0.0712	0.0175	0.0644	0.0642	0.8373	-1.3366	20.0
0.0893	18.0	5418	0.4924	0.0871	0.0196	0.0787	0.0787	0.837	-1.873	20.0
0.08	19.0	5719	0.4934	0.0871	0.0196	0.0787	0.0787	0.837	-1.873	20.0
0.0706	20.0	6020	0.4945	0.0871	0.0196	0.0787	0.0787	0.837	-1.873	20.0

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2