bart-large-cnn-finetuned-scope-summarization

This model is a fine-tuned version of facebook/bart-large-cnn on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1309
• Rouge1: 52.6236
• Rouge2: 39.8632
• Rougel: 43.4607
• Rougelsum: 43.3561

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: 5.6e-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

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
0.5104	1.0	34	0.2207	41.732	26.7717	31.2807	31.3611
0.2181	2.0	68	0.2001	44.5268	30.0523	34.7912	35.0095
0.1824	3.0	102	0.1995	45.4038	32.6808	36.3856	36.4004
0.1851	4.0	136	0.1728	48.85	35.9202	39.2826	39.1813
0.1692	5.0	170	0.1663	47.1374	34.5505	37.8192	37.8176
0.164	6.0	204	0.1594	50.3895	37.8751	40.4181	40.3778
0.1534	7.0	238	0.1526	50.7178	38.8207	41.5719	41.6111
0.1421	8.0	272	0.1424	51.3382	38.6796	40.4545	40.3891
0.1423	9.0	306	0.1354	53.8161	41.0736	45.1571	45.0427
0.1336	10.0	340	0.1309	52.6236	39.8632	43.4607	43.3561

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1