bart-large-cnn-finetuned-scope1-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: 0.0612
• Rouge1: 55.9874
• Rouge2: 41.0458
• Rougel: 47.6072
• Rougelsum: 47.5635

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: 16
• eval_batch_size: 16
• 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
No log	1.0	17	0.1238	46.7806	30.4394	36.8259	36.8757
0.4762	2.0	34	0.1058	49.4907	32.4075	39.352	39.161
0.4762	3.0	51	0.0899	54.1557	35.6198	41.6488	41.4013
0.1104	4.0	68	0.0867	53.237	36.766	42.8508	42.7151
0.1104	5.0	85	0.0773	57.4084	39.3354	45.068	44.9505
0.0914	6.0	102	0.0736	56.9111	41.3118	48.1607	47.9965
0.0914	7.0	119	0.0699	58.6135	42.3985	48.7923	48.4873
0.0785	8.0	136	0.0673	59.5593	43.9205	51.7275	51.5617
0.0785	9.0	153	0.0618	62.0583	47.3928	53.3198	53.1472
0.0702	10.0	170	0.0612	55.9874	41.0458	47.6072	47.5635

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1