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: 0.1165
• Rouge1: 55.7822
• Rouge2: 41.9699
• Rougel: 47.2427
• Rougelsum: 47.1372

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
0.613	1.0	35	0.2146	42.8394	24.2808	31.8024	31.7805
0.1987	2.0	70	0.1989	46.2492	30.694	36.6481	36.5135
0.1939	3.0	105	0.1792	47.7166	31.8667	38.5667	38.526
0.1663	4.0	140	0.1642	49.6835	34.7278	39.5294	39.4623
0.1711	5.0	175	0.1555	51.6538	35.9915	40.1589	40.1665
0.1577	6.0	210	0.1443	50.4306	35.9713	40.4836	40.4492
0.1511	7.0	245	0.1367	55.8887	43.2295	49.0124	48.9803
0.1425	8.0	280	0.1306	56.2433	41.4182	45.9078	45.9027
0.1255	9.0	315	0.1191	57.3464	43.7543	47.335	47.3058
0.1299	10.0	350	0.1165	55.7822	41.9699	47.2427	47.1372

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1