SocialScienceBARTMainSections

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: 4.5267
• Rouge1: 51.5502
• Rouge2: 19.1289
• Rougel: 36.9981
• Rougelsum: 48.0056
• Bertscore Precision: 81.4667
• Bertscore Recall: 83.8704
• Bertscore F1: 82.647
• Bleu: 0.1571
• Gen Len: 194.5169

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: 5e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bertscore Precision	Bertscore Recall	Bertscore F1	Bleu	Gen Len
5.9423	0.1332	100	5.8537	44.7886	15.5028	32.2161	41.724	78.6884	82.051	80.3281	0.1282	194.5169
5.5219	0.2664	200	5.3814	45.5534	16.1718	32.9346	42.758	79.0971	82.3765	80.6972	0.1323	194.5169
5.1742	0.3997	300	5.0879	48.1215	17.0033	34.0793	44.5274	79.1451	82.8315	80.939	0.1387	194.5169
5.0337	0.5329	400	4.9042	49.2783	17.5741	34.9739	45.4337	80.0738	83.2616	81.6311	0.1447	194.5169
4.8596	0.6661	500	4.7692	50.3917	17.9196	35.6188	47.0232	80.8885	83.4241	82.1326	0.1475	194.5169
4.7917	0.7993	600	4.6321	51.7348	19.0125	36.6567	47.9429	81.3534	83.827	82.5677	0.1557	194.5169
4.5184	0.9326	700	4.5267	51.5502	19.1289	36.9981	48.0056	81.4667	83.8704	82.647	0.1571	194.5169

Framework versions

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