BBC_News_summary_model_trained_on_reduced_data

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

• Loss: 0.1620
• Rouge1: 0.7253
• Rouge2: 0.6345
• Rougel: 0.5335
• Rougelsum: 0.5337
• Generated Length: 124.1525

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: 2e-05
• 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
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Generated Length
0.2107	1.0	501	0.1603	0.6533	0.5521	0.4925	0.4923	110.139
0.1133	2.0	1002	0.1495	0.6886	0.5875	0.5131	0.5138	125.2915
0.0719	3.0	1503	0.1620	0.7253	0.6345	0.5335	0.5337	124.1525

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1