bart_samsum

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

• Loss: 1.5520
• Rouge1: 53.0152
• Rouge2: 28.029
• Rougel: 43.8864
• Rougelsum: 48.7945
• Gen Len: 30.3138

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.4812	1.0	921	1.5084	52.6238	27.9144	43.3378	48.77	30.4225
1.1068	2.0	1842	1.4507	53.2142	28.5004	44.382	49.228	28.6264
0.9224	3.0	2763	1.5031	52.8334	27.9492	43.7939	48.714	29.5751
0.7957	4.0	3684	1.5520	53.0152	28.029	43.8864	48.7945	30.3138

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1