bart-fine-tuned-on-summarization

This model is a fine-tuned version of ccdv/lsg-bart-base-16384-mediasum on the pubmed-summarization dataset. It achieves the following results on the evaluation set:

• Loss: 2.7293

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: 100
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
3.4477	0.2	100	3.1109
3.0893	0.4	200	2.8719
2.8441	0.6	300	2.7832
2.9203	0.8	400	2.7402
2.9796	1.0	500	2.7293

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2