finetuned_bartbase_on_medi_data

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: 0.1259
• Rouge1: 0.8208
• Rouge2: 0.6644
• Rougel: 0.7467
• Rougelsum: 0.7542
• Gen Len: 75.0167

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: 2.660730299084495e-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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	175	0.1297	0.812	0.6479	0.7393	0.7477	76.0533
No log	2.0	350	0.1037	0.817	0.6546	0.7393	0.7497	77.2933
0.0519	3.0	525	0.1095	0.8196	0.6656	0.7504	0.7588	75.38
0.0519	4.0	700	0.1157	0.8141	0.6539	0.7397	0.7494	76.3633
0.0519	5.0	875	0.1259	0.8208	0.6644	0.7467	0.7542	75.0167

Framework versions

• Transformers 4.30.0
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.13.3