bart_base

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

• Loss: 2.5401
• Rouge1: 0.2403
• Rouge2: 0.0714
• Rougel: 0.1924
• Rougelsum: 0.1922
• Gen Len: 18.1163

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: 1
• eval_batch_size: 1
• seed: 42
• 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
No log	1.0	99	2.5104	0.189	0.053	0.154	0.1535	13.3023
No log	2.0	198	2.4941	0.2498	0.0781	0.1972	0.197	18.2326
No log	3.0	297	2.5144	0.2394	0.0652	0.1914	0.192	18.4419
No log	4.0	396	2.5401	0.2403	0.0714	0.1924	0.1922	18.1163

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3