bart-cnn-samsum-finetuned

This model is a fine-tuned version of facebook/bart-large-cnn on the 'samsum' data set. It achieves the following results on the evaluation set:

• Loss: 0.1330

Model description

Experiments with simple training on an existing model; for my personal blog.

Intended uses & limitations

To read conversations and give them summaries, to some degree.

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.0981	1.0	37	0.1360
0.1009	2.0	74	0.1330
0.0957	3.0	111	0.1330

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2