This dataset consists of text articles and their corresponding summaries, designed for text summarization tasks.
The training example was manually created, while the test examples were extracted from the existing CNN/Daily Mail dataset.
No extensive preprocessing was done on the data. Articles were taken directly from the CNN/Daily Mail dataset.
1from datasets import load_from_disk
2
3train_dataset = load_from_disk('./results/custom_dataset/train')
4test_dataset = load_from_disk('./results/custom_dataset/test')
