search

nlp-with-transformers / notebooks

Jupyter notebooks for the Natural Language Processing with Transformers book
https://transformersbook.com/
Apache License 2.0
3.92k stars 1.23k forks source link

emotion dataset #89

Open pnugues opened 1 year ago

pnugues commented 1 year ago

Information

The problem arises in chapter:

Describe the bug

Collecting the datasets with the statements: from datasets import list_datasets all_datasets = list_datasets() from the notebook fails to find an emotion dataset.

Running the statement: [item for item in all_datasets if 'emotion' in item][:5] here are the first items of the list I obtain: ['dair-ai/emotion', 'go_emotions', 'Mansooreh/sharif-emotional-speech-dataset', 'Pyjay/emotion_nl', 'SetFit/emotion'] but no emotion.

As a consequence, the subsequent statements: from datasets import load_dataset emotions = load_dataset("emotion") simply crash.

hoangminhtoan commented 1 year ago

You should change "emotion" -> "SetFit/emotion" as bellow.

emotions = load_dataset("SetFit/emotion")

pnugues commented 1 year ago

Thank you for the workaround. Nonetheless, this is a bug that should be corrected in the notebook.