nlp 0.4.0

HuggingFace/NLP is an open library of NLP datasets.

Stars: 21193, Watchers: 21193, Forks: 3099, Open Issues: 1061

The huggingface/datasets repo was created 5 years ago and the last code push was 21 hours ago.
The project is extremely popular with a mindblowing 21193 github stars!

How to Install nlp

You can install nlp using pip

pip install nlp

or add it to a project with poetry

poetry add nlp

Package Details

Author

HuggingFace Inc.

License

Apache 2.0

Homepage

https://github.com/huggingface/nlp

PyPi:

https://pypi.org/project/nlp/

GitHub Repo:

https://github.com/huggingface/nlp

Classifiers

• Scientific/Engineering/Artificial Intelligence

Related Packages

Errors

A list of common nlp errors.

Code Examples

Here are some nlp code examples and snippets.

GitHub Issues

The nlp package has 1061 open issues on GitHub

• Use Sequence instead of list in Dataset.from_parquet type hints
• [CUDA Tensors Not working in ~v4.5.0] set_format(type="torch", device="cuda") returns cpu
• Is the 10k files / folder limit a hard limit for a dataset repo?
• all_exhausted_without_replacement working same as first_exhausted
• #5354: replace list with Sequence in from_parquet type hints
• feat: Add GenBank file format support for biological sequence data
• docs: clarify documentation build instructions
• json: add optional return_file_name parameter
• MMLU get_dataset_config_names provides different lists of subsets in online and offline modes
• Question: Is there a faster way to push_to_hub for large image datasets?
• Remove Python 3.7 and Python 2 code paths from _dill.py
• Improve readability and documentation of indexing integration tests
• datasets.load_from_disk progress bar optional manual control
• Fix duplicate log messages by disabling log propagation by default
• Bug fix: Add HDFS hostname to protocol prefix

See more issues on GitHub