Contents

autogluon 0.6.1

0

AutoML for Image, Text, and Tabular Data

AutoML for Image, Text, and Tabular Data

Stars: 5187, Watchers: 5187, Forks: 678, Open Issues: 224

The autogluon/autogluon repo was created 3 years ago and was last updated Yesterday.
The project is extremely popular with a mindblowing 5187 github stars!

How to Install autogluon

You can install autogluon using pip

pip install autogluon

or add it to a project with poetry

poetry add autogluon

Package Details

Author
AutoGluon Community
License
Apache-2.0
Homepage
https://github.com/autogluon/autogluon
PyPi
https://pypi.org/project/autogluon/
GitHub Repo
https://github.com/awslabs/autogluon

Classifiers

  • Scientific/Engineering/Artificial Intelligence
  • Scientific/Engineering/Image Recognition
  • Scientific/Engineering/Information Analysis
  • Software Development
No  autogluon  pypi packages just yet.

Errors

A list of common autogluon errors.

Code Examples

Here are some autogluon code examples and snippets.

GitHub Issues

The autogluon package has 224 open issues on GitHub

  • Various HPO Cleanup / Fixes
  • NN_TORCH incorrectly implements use_batchnorm=True
  • Handle nan in categorical feature in text feature generator
  • NeuralNetFastAI runs extremely slow
  • How do I set confidence when I use test data to predict? Are there relevant parameters?
  • [WIP]ARM/M1 dependency upgrade and doc
  • Text Prediction only be handled in English? Do you support Chinese?
  • Add scikit-learn compatible API
  • TabularPredictor config helper
  • Consider integrating cleanlab - cleans noisy labels (enhancement)
  • [v0.4] Add documentation for HPO / Search Spaces
  • [v0.4] Refactor model valid features to be allow instead of exclude
  • Feature Tools integration
  • Included Model Types
  • No GPU is detected in the machine and we will not proceed to run TextPredictor because they will train too slowly with only CPU.

See more issues on GitHub

Related Packages & Articles

gluonts 0.11.6

GluonTS is a Python toolkit for probabilistic time series modeling, built around MXNet.

h2o 3.38.0.3

H2O, Fast Scalable Machine Learning, for python

datasets 2.8.0

HuggingFace community-driven open-source library of datasets

thinc 8.1.6

A refreshing functional take on deep learning, compatible with your favorite libraries

nlp 0.4.0

HuggingFace/NLP is an open library of NLP datasets.