Dependencies

---

Python & OS

As of the moment, ATOM supports the following Python versions:

• Python 3.7
• Python 3.8
• Python 3.9
• Python 3.10

And operating systems:

• macOS
• Unix-like (Ubuntu, Fedora, etc...)
• Windows 8.1+

Packages

Required

ATOM is built on top of several existing Python libraries. These packages are necessary for its correct functioning.

• numpy (>=1.21.5)
• scipy (>=1.7.3)
• pandas (>=1.3.5)
• pandas-profiling (>=3.2.0)
• explainerdashboard (>=0.3.8.2)
• mlflow (>=1.26)
• dill (>=0.3.5)
• tqdm (>=4.64)
• joblib (>=1.1.0)
• typeguard (>=2.13)
• scikit-learn (>=1.0.1)
• scikit-optimize (>=0.9.0)
• nltk (>=3.7)
• tpot (>=0.11.7)
• category-encoders (>=2.4.1)
• imbalanced-learn (>=0.9.0)
• featuretools (>=1.10.0)
• gplearn (>=0.4.1)
• zoofs (>=0.1.24)
• matplotlib (>=3.5.0)
• seaborn (>=0.11.0)
• shap (>=0.41)
• wordcloud (>=1.8.1)
• schemdraw (>=0.14)

Optional

You can install some optional packages to be able to use some well-known machine learning estimators that are not provided by sklearn but are among ATOM's predefined models. Install them using pip install atom-ml[models].

• xgboost (>=0.90)
• lightgbm (>=3.3.2)
• catboost (>=1.0.4)

Development

The development dependencies are not installed with the package, and are not required for any of its functionalities. These libraries are only necessary to contribute to the project. Install them using pip install atom-ml[dev].

• isort (>=5.10.1)
• flake8 (>=4.0.1)
• pytest (>=7.1.0)
• pytest-cov (>=3.0.0)
• tensorflow (>=2.3.1)
• mkdocs (>=1.2.3)
• mkdocs-material (>=8.2.5)
• mkdocs-jupyter (>=0.20.1)