deepchem

Introduction to DeepChem

DeepChem is an open-source project designed to make deep learning accessible to fields such as drug discovery, materials science, quantum chemistry, and biology. This platform provides a comprehensive toolchain that enables researchers and enthusiasts to develop and apply machine learning models to a wide range of scientific problems.

Project Overview

DeepChem supports deep learning frameworks such as TensorFlow, PyTorch, and JAX, providing flexibility for various computational needs. Its main aim is to establish an open platform where the latest advancements in machine learning can be applied in life sciences effectively.

Requirements

DeepChem is compatible with Python versions 3.7 through 3.10 and depends on several core packages, including:

• Joblib
• NumPy
• Pandas
• Scikit-learn
• SciPy
• RDKit

Additionally, there are numerous optional packages that might be required depending on the specific functionalities or models a user desires to utilize. Details on these can be found in DeepChem's documentation.

Installation

DeepChem can be installed in various ways to suit different user preferences and operational environments:

• Stable Version: Installable via pip or conda.

  • pip install deepchem
  • conda install -c conda-forge deepchem

• Nightly Build: For those interested in cutting-edge features, the nightly build can be installed using pip install --pre deepchem.

• Docker: DeepChem also offers Docker images for those who prefer containerized environments. This includes both tagged versions and the latest builds.

• From Source: Users who wish to have all aspects of DeepChem or contribute to its development can install directly from the source.

Getting Started

DeepChem provides a wealth of tutorial materials aimed at guiding users from basic to more advanced skills in molecular machine learning and computational biology. These tutorials are accessible and can be run on platforms like Google Colab. Beyond tutorials, there are many examples that users can adapt for their specific research needs.

Community and Support

The DeepChem community flourishes with contributions from a diverse group of scientists and developers. For discussions, support, or feature requests, the community is active on Discord and has a discussion forum available. This collaborative environment provides a perfect avenue for learning, questions, and community-driven development.

Contributing

DeepChem welcomes contributions from anyone interested. The project is managed by volunteers who are united by a passion for open-source solutions in life sciences.

Citing DeepChem

Researchers who incorporate DeepChem into their work are encouraged to cite the book "Deep Learning for the Life Sciences" by the DeepChem core team. This book acts as a comprehensive resource for understanding how deep learning can be applied across various life science disciplines.

By fostering an inclusive and resourceful community, DeepChem stands as a pivotal tool in leveraging deep learning for scientific advancement.