Combining Conda + Pip for Machine Learning and AI

Best practices for installing and managing packages

Conda, our cross-platform environment and package manager, can be used to install most machine learning, AI, and deep learning software in a consistent and reliable manner. But occasionally, software is not yet available as a conda package but is available on PyPI. In these cases, conda can be combined with pip.

Getting conda to cooperate with pip often can be a challenge, however. In this Anaconda webinar, Software Engineer Jonathan Helmus will discuss common pitfalls that occur when these tools are used together and recommend best practices to avoid these problems.

Meet Our Speaker

  • Jonathan Helmus
    Software Engineer, Anaconda

    Jonathan is a member of the Distribution team at Anaconda, where he helps maintain the large collection of conda packages including many popular AI, deep learning, and GPU accelerated libraries. He assists clients in creating customized packages and data science environments using conda. Jonathan also is a core member of the conda-forge organization, the creator and maintainer of berryconda, and a contributor to a number of open source projects in the PyData ecosystem.

    Prior to joining Anaconda, Jonathan worked as a scientist and software engineer at Argonne National Laboratory, where he researched algorithms for processing weather radar data.

    Jonathan received his PhD in Chemical Physics from The Ohio State University, and a BS in Chemistry from Michigan Technological University.

*Required. Your information will be processed according to Anaconda’s Privacy Policy.