This topic lists some limitations associated with customized ML Runtime images.

  • The contents of certain pre-existing standard directories such as /home/cdsw, /tmp, and so on, cannot be modified while creating customized non-PBJ ML Runtimes. This means any files saved in these directories will not be accessible from sessions that are running on customized ML Runtimes.

    Workaround: Create a new custom directory in the Dockerfile used to create the customized ML Runtime, and save your files to that directory.

For PBJ Runtimes, note the following limitations:

  • PBJ Runtimes work as models only with R and Python kernels.