Here’s a great walkthrough of how to create a docker image that exclusively uses a custom model to run predictions and returns back the result.

This provides context isolation for the application running the model and avoids any non-reproducibility issue.

When dealing with python versioning you can easily get lost with all the possible combinations of your favorite library version, and the versions of the interpreter. Many different projects, specially the ones involving Tensorflow and Keras, require a very specific library version to run. Many other projects are not […]