What options do you have for distributing Python and R code and still keeping some control over the intellectual property (IP). In this video I look at some of the high level options, such as API’s, Docker Images, Compiled code, and encryption. If there is interest I may follow this up with more technical low-level tutorials on how to protect code in Python.

0:22 What code should you not share?
0:55 Protecting machine learning code
1:16 What are the parts of a model deploy
1:40 Scoring code vs. training code
2:44 What about lookup tables and other data?
3:02 Deploy in the cloud or on the edge?
3:12 Deploying behind an API
4:28 Protecting access to the API
4:50 Authentication and Throttling
5:24 Advsarial Example Attack
6:20 Edge Deployment
7:50 Preventing copying and modification
8:00 Compiled languages
8:23 Does Docker offer protection?
9:50 Protecting Binary files
10:30 What about encryption?

Subscribe: https://www.youtube.com/c/HeatonResearch/featured

#python #r #machine-learning