3 ms·
Not to mention dependency management! Since a lot of ML code is in Python this ends up being a very tricky thing to handle at scale (especially if you need to u
by alfalfasprout 4y ago
Not to mention dependency management! Since a lot of ML code is in Python this ends up being a very tricky thing to handle at scale (especially if you need to update dependencies, etc.)
- pharmakom 4y agoI didn’t care much for Docker until I started working in the Python ecosystem.