Clumsy package and build systems have been C++ biggest problems.
You could not ask a data scientist to use C++.
All the top ML libs written in C++ had to be wrapped in Python.
I once had to deploy many expensive GPU VMs to server up ML models on slow Python web servers.
Running up cloud bill 💸
You could not ask a data scientist to use C++.
All the top ML libs written in C++ had to be wrapped in Python.
I once had to deploy many expensive GPU VMs to server up ML models on slow Python web servers.
Running up cloud bill 💸
Comments
Python is great for ML development, but it's nice to have the option to serve the ML models from a C++ web server.
I tested Conan on C/C++ game libraries: raylib, SDL, SFML and easily got them running 😍