Allow eager execution and distributed training over Multi-GPU systems

tyagi-iiitv / PointPillars

GNU General Public License v3.0

105 stars 47 forks source link

Allow eager execution and distributed training over Multi-GPU systems #25

Open ma7555 opened 3 years ago

ma7555 commented 3 years ago

Fixes hardcoded batch_size during network creation.

Allow eager execution

Fixed C++ module compilation over Python v3.8+ for win10 64bit (updating pybind)

tyagi-iiitv commented 3 years ago

Hi @ma7555, thanks for your contributions. I am not actively working on this project now, but to support continued collaboration and efforts to make this work, it'll be great if you could be a developer in this repo. That'll support continued development in this project.

ma7555 commented 3 years ago

I can help when i can