search

devendrachaplot / DeepRL-Grounding

Train an RL agent to execute natural language instructions in a 3D Environment (PyTorch)
https://sites.google.com/view/gated-attention/home
MIT License
237 stars 37 forks source link

No GPU #12

Closed mmwebster closed 3 years ago

mmwebster commented 3 years ago

Any reason why cuda/gpus aren't used here?

devendrachaplot commented 3 years ago

Hi, This codebase uses the A3C algorithm for reinforcement learning, which requires a separate copy of the model for each thread, and thus, typically uses only CPUs. You can check the A3C paper for more details: https://arxiv.org/pdf/1602.01783.pdf

mmwebster commented 3 years ago

Ah I see. Thank you!