- Delete dockerfile and docker compose
- copy dockerfile and compose in either mac image for testing on a non gpu machine (not just mac), or gpu image to train on a gpu machine
- only if using gpu download cudnn 8.4.x and place it in extras/lib
- run docker compose up -d or docker build to make the image
- for gpu, you'll need to install nvidia drivers and nvidia-docker on your machine
- Be aware that this is very shallow detail, there maybe errors or edit that you'll run into that i cannot think about at the moment, for e.g, your cudnn download in extras/lib, will have to be written into extras/shell/cuda.sh, so that it can be installed. Other similar issues may exist. Be ware.
-
Notifications
You must be signed in to change notification settings - Fork 0
Mo379/TensorNets
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Applying TensorNetworks to the problem of MultiAgent Reinforcement Learning
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published