This repository contains code for Bayesian 3D inverse graphics.
The b3d/bayes3d/
subdirectory contains code for the bayes3d
project, and the b3d/chisight/
subdirectory contains code for post-bayes3D ChiSight systems (currently, SAMA4D).
Run the following command to launch your instance:
export ZONE="us-west1-a"
# Make sure to replace the value with a unique name!
export INSTANCE_NAME="your-instance-name-here"
gcloud compute instances create $INSTANCE_NAME \
--zone=$ZONE \
--image-family="common-cu123-ubuntu-2204-py310" \
--image-project=deeplearning-platform-release \
--maintenance-policy=TERMINATE \
--boot-disk-size=400GB \
--machine-type g2-standard-32 \
--accelerator="type=nvidia-l4,count=1" \
--metadata="install-nvidia-driver=True"
to ssh into your new machine, run:
gcloud compute config-ssh
ssh $INSTANCE_NAME.$ZONE.$PROJECT_ID
Add the following to your bash_rc:
export XLA_PYTHON_CLIENT_PREALLOCATE="false"
export XLA_PYTHON_CLIENT_ALLOCATOR="platform"
b3d
's GenJAX dependency is hosted in Google Artifact Registry. To configure your machine to access the package:
- Check if you can access the GenJAX Users
Group, and if not, run
\invite-genjax <google-account-email>
in any channel in the the probcomp Slack - Install the Google Cloud command line tools.
- Run
gcloud auth application-default login
. (This command needs to be rerun every time your machine reboots.)
Next, on the machine, close the b3d
repo:
git clone https://github.com/probcomp/b3d.git
cd b3d
Create and activate b3d
conda environment:
conda create -n b3d python=3.12
When that completes, run the install script:
bash -i install.sh
Then activate the conda environment:
conda activate b3d
Tunnel port 8812
for Rerun visualization by adding the RemoteForward
line to your ssh config:
Host xx.xx.xxx.xxx
HostName xx.xx.xxx.xxx
IdentityFile ~/.ssh/id_rsa
User thomasbayes
RemoteForward 8812 127.0.0.1:8812
Install rerun on local machine pip install rerun-sdk
and open viewer:
rerun --port 8812