Environment Setup

Before you begin, please ensure that the following environment variables are correctly set: UDATADIR, UPRJDIR, and UOUTDIR. The operations you perform will only modify these three directories on your device.

Here's an example setup:

!!! Create a .env file in root dir of this project

# Put the following in the .env file in the project root# Example directory pathsexport UDATADIR=~/data # directory for datasetexport UPRJDIR=~/code # directory for codeexport UOUTDIR=~/output # directory for outputs such as logs# Example API Key and Worker settingsexport WANDB_API_KEY="xxx360492802218be41f76xxxx"# your Weights & Biases API keyexport NUM_WORKERS=0 # number of workers to use# Create directories if they do not exist
mkdir -p $UDATADIR$UPRJDIR$UOUTDIR

Using Docker

If you prefer to use Docker, you can find the Dockerfile in the .devcontainer directory. Please refer to Docker's documentation if you need guidance on building a Docker image and running a container.

RUN

For entry - train_diffuser

python entry/entry.py \
experiment=train_diffuser \
trainer.save_freq=100

For entry - plan_guided

python entry/entry.py \
experiment=plan_guided \
diffusion.epoch=latest \
policy.scale_grad_by_std=true \
guide._target_=diffuser.sampling.NoTrainGuideLonger \
trainer.custom_target=bl2tr \
policy.scale=0.1

For entry - train_values

python entry/entry.py \
experiment=train_values \
trainer.save_freq=100