Visual novelty dataset for the Polycraft domain.
If you do not have Pipenv installed, run the following in Python 3.10:
pip install pipenv
The dependencies can be installed within a Pipenv with the following commands:
pipenv install --categories "packages torch_cpu"
PyTorch may require different versions depending on the machine it is running on. The default command is for non-CUDA machines while swapping torch_cpu for torch_cu116 installs PyTorch for CUDA 11.6. If a non-default version of PyTorch is required then generate the appropriate Pip command on the PyTorch website then run it within the Pipenv by prepending pipenv run to it.
The following instructions are not recommended unless you are unable to install a Python version compatible with the Pipenv.
For this installation PyTorch will be installed in the Conda environment using the appropriate command according to the PyTorch website. For example:
conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch
Then the Pipenv must be configured for the Conda environment before install:
pipenv --python=$(conda run which python) --site-packages
pipenv install
To run unit tests, run the following command:
pipenv run python -m pytest