Skip to content

Latest commit

 

History

History
75 lines (49 loc) · 2.63 KB

installation.md

File metadata and controls

75 lines (49 loc) · 2.63 KB

Installation

We built our model with the mmdetection3d project. Users may refer to their getting_started documentation for more details on configurating local environment.


Download REDFormer Model

Download our project repository and move into the project root directory.

cd && git clone --depth=1 -b master https://https://github.com/PurdueDigitalTwin/REDFormer
cd REDFormer

Create Virtual Environment

Next we setup a virtual environment in conda for running the model.

Install conda package

Our model runs on a local virtual environment. We recommend Anaconda or Miniconda for this purpose. Please follow their instructions to install locally.We highly recommend.

Create conda environment

Now we create a virtual environment using conda for our model to run inside.

conda env create -f environment.yml

Activate environment and install mmdetection3d library

Use the following steps to install the mmdetection3d library locally.

cd && git clone https://github.com/open-mmlab/mmdetection3d.git && cd mmdetection3d
git checkout 1.0
conda activate redformer
pip install -e .

Caution:

If the aforementioned steps do not provide you with the correct version of the mmdet3d library, please refer to this page for detailed instructions.

To quite the virtual environment, use the following scripit:

conda deactivate redformer

Build the Project

Finally, we recommend you to install our project as a local PyPI development package. Inside the project root directory, run the following script to install the project:

pip install -e .

That's it! Now you are all set to go.