LayoutParser Installation Video

[PatrickGold555]

After reading the issues board it seems like a lot of people are facing similar issues.

I have personally watched videos the LP team have posted on Youtube many times, but it seems like there is underlying configs / package installations that are not displayed or pointed out in the videos. from the issues board a lot of the issues seems to revolve around the detectron2 package.

I believe this project has real potential and a step by step installation video on different OS will be helpful, maybe even a community based instruction video set if others have managed to get this working locally and not in GoogleColab.

I have tried Ubuntu 22.04 as a system installation, but have resorted to using Ubuntu-22.04 within a WSL2 environment on windows, because i can quickly delete my WSL instance and start from scratch.

I Have created a list of packages i install on a fresh Ubuntu instance to try and LayoutParser installed, however i have the same issues others do (Detectron2) i hope my list can help others that are facing similar issues.

sudo apt-get update
sudo apt-get upgrade -y
sudo apt update
sudo apt install python3.10-venv -y
sudo apt install software-properties-common
sudo apt install build-essential -y
sudo apt-get install manpages-dev
sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt install build-essential zlib1g-dev libncurses5-dev libgdbm-dev libnss3-dev libssl-dev libreadline-dev libffi-dev libsqlite3-dev wget libbz2-dev -y
sudo apt install python3.9 -y
sudo apt install python3.9-venv -y
sudo apt-get purge build
sudo apt-get update
sudo apt install poppler-utils
sudo apt-get install libpoppler-dev -y

wget https://developer.download.nvidia.com/compute/cuda/repos/wsl-ubuntu/x86_64/cuda-wsl-ubuntu.pin
sudo mv cuda-wsl-ubuntu.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget https://developer.download.nvidia.com/compute/cuda/11.8.0/local_installers/cuda-repo-wsl-ubuntu-11-8-local_11.8.0-1_amd64.deb
sudo dpkg -i cuda-repo-wsl-ubuntu-11-8-local_11.8.0-1_amd64.deb
sudo cp /var/cuda-repo-wsl-ubuntu-11-8-local/cuda-*-keyring.gpg /usr/share/keyrings/
sudo apt-get update
sudo apt-get -y install cuda

Paste the following 2 lines at the bottom of your bashrc. file in your home directory
Then restart your machine and type [nvcc --version] to check if it correctly displays the Cuda drivers
(NO NOT RELY ON nvidia-smi command for cuda version)
If using a different CUDA version please change the CUDA version in the below 2 lines

export PATH=/usr/local/cuda-11.8/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-11.8/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

python3.9 -m venv lp

cd lp
source bin/activate

pip install --upgrade pip
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
pip install gcc7
pip install python-poppler
pip install compiletools
pip install jupyter
pip install numpy
pip install matplotlib
pip install spacy
pip install pandas
pip install wheel
pip install cython
pip install ninja
pip install poppler
pip install pycocotools
pip install funcy
pip install bs4
pip install scikit-learn
pip install imagesize
pip install tqdm
pip install tensorboard
pip install setuptools
pip install opencv-python
pip install layoutparser
pip install "layoutparser[effdet]"
pip install "layoutparser[paddledetection]"
pip install "layoutparser[ocr]"

git clone https://github.com/facebookresearch/detectron2.git
python -m pip install -e detectron2