search

Lydorn / Polygonization-by-Frame-Field-Learning

This repository contains the code for our fast polygonal building extraction from overhead images pipeline.
BSD 3-Clause "New" or "Revised" License
294 stars 70 forks source link

create virtual environment by using the environment.yml on windows 10 #14

Open XiaoyuSun-hub opened 3 years ago

XiaoyuSun-hub commented 3 years ago

Hi, I have just tried to install environment by enviroment.yml you provided, there is an error: image Could you please check that? thank you !

Lydorn commented 3 years ago

Hi, I think it does not work because I exported that enviroment.yml from Linux and it is not compatible with other platforms. You can have a look at the Dockerfile to see what to install. The main thing is to install pytorch 1.4 first and go from there.

XiaoyuSun-hub commented 3 years ago

@Lydorn , thank you for your reply, After I install packages in Dockerfile, Do I need to run setup.py, or install packages in setup.sh?

ashokdahal commented 3 years ago

@XiaoyuSun-hub To make it easier, the following packages are required to make it run in windows (or any environment) with anaconda. you can put it on loop and install or just run them all through anaconda powershell. conda install pyproj conda install libgdal conda install gdal conda install overpy conda install shapely conda install jsmin conda install tqdmy conda install scikit-learn conda install multiprocess pip install git+https://github.com/Lydorn/kornia@7bcb52125917eedee867ec93ed56c289019bd7d2 conda config --add channels conda-forge conda config --set channel_priority false conda install rasterio conda install scikit-image conda install geojson conda install tensorboard conda install fiona // visual studio build tool required required// pip install -U --no-cache-dir cython pip install --no-cache-dir "git+https://github.com/cocodataset/cocoapi.git#egg=pycocotools&subdirectory=PythonAPI" conda install tifffile pip install Pillow

conda install future conda install descartes conda install -c conda-forge opencv=4.5.0 conda install -c conda-forge skan

1.7.0 with cuda // conda install pytorch==1.7.0 torchvision==0.8.0 torchaudio==0.7.0 cudatoolkit=11.0 -c pytorch

pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" .

conda install numba=0.48.0

pip install torch-scatter -f https://pytorch-geometric.com/whl/torch-1.7.0+cu110.html //change version with matching to your anaconda version.