CosmiQ/cw-nets

cw-nets: built for inference against large scale geotiffs

# This repository is no longer being updated. Future development of code tools for geospatial machine learning analysis will be done at https://github.com/cosmiq/solaris.

cw-nets

Segmentation Nets designed for use with SpaceNet datasets and other remote sensing data

An example of the output of this tool can be found at https://cwnets-demo.netlify.com/

Installation

Using conda

Create Virtual Environment ` conda create -n cw-nets python-3.6 pip cython `

Install geospatial requirements ``` conda install --name cw-nets

rtree gdal

```

Install Deep Learning Frameworks: ` conda install pytorch torchvision cuda91 -c pytorch conda install opencv scikit-image `

Install CosmiQ tools ` pip install git+https://github.com/CosmiQ/cw-tiler.git@dataset_creation pip install git+https://github.com/CosmiQ/cw-nets.git@pytorch_generator `

Example

python create_mask.py --raster_path s3://spacenet-dataset/AOI_2_Vegas/srcData/rasterData/AOI_2_Vegas_MUL-PanSharpen_Cloud.tif

--output_name AOI_2_Vegas_v11.tif --data_output $OUTPUT_PATH --model_path weights/deepglobe_buildings.pt --cell_size 200 --stride_size 190 --tile_size 650

Dependencies

• cw-tiler https://github.com/CosmiQ/cw-tiler
• numpy
• tqdm
• shapely
• rasterio
• opencv
• scikit-image
• scikit-learn
• tensorflow
• keras
• torch
• torchvision

License

See LICENSE.txt.

Authors

See AUTHORS.txt.

Changes

See CHANGES.txt.