This repository contains sample Python models for use with SAS Detection Architecture. See SAS Detection Architecture product documentation for how to import and use the models.
MLServer bring-your-own environment requires a conda environment packed into a tarball using conda-pack on the same architecture as runtime container. Here you can use any x86_64 linux docker image as <x86_64_LINUX_IMAGE>. We are using Miniforge3 as the conda distribution.
docker run --user root --rm -it -v $(pwd)/PYTEST_0200:/PYTEST_0200 --entrypoint=/bin/bash <x86_64_LINUX_IMAGE>
# install miniforge inside the container
curl -L https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-Linux-x86_64.sh -O
bash Miniforge3-Linux-x86_64.sh -b && export PATH=$PATH:/root/miniforge3/bin
cd /PYTEST_0200 && \
conda env create -f environment.yml && \
conda install -y conda-pack && \
conda pack --ignore-missing-files -f -n pytest -o environment.tar.gz
This project is licensed under the Apache 2.0 License.
This is a sample repository, the maintainers are not currently accepting contributions.