Demonstration of aws lambda: HTTP based time series forecasting service.
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
pip uninstall -y matplotlb
find "$VIRTUAL_ENV/lib64/python3.7/site-packages" -name "test" | xargs rm -rf
find "$VIRTUAL_ENV/lib64/python3.7/site-packages" -name "tests" | xargs rm -rf
rm -rf "$VIRTUAL_ENV/lib64/python3.7/site-packages/pystan/stan/src"
rm -rf "$VIRTUAL_ENV/lib64/python3.7/site-packages/pystan/stan/lib/stan_math/lib"
find "$VIRTUAL_ENV/lib64/python3.7/site-packages" -name "*.so" | xargs strip --strip-debug
Now prepare the "vendor" subdirectory:
mkdir vendor
cp -r $VIRTUAL_ENV/lib64/python3.7/site-packages/* vendor/