Advanced features

Use float instead of double

The fast version is based on C, which defaults to the double datatype. It is possible to compile the C code to use float or integers if that is required (requires a C compiler and make):

git clone https://github.com/wannesm/dtaidistance.git
cd dtaidistance/dtaidistance/jinja
make float
# make int
cd ../..
make build
pip install .