treora/cdtw

NumPy-based constrained Dynamic Time Warping (cDTW) module

cdtw's Issues

Memory Leak

Thanks. This is very helpful! :)
Anyway, If you have to calculate huge distance matrices there is a problem with not freeing memory.
I added
free(curr_column);
free(prev_column);
to the end of my copy of the procedure.

NumPy dependency unspecified in setup.py

NumPy is needed in two ways: the numpy package needs to be installed, and its C headers are used to compile the cdtw code. But getting both working at once seems tricky, and currently not worth my time to fix.

See:

• http://stackoverflow.com/questions/21605927/why-doesnt-setup-requires-work-properly-for-numpy
• http://stackoverflow.com/questions/2379898/make-distutils-look-for-numpy-header-files-in-the-correct-place

Workaround

If not already installed, install numpy manually (pip install numpy) before building this package.