garrettwrong / cuTWED

Linear memory CUDA Time Warp Edit Distance
GNU General Public License v3.0
28 stars 4 forks source link

Separate malloc and memcpy #2

Closed garrettwrong closed 4 years ago

garrettwrong commented 4 years ago

Move malloc and memcpy to their own function. This should expose the core twed to use provided gpu memory (and/or gpuarrays in the python bindings). Also create and wrap a host version that manages the copies as it is now.