Calculate_feature_matrix only uses functions from Pandas/Numpy/Scipy for increasing performance over raw Python and numba or cython helps specifically in PandasBackend class and in individual feature computation functions.It’s complexity is more for installation because it requires a compiled C code or a C compiler.
Due to this complexity it is low in our priority list. So, we are focused on scaling our small datasets towards larger datasets which demands parallelization rather than subroutine optimization.
If you want to know more about Machine Learning then watch this video:
Gain practical exposure with data science projects in Intellipaat's Data Science course online.
Back