Performant Python

Performant Python

Once you have working code that you have tested to demonstrate that it is working correctly. You might want to explore how efficient the code runs. This lesson will explore numpy, numba, scipy and Cython. Each of these explore different methods to (attempt to) improve the performance of Python to comparable levels with compiled languages such as C and Fortran.

Prerequisites:

To complete this lesson you should be familiar with:

• Writing numerical Python codes

Schedule:

Time	Episode	Description
0:00	Review of Nummpy	Review Numpy arrays
0:20	Copying and accessing arrays	How to copy and access subsets of arrays
0:40	Manipulating Array	Manipulating Arrays
1:00	File handling	Reading and Writing Files with Numpy
1:15	Something Random	Something random in Numpy
1:30	Vectorisation	Vector operations (for free) in Numpy)
1:45	Performance	Performance with Python
2:00	Numba	Numba jit library
2:30	Scipy	Scipy
3:00	Cython	Cython

Be sure to edit the links above.

Setup:

Running notebooks on azure and clone the library and Run on Free ...