Chapter 4. Profiling and Optimization

In this chapter, we will cover the following topics:

• Evaluating the time taken by a statement in IPython
• Profiling your code easily with cProfile and IPython
• Profiling your code line-by-line with line_profiler
• Profiling the memory usage of your code with memory_profiler
• Understanding the internals of NumPy to avoid unnecessary array copying
• Using stride tricks with NumPy
• Implementing an efficient rolling average algorithm with stride tricks
• Making efficient array selections in NumPy
• Processing huge NumPy arrays with memory mapping
• Manipulating large arrays with HDF5 and PyTables
• Manipulating large heterogeneous tables with HDF5 and PyTables