Python Basics

Python differs from C++ or Fortran in several ways:

• Python is interpreted: There is no compilation step that turns the source code into machine instructions. Instead, you run python programs through an interpreter, usually called python.

  In our case, Jupyter will be our interface to the python interpreter.

• Python is dynamically typed: You don’t need to declare what type of data a variable will hold ahead of time.

• Python manages memory automatically: you don’t have to explicitly free memory when it is no longer needed. Python uses reference counting to determine if memory is still needed.

  Note: with modern C++, if you use the containers defined in the standard library (like vector), then memory management is largely taken care of for you as well. But it is still easy to get yourself in trouble in C++ if you are not careful.

• Python code is in general slower than compiled C++ or Fortran code. This is because it is interpreted. However, many libraries that we use through python are built in C and if we use them properly, we can recover a lot of the speed difference compared to a compiled language.