View on GitHub

IPython Deep Dive: Python Data Science Handbook, Ch. 1

Reading: "IPython: Beyond Normal Python", by Jake VanderPlas — Python Data Science Handbook, Chapter 1.

1. Help & Documentation

• object? — shows the docstring for any object, function, or method
• object?? — shows the underlying source code, when available

2. Tab Completion

• object. + TAB — lists an object's attributes and methods
• Wildcard matching with * — e.g. str.*find*?

3. Magic Commands

• %timeit / %%timeit — time a line / a whole cell
• %history — show past commands
• %run — execute an external .py script
• %load — load a script's contents into a cell

4. Input / Output History

• In[] and Out[] — every input and output is numbered and stored
• _ — the last output; __ — the one before that
• Out[2] — re-access any earlier output by its number

5. Shell Access from IPython

• !ls, !pwd — run any shell command with a ! prefix
• contents = !ls — capture shell output into a Python variable
• !cp {filename} {dest} — pass Python variables back into the shell

6. Errors & Debugging

• %xmode — control how much detail a traceback shows
• %debug — drop into an interactive debugger at the point of the last error

7. Profiling & Timing

• %prun — profile a statement, function by function
• %lprun — line-by-line profiling (needs line_profiler)
• %memit — measure memory use of a statement