Section 1.1: %who - Discover Your Variables

What does the command %who do? This particular magic command lists all the variables currently defined in your Jupyter Notebook environment. For instance, consider the following code snippet:

import seaborn as sns

df = sns.load_dataset('mpg')

a = 'simple'

b = 2

When you input %who in your Jupyter Notebook cell, it will display all existing variables.

As illustrated in the image above, all the variables, including those pre-existing in the environment, are shown. If you're interested in viewing a specific variable type, simply append the object type to the command, such as %who str.

Section 1.2: %timeit - Measure Execution Speed

The %timeit command is quite intriguing. It helps you assess the execution speed of your code by running it multiple times and calculating the average along with the standard deviation of the execution time. For example, try this:

import numpy as np

%timeit np.random.normal(size=1000)

By using the %timeit command, you can observe that the execution time varies around 341 ns for each run. This command is particularly useful for evaluating the stability of your code and its looping processes.

Section 2.1: %store - Share Variables Across Notebooks

Have you ever needed to transfer variables from one Jupyter Notebook to another? The %store command simplifies this task. Instead of saving the variables as pickled objects, you can use %store to transfer them seamlessly.

For instance, if you wish to store the 'df' variable from your current notebook, just type %store df. Later, in a new notebook, you can retrieve it by entering %store -r df.

Section 2.2: %prun - Analyze Function Performance

The %prun command is another time-related magic command that evaluates the execution time of each function in your program. It provides a detailed table showing how many times each internal function was called, along with the time taken for each call and the cumulative time.

For example, execute:

%prun sns.load_dataset('mpg')

Section 2.3: %history - Review Your Commands

Have you ever lost track of the commands you've executed during your analysis? The %history command provides a log of your activity, allowing you to trace back your previous actions.

Simply type %history in your Jupyter Notebook cell and check the output.

Section 2.4: %pinfo - Get Detailed Information

When working with new objects or packages, you might want more detailed information. The %pinfo command helps you retrieve comprehensive details about your objects.

For instance, running %pinfo df will display all relevant information concerning the DataFrame object.

Section 2.5: %%writefile - Save Code as a Python File

Even though Jupyter Notebook may not be the ideal environment for development, you can still save your code as a Python file using the %%writefile command. For example:

%%writefile test.py

def number_awesome(x):

return 9

After executing this command, check your directory for the newly created Python file.

Section 2.6: %pycat - Read Python Files into Jupyter

If you want to read a Python file back into your Jupyter Notebook, the %pycat command can help. For example, to read the previous Python file:

%pycat test.py

A popup will display all the code within the Python file directly in your Jupyter Notebook.

Section 2.7: %quickref - Quick Reference for Magic Commands

Lastly, the %quickref command is invaluable, as it provides a comprehensive overview of all magic commands available in Jupyter Notebook. Running %quickref will present you with detailed explanations of each command.