Plotting and Programming in Python: Reference

Key Points

Reference

Running and Quitting

• Python files have the .py extension.
• Can be written in a text file or a Jupyter Notebook.
  • Jupyter notebooks have the extension .ipynb
  • Jupyter notebooks can be opened from Anaconda or through the command line by entering $ jupyter notebook
    • Markdown and HTML are allowed in markdown cells for documenting code.

Variables and Assignment

• Variables are stored using =.
  • Strings are defined in quotations '...'.
  • Integers and floating point numbers are defined without quotations.
• Variables can contain letters, digits, and underscores _.
  • Cannot start with a digit.
  • Variables that start with underscores should be avoided.
• Use print(...) to display values as text.
• Can use indexing on strings.
  • Indexing starts at 0.
  • Position is given in square brackets [position] following the variable name.
  • Take a slice using [start:stop]. This makes a copy of part of the original string.
    • start is the index of the first element.
    • stop is the index of the element after the last desired element.
• Use len(...) to find the length of a variable or string.

Data Types and Type Conversion

• Each value has a type. This controls what can be done with it.
  • int represents an integer
  • float represents a floating point number.
  • str represents a string.
• To determine a variables type, use the built-in function type(...), including the variable name in the parenthesis.
• Modifying strings:
  • Use + to concatenate strings.
  • Use * to repeat a string.
  • Numbers and strings cannot be added to on another.
    • Convert string to integer: int(...).
    • Convert integer to string: str(...).

Built-in Functions and Help

• To add a comment, place # before the thing you do not with to be executed.
• Commonly used built-in functions:
  • min() finds the smallest value.
  • max() finds the largest value.
  • round() rounds off a floating point number.
  • help() displays documentation for the function in the parenthesis.
    • Other ways to get help include holding down shift and pressing tab in Jupyter Notebooks.

Libraries

• Importing a library:
  • Use import ... to load a library.
  • Refer to this library by using module_name.thing_name.
    • . indicates ‘part of’.
• To import a specific item from a library: from ... import ...
• To import a library using an alias: import ... as ...
• Importing the math library: import math
  • Example of referring to an item with the module’s name: math.cos(math.pi).
• Importing the plotting library as an alias: import matplotlib as mpl

Reading Tabular Data into DataFrames

• Use the pandas library to do statistics on tabular data. Load with import pandas as pd.
  • To read in a csv: pd.read_csv(), including the path name in the parenthesis.
    • To specify a column’s values should be used as row headings: pd.read_csv('path', index_col='column name'), where path and column name should be replaced with the relevant values.
• To get more information about a DataFrame, use DataFrame.info, replacing DataFrame with the variable name of your DataFrame.
• Use DataFrame.columns to view the column names.
• Use DataFrame.T to transpose a DataFrame.
• Use DataFrame.describe to get summary statistics about your data.

Pandas DataFrames

• Select data using [i,j]
  • To select by entry position: DataFrame.iloc[..., ...]
    • This is inclusive of everything except the final index.
  • To select by entry label: DataFrame.loc[..., ...]
    • Can select multiple rows or columns by listing labels.
    • This is inclusive to both ends.
  • Use : to select all rows or columns.
• Can also select data based on values using True and False. This is a Boolean mask.
  • mask = subset > 10000
  • We can then use this to select values.
• To use a select-apply-combine operation we use data.apply(lambda x: x > x.mean()) where mean() can be any operation the user would like to be applied to x.

Plotting

• The most widely used plotting library is matplotlib.
  • Usually imported using import matplotlib.pyplot as plt.
  • To plot we use the command plt.plot(time, position).
  • To create a legend use plt.legend(['label1', 'label2'], loc='upper left')
    • Can also define labels within the plot statements by using plt.plot(time, position, label='label'). To make the legend show up, use plt.legend()
  • To label x and y axis plt.xlabel('label') and plt.ylabel('label') are used.
• Pandas DataFrames can be used to plot by using DataFrame.plot(). Any operations that can be used on a DataFrame can be applied while plotting.
  • To plot a bar plot data.plot(kind='bar')

import matplotlib.puplot as plot
plt.plot(time, position, label='label')
plt.xlabel('x axis label')
plt.ylabel('y axis label')
plt.legend()

Lists

• Defined within [...] and separated by ,.
  • An empty list can be created by using [].
• Can use len(...) to determine how many values are in a list.
• Can index just as done in previous lessons.
  • Indexing can be used to reassign values list_name[0] = newvalue.
• To add an item to a list use list_name.append(), with the item to append in the parenthesis.
• To combine two lists use list_name_1.extend(list_name_2).
• To remove an item from a list use del list_name[index].

For Loops

• Start a for loop with for number in [1, 2, 3]:, with the following lines indented.
  • [1, 2, 3] is considered the collection.
  • number is the loop variable.
  • The action following the collection is the body.
• To iterate over a sequence of numbers use range(start, end)

for number in range(0,5):
    print(number)

Conditionals

• Defined similarly to a loop, using if variable conditional value:.
  • For example, if variable > 5:.
• Use elif: for additional tests.
• Use else: for when if statement is not true.
• Can combine more than one conditional by using and or or.
• Often used in combination with for loops.
• Conditions that can be used:
  • == equal to.
  • >= greater than or equal to.
  • <= less than or equal to.
  • > greater than.
  • < less than.

for m in [3, 6, 7, 2, 8]:
    if m > 5:
        print(m, 'is large')
    elif m == 5:
        print(m, 'is 5')
    else:
        print(m, 'is small')

Looping Over Data Sets

• Use a for loop: for filename in [file1, file2]:
• To find a set of files using a pattern use glob.glob
  • Must import first using import glob.
  • * indicates “match zero or more characters”
  • ? indicates “match exactly one character”
    • For example: glob.glob(*.txt) will find all files that end with .txt in the current directory.
• Combine these by writing a loop using: for filename in glob.glob(*.txt):

for filename in glob.glob(*.txt):
  data = pd.read_csv(filename)

Writing Functions

• Define a function using def function_name(parameters):. Replace parameters with the variables to use when the function is executed.
• Run by using function_name(parameters).
• To return a result to the caller use return ... in the function.

def add_numbers(a, b):
    result = a + b
    return result

add_numbers(1, 4)

Variable Scope

• A local variable is defined in a function and can only be seen and used within that function.
• A global variable is defined outside of a function and can be seen or used anywhere after definition.

Programming Style

• Document your code.
• Use clear and meaningful variable names.
• Follow the PEP8 style guide when setting up your code.
• Use assertions to check for internal errors.
• Use docstrings to provide help.

Glossary

Arguments

Values passed to functions.

Array

A container holding elements of the same type.

Boolean

An object composed of True and False.

DataFrame

The way Pandas represents a table; a collection of series.

Element

An item in a list or an array. For a string, these are the individual characters.

Function

A block of code that can be called and re-used elsewhere.

Global variable

A variable defined outside of a function that can be used anywhere.

Index

The position of a given element.

Jupyter Notebook

Interactive coding environment allowing a combination of code and markdown.

Library

A collection of files containing functions used by other programs.

Local Variable

A variable defined inside of a function that can only be used inside of that function.

Mask

A boolean object used for selecting data from another object.

Method

An action tied to a particular object. Called by using object.method.

Modules

The files within a library containing functions used by other programs.

Parameters

Variables used when executing a function.

Series

A Pandas data structure to represent a column.

Substring

A part of a string.

Variables

Names for values.