This is the list of commands which I know in Python language
help(variable_name.methodname)
pydoc <commmand>
Jupyter Notebook / Jupyterlab
?variable_name.methodname
To find the execution time
%%time
To see what a particular function does, we can press SHIFT+Tab and check
Statements: A statement is an instruction that can be executed. Every line of code we have written so far is a statement e.g. assigning a variable, calling a function, conditional statements using
if,else, andelif, loops usingforandwhileetc.
Expressions: An expression is some code that evaluates to a value. Examples include values of different data types, arithmetic expressions, conditions, variables, function calls, conditional expressions, etc.
Storing multiple data in one line
a, b, c = 12, 13, 14
d = e = f = 15
We can use single quotes inside a string written with double quotes, and vice versa.
text1 = "I thought to myself 'This changes everything'"
text2 = 'I thought to myself "This changes everything"'
To use a double quote within a string written with double quotes, escape the inner quotes by prefixing them with the \ character.
text3 = "I thought to myself \"This changes everything\""
text4 = 'I thought to myself \'This changes everything\''
Strings created using single or double quotes must begin and end on the same line. To create multiline strings, use three single quotes ''' or three double quotes """ to begin and end the string. Line breaks are represented using the newline character \n.
text5 = """I thought to myself
This changes everything"""
type(10/3) # float
type(10/2) # float
type(10//2) # int
Booleans are automatically converted to
intwhen used in arithmetic operations.Trueare converted to1andFalseare converted to0.
Only the following values are evaluated as False, (falsy values):
- The
Falseitself - The integer
0 - The float
0.0 - The empty value
None - The empty text
"" - The empty list
[] - The empty tuple
() - The empty dictionary
{} - The empty set
set() - The empty range
range(0)
Any value in Python can be converted to a boolean using bool() function
text6 = bool(range(10)) # True is stored in text6
The None type includes a single value
None, used to indicate the absence of a value.Nonehas the typeNoneType. It is often used to declare a variable whose value may be assigned later
nothing = None
text7 = type(nothing) # stores "NoneType" in text7
String
multiline_string = """a
b"""
text8 = list(multiline_string) # stores ['a','\n','b']
Methods: Methods are functions associated with data types and are accessed using the
.notation e.g.variable_name.method()or"a string".method(). Methods are a powerful technique for associating common operations with values of specific data types.
cost_of_ice_bag = 1.25
profit_margin = .2
number_of_bags = 500
text9 = """If a grocery store sells ice bags at $ {} per bag, with a profit margin of {} %,
then the total profit it makes by selling {} ice bags is $ {}."""
text10 = text9.lower()
text10 = text9.upper()
text10 = text9.capitalize() # changes the first character to uppercase
text10 = text9.replace("sample1", "sample2") # sample1 will be replaced by sample2, it only returns the value once, it does not change the orignal text fully
text10 = text9.split("anysymbol") # this outputs a list
text10 = text9.strip() # removes whitespace in the beginning and ending of the string
text10 = text9.format(parameter1, parameter2, ...) # See the example below, placeholder `{}` in text9 is neccessary
text10 = text9.format(cost_of_ice_bag, profit_margin*100, number_of_bags, total_profit)
print(text10)
print("""If a grocery store sells ice bags at $ {} per bag, with a profit margin of {} %,
then the total profit it makes by selling {} ice bags is $ {}.""".format(cost_of_ice_bag, profit_margin*100, number_of_bags, total_profit))
print('This is a value {:.0f}'.format(32.3456)) # 32
print('This is a value {:.1f}'.format(32.3456)) # 32.3
List
list1 = ['a','b','c','d']
text11 = list1.pop(3) # text11 stores 'd' but deletes the value from list1
list2 = list1.copy() # Changing the content of list2 will not affect the content of list1
Tuple
single_element_tuple1 = 4, # it stores as (4,)
single_element_tuple1 = (3,)
not_tuple = (4) # it stores as 4
a_tuple = 34, True, None, "Mubin", 23.3, "Moriom"
point = (3,4)
point_x, point_y = point # point_x = 3 and point_y = 4
# Converting a list into a tuple and vice verse, use tuple() and list() respectively
Dictionaries
# Dictionaries can also be created using dict function
person1 = dict( name = 'Mubin', gender = 'male', age = 20, married = False)
print(person1) # outputs {'name' : 'Mubin', 'gender' : 'male', 'age' : 20, 'married' : False}
# We can use get method to access the value associated with a key
text = person1.get('name')
To remove a key and associated value from a dictionary, use the pop method
text = person1.pop('age') # stores the value associated with the key 'age'
list1 = person1.keys() # it stores dict_keys(['name', 'gender', 'age', 'married'])
list2 = person1.values() # it stores dict_values(['Mubin', 'male', 20, False])
list3 = person1.items() # it stores dict_items([('name', 'Mubin'), ('gender', 'male'), ('age', 20), ('married', False)])
list1 = list(list1) # now we can access the values from list1 treating it as a list, like list1[0] outputs 'name'
Pass Statement
if num % 2 == 0:
pass # it does nothing
elif num % 3 == 0:
print("Divisible by 3")
Enumerate function
a_list = ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday']
for i in range(len(a_list)):
print('The value at position {} is {}.'.format(i, a_list[i]))
for i, val in enumerate(a_list):
print('The value at position {} is {}.'.format(i, val))
Pair
for x, y in zip(x_cordinate, y_coordinate):
result += x^2 + y^2
def cube(y):
return y*y*y
lambda_cube = lambda y: y*y*y
print(cube(5))
print(lambda_cube(5))
a,b = 1, 2
print("a" if a>b else "b")
[on_true] if [expression] else [on_false]
import random
a, b = random.randint(2,10), random.randint(0,10)
result1 = "a" if a>b else "b"
print(result)
([on_false], [on_true]) [expression] # Using tuples
result2 = (b,a) [a>b]
result3 = (f"b:{b}", f"a:{a}")[a>b]
{True:[on_true],False:[on_false]}[expression] # Using dictionaries
result3 = {True:f"b:{b}",False:f"a:{a}"}[a<b]
print(result3)
result4 = {False:f"b:{b}",True:f"a:{a}"}[a<b]
print(result4)
(lambda: [on_true], lambda: [on_false])expression # Using lambda function
result5 = (lambda: f"a:{a}", lambda: f"b:{b}")[a>b]()
print(result5)
result6 = "Less than or equal to zero" if a<=0 else "Between 0 and 3" if a>0 and a<3 else "Greater than 2"
print(result6)
try:
print(x)
except:
print("An exception occurred")
NameError
try:
print(x)
except NameError:
print("Variable is not defined")
except:
print("Something else went wrong!")
elsekeyword is used to define a block of code to be executed if no errors were raised
try:
print("Hello!")
except:
print("Something went wrong!")
else:
print("Nothing went wrong!")
finallykeyword is used to define a block of code to be executed regardless if thetryblock raises an error or not
try:
print(x)
except:
print("Something went wrong!")
finally:
print("The try-except is finished!")
An example of trying to open and write an file that is not writable
try:
f = open("demofile.txt")
try:
f.write("Moriomubin")
except:
print("Something went wrong when writing to the file!")
finally:
f.close()
except:
print("Something went wrong when opening the file")
iskeyword is used to test whether two variables belong to the same object. The test will returnTrueif the two objects are same else it will returnFalseeven if the two objects are 100% equal. Note: the==operator is used to test if the two objects are same.
a = 10
b = 10
if a is b:
print(True)
else:
print(False)
# Prints True
x = 'Mubin'
y = 'Mubin'
if x is y:
print(True)
else:
print(False)
# Prints True
ab = ['a', 'b', 'c']
ac = ['a', 'b', 'c']
if ab is ac:
print(True)
else:
print(False)
# Prints False
print(ab is ac) # outputs False
print(ab == ac) # outputs True
Raising an exception
x = -1
if x < 0:
raise Exception("Sorry, no numbers below zero!")
y = 'hello'
if not type(y) is int:
raise TypeError("Only integers are allowed!")
Named Arguments: Python allows to enter named arguments when calling a function
def emi_calculate(amount, duration):
return amount/duration
loan_calculate(amount = 12000, duration = 12)
A docstring describes what the function does, and provide some explanations about the arguments. It is used by the
helpfunction.
def loan_emi(amount, duration, rate, down_payment=0):
"""Calculates the equal montly installment (EMI) for a loan.
Arguments:
amount - Total amount to be spent (loan + down payment)
duration - Duration of the loan (in months)
rate - Rate of interest (monthly)
down_payment (optional) - Optional intial payment (deducted from amount)
"""
loan_amount = amount - down_payment
try:
emi = loan_amount * rate * ((1+rate)**duration) / (((1+rate)**duration)-1)
except ZeroDivisionError:
emi = loan_amount / duration
emi = math.ceil(emi)
return emi
help(loan_emi)
# Outputs the docstring
import math
a = math.ceil(1.2)
from math import ceil
a = ceil(1.2)
import os
os.cwd() # Checking current present directory
os.listdir() # Checking the list of files in a directory
os.listdir('.') # checking the relative path
os.makedirs('./foldername', exist_ok = True) # Creating a new directory in the present directory. If exist_ok is False (the default), a FileExistsError is raised if the target directory already exists.
a = 'foldername' in os.listdir('.') # Stores True if the folder name is present in the present directory
b = os.listdir('./foldername') # Absolute path, list of files inside a directory of present directory. If that directory is empty, empty list is stored in b
Reading from a file
file1 = open('filename.filetype', mode = 'r') # Filename can have absolute path starting with './', there are different modes, check those with help function
file1_contents = file1.read()
print(file1_contents)
file1.close() # After closing the file, file1.read() will not work
Closing files automatically using
with
with open('./demofolder/filename.txt', mode = 'r') as file2:
file2_contents = file2.read()
print(file2_contents)
Reading a file line by line
with open('./demofolder/filename.txt', mode = 'r') as file2:
file2_contents = file2.readlines() # Each line is stored as an element of a list
print(file2_contents)
Processing data from files
def parse_headers(header_line):
return header_line.strip().split(',')
def parse_values(data_line):
values = []
for item in data_line.strip().split(','):
if item == '':
values.append(0.0)
else:
try:
values.append(float(item))
except ValueError:
values.append(item)
return values
def create_item_dict(values, headers):
result = {}
for value, header in zip(values, headers):
result[header] = value
return result
def read_csv(path):
result = []
# Open the file in read mode
with open(path, 'r') as f:
# Get a list of lines
lines = f.readlines()
# Parse the header
headers = parse_headers(lines[0])
# Loop over the remaining lines
for data_line in lines[1:]:
# Parse the values
values = parse_values(data_line)
# Create a dictionary using values & headers
item_dict = create_item_dict(values, headers)
# Add the dictionary to the result
result.append(item_dict)
return result
def write_csv(items, path):
# Open the file in write mode
with open(path, 'w') as f:
# Return if there's nothing to write
if len(items) == 0:
return
# Write the headers in the first line
headers = list(items[0].keys())
f.write(','.join(headers) + '\n')
# Write one item per line
for item in items:
values = []
for header in headers:
values.append(str(item.get(header, "")))
f.write(','.join(values) + "\n")
with open('./data/emis2.txt', 'w') as f:
for loan in loans2:
f.write('{},{},{},{},{}\n'.format(
loan['amount'],
loan['duration'],
loan['rate'],
loan['down_payment'],
loan['emi']))
To install numpy
!pip install numpy --upgrade --quiet
Array
import numpy as np
a = np.array([23,32,22,33])
b = np.array([.4,.5,.6,.7])
ab = a*b
ab_dot = np.dot(a,b) # Dot product of the arrays a and b
ab_sum = ab.sum() # Same as ab_sum = (a*b).sum()
type(a) # numpy.ndarray
a.shape # (4,)
a.dtype # dtype('int32')
aa = np.array([
[12, 13, 14, 15],
[23, 24, 25, 26],
[34, 35, 36, 37],
[45, 46, 47, 48]
])
aa.shape # (4,4) == (outer dimension, inner dimension)
aaa_matrixmul1 = np.matmul(aa, a)
aaa_matrixmul2 = aa @ a
Urllib.request Module
import urllib.request
urllib.request.urlretrieve(
'https://gist.github.com/BirajCoder/a4ffcb76fd6fb221d76ac2ee2b8584e9/raw/4054f90adfd361b7aa4255e99c2e874664094cea/climate.csv',
'climate.txt')
from urllib.request import urlretrieve
urlretrieve('https://gist.github.com/BirajCoder/a4ffcb76fd6fb221d76ac2ee2b8584e9/raw/4054f90adfd361b7aa4255e99c2e874664094cea/climate.csv', 'climate.txt')
# urlretrieve('https://sample.url', './demofolder/climate.txt')
Converting txt into array
data = np.genfromtxt('filename.txt', delimiter = ',', skip_header = 1)
Writing txt into a file
np.savetxt('filename.txt', variable_name, fmt = '%.2f', delimiter = ',', header = 'header1, header2, header3', comments = '')
Reshaping the array
a_reshaped1 = a.reshape(2, 2)
a_reshaped2 = a.reshape(4, 1) # (outer dimension, inner dimension) == (row, column)
Concatenate function
a = np.array([[1,2],[3,4]]) #(2, 2)
b = np.array([[5,6]]) #(1, 2)
concated1 = np.concatenate((a,b), axis = 0) #a and b has to be same dimensional array
concated2 = np.concatenate((a,b.T), axis = 1) #np.concatenate((a, b.reshape(2,1)), axis = 1)
concated3 = np.concatenate((a,b), axis = None) #a and b does not need to be in same dimension
Mathematics:
np.sum,np.exp,np.round, arithemtic operators
Array manipulation:
np.reshape,np.stack,np.concatenate,np.split
Linear Algebra:
np.matmul,np.dot,np.transpose,np.eigvals
Statistics:
np.mean,np.median,np.std,np.max
NumPy Documentation : numpy
Forming arrays
# Range with start, end and step
a1 = np.arange(5) # array([0, 1, 2, 3, 4])
a2 = np.arange(3, 20) # array([ 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19])
a2 = np.arange(3, 20, 4) # array([ 3, 7, 11, 15, 19])
b1 = np.ones(4) # array([1., 1., 1., 1.])
b2 = np.ones((4,)) # array([1., 1., 1., 1.])
b3 = np.ones((3,3)) # array([[1., 1., 1.],[1., 1., 1.],[1., 1., 1.]])
b4 = np.ones([2, 2, 3]) # array([[[1., 1., 1.],[1., 1., 1.]],[[1., 1., 1.],[1., 1., 1.]]])
c1 = np.zeros(4) # array([0., 0., 0., 0.])
c2 = np.zeros((4,)) # array([0., 0., 0., 0.])
c3 = np.zeros((4,2)) # array([[0., 0.],[0., 0.],[0., 0.],[0., 0.]])
c4 = np.zeros([2,2,3]) # array([[[0., 0., 0.],[0., 0., 0.]],[[0., 0., 0.],[0., 0., 0.]]])
# Identity Matrix
d = np.eye(3) # array([[1., 0., 0.],[0., 1., 0.],[0., 0., 1.]])
# Random Vector forms from uniform distribution [0,1)
e1 = np.random.rand(4) # array([0.67720034, 0.78128189, 0.9746822 , 0.66433952])
# Random matrix
e2 = np.random.rand(2,3) # array([[0.21728667, 0.87190509, 0.89748379],[0.47179849, 0.83554905, 0.86015455]])
f1 = np.random.randn(4) # standard normal distribution (includes negative numbers) # array([-0.46426101, 0.36486801, 1.03339286, 1.29384229])
f2 = np.random.randn(2,3) # array([[ 1.33146919, -1.95268722, 1.55484275],[ 0.71057842, -0.46610497, 1.37420616]])
# a n-dimensional array with a specific number
g1 = np.full([2,3], 20)
g2 = np.full([2, 3, 4], 21)
g3 = np.full([2,3,4,5], 22)
# Equally spaced numbers in a range
h1 = np.linspace(3, 18, 7) # array([ 3. , 5.5, 8. , 10.5, 13. , 15.5, 18. ])
Scaling
a1 + 2 # Adds 2 to each element of the the array a1 a1 - 2 # Subtracts 2 from each element of the array a1 a1 % 2 # Find the modulus from each element of the array a1 a1 / 2 # Division by scalar a1 * a2 # Element-wise multiplication
Broadcasting
Broadcasting only works if one of the arrays can be replicated to match the array's shape.
a = np.arange(3) + 5 # a = array([5, 6, 7]) which is array([0, 1, 2]) + array([5, 5, 5])
a = array([0,1,2]) + 5
b = np.ones((3,3)) + np.arange(3) # b = array([[1,2,3],[1,2,3],[1,2,3]]) which is array([[1,1,1],[1,1,1],[1,1,1]]) + array([[0,1,2],[0,1,2],[0,1,2]])
b = array([[1,1,1],[1,1,1],[1,1,1]]) + array([0,1,2])
c = np.arange(3).reshape(3,1) + np.arange(3) # c = array([[0,1,2],[1,2,3],[2,3,4]]) which is array([[0,0,0],[1,1,1],[2,2,2]]) + array([[0,1,2],[0,1,2],[0,1,2]])
c = array([[0],[1],[2]]) + array([0,1,2])
Array Comparison
arr1 = np.array([[1, 2, 3],[3, 4, 5]])
arr2 = np.array([[2, 2, 3],[1, 2, 5]])
mb = arr1 == arr2 # mb = array([[False, True, True],[False, False, True]]) same rule applies to other comparison operators such as !=, >, >=, <, <=
ms = (arr1 == arr2).sum()
Array Indexing & Slicing
arr3 = np.array(
[[[11, 12, 13, 14],
[13, 14, 15, 19]],
[[15, 16, 17, 21],
[63, 92, 36, 18]],
[[98, 32, 81, 23],
[17, 18, 19.5, 43]]])
arr3.shape # (3, 2, 4)
a = arr3[0] # a = array([[11., 12., 13., 14.],[13., 14., 15., 19.]])
b = arr3[0][0] # b = array([11., 12., 13., 14.]) same as arr3[0, 0]
c = arr3[0][0][0] # c = 11.0 same as arr3[0,0,0]
# Subarray using ranges
aa = arr3[1:, :1, :3] # aa = array([[[15,16,17]],[[98,32,81]]])
aa.shape # (2,1,3)
# Mixing indices and ranges
ab = arr3[1:, 0, :3] # ab = array([[15,16,17]],[[98,32,81]])
ab.shape # (2,3)
ac = arr3[1:, 0, 0] # ac = array([15, 98])
ac.shape # (2,)
To install Panda module
!pip install pandas --upgrade --quiet
Initial code
import panda as pd
covid_df = pd.read_csv('filename.csv')
type(covid_df) # Data type is DataFrame, full form is pandas.core.frame.DataFrame
print(covid_df) # The data is printed as a table, together with indexing
covid_df.info() # Summary of the data is returned, (Data type, Column Header, Number of Rows and many more)
covid_df.describe() # Statistical information for numerical columns (mean, standard deviation, minimum/maximum values, and the number of non-empty values)
covid_df.columns # Get the list of column names
covid_df.shape # Get the number of rows & columns as a tuple
Retrieving data from DataFrame
Panda data storing format is similar to a dictionary, not a list of dictionary
a = covid_df['date'] # a series of data is stored in a
Each column is represented using a data structure called Series
type(covid_df['date']) # pandas.core.series.Series
covid_df['new_cases'][246] # Retrieving specific element
covid_df.at[246, 'new_cases'] # Retrieving specific element by at method
covid_df.date # Pandas also allows accessing columns as properties of the dataframe using the `.` notation instead of `[]`
covid_df[['date', 'new_cases']] # Retrieving a list of columns within the indexing notation []
cases_df = covid_df[['date', 'new_cases']] # cases_df and covid_df both point to the same data in the computer's memory. Changing one will change the other.
covid_df_copy = covid_df.copy() # covid_df_copy is completely separate from covid_df
covid_df.loc[243] # To access a specific row of data
covid_df.loc[108:113]
type(covid_df.loc[243]) # pandas.core.series.Series
covid_df.head(5) # To view the first few rows of data
covid_df.tail(4) # To view the last few rows of data
b = covid_df.at[0, 'new_tests'] # nan is stored
type(covid_df.at[0, 'new_tests']) # numpy.float64
covid_df.new_tests.first_valid_index() # To find the first index that doesn't contain a NaN value using first_valid_index method.
covid_df.sample(10) # To retrieve a random sample of rows from the data frame.
total_cases = covid_df.new_cases.sum() # To find the total value
Querying and sorting rows
high_new_cases = covid_df.new_cases > 1000 # high_new_cases is a series containing True and False boolean values
covid_df[covid_df.new_cases > 1000] # Querying a subset of rows satisfying the chosen criteria using boolean expressions
df['pos_rate'] = df.new_cases/df.new_tests # Adding new columns by combining data from existing columns
covid_df.drop('positive_rate') # Removing one or more columns from the data frame
covid_df.drop(columns=['positive_rate'], inplace=True)
sort_values # Sorting the rows of a data frame using column values
covid_df.sort_values('new_cases').head(10)
covid_df.sort_values('new_deaths', ascending=False).head(10)
covid_df.at[172, 'new_cases'] = ... # Replacing a value within the data frame
Changing display option
from IPython.display import display
with pd.option_context('display.max_rows', 100):
display(covid_df[covid_df.new_cases > 1000])
Working with dates
covid_df['date'] = pd.to_datetime(covid_df.date) # To convert it into a column of datetime datatype using the pd.to_datetime
# To extract different parts of the datatime datatype using the DatetimeIndex class
covid_df['year'] = pd.DatetimeIndex(covid_df.date).year
covid_df['month'] = pd.DatetimeIndex(covid_df.date).month
covid_df['day'] = pd.DatetimeIndex(covid_df.date).day
covid_df['weekday'] = pd.DatetimeIndex(covid_df.date).weekday
Finding different statistics
covid_df.new_cases.mean()
covid_df[covid_df.weekday == 6].new_cases.mean()
Grouping and Aggregation
We have to use the groupby function to create a group for each month, select the columns we wish to aggregate, and aggregate them using the sum (any statistics) method.
covid_month_df = covid_df.groupby('month')[['new_cases', 'new_deaths', 'new_tests']].sum()
covid_month_mean_df = covid_df.groupby('month')[['new_cases', 'new_deaths', 'new_tests']].mean()
We can use the cumsum method to compute the cumulative sum of a column as a new series
covid_df['total_deaths'] = covid_df.new_deaths.cumsum()
covid_df['total_tests'] = covid_df.new_tests.cumsum() + initial_tests # NaN values in the total_tests column remain unaffected.
Merging data from multiple source
To merge two data frames, we need at least one common column.
# To add the columns from `locations_df` into `covid_df` using the `.merge` method.
merged_df = covid_df.merge(locations_df, on="location")
Writing back to files
To write the data from the data frame into a file, we can use the to_csv function.
result_df.to_csv('results.csv') # Index column from the data frame also gets stored
result_df.to_csv('results.csv', index=None) # Index column from the data frame is ignored
!pip install matplotlib seaborn --upgrade --quiet
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline # Without this command, sometimes plots may show up in pop-up windows.
Line Chart
yield_apples = [0.895, 0.91, 0.919, 0.926, 0.929, 0.931]
years = [2010, 2011, 2012, 2013, 2014, 2015]
plt.plot(yield_apples); # A semicolon (;) at the end to avoiding showing some output and display just the graph.
apples = [0.895, 0.91, 0.919, 0.926, 0.929, 0.931, 0.934, 0.936, 0.937, 0.9375, 0.9372, 0.939]
oranges = [0.962, 0.941, 0.930, 0.923, 0.918, 0.908, 0.907, 0.904, 0.901, 0.898, 0.9, 0.896 ]
years = range(2000, 2012)
plt.plot(years, apples) # Customizing the X-axis
plt.plot(years, oranges) # Plotting Multiple Lines in a single chart
plt.xlabel('Year') # Axis Labels
plt.ylabel('Yield (tons per hectare)')
plt.title("Crop Yields in Kanto") # Chart Title
plt.legend(['Apples', 'Oranges']) # Chart Legend
plt.plot(years, apples, marker='o') # Line Markers
The plt.plot function supports many arguments for styling lines and markers:
color or c: Set the color of the line (supported colors)
linestyle or ls: Choose between a solid or dashed line
linewidth or lw: Set the width of a line
markersize or ms: Set the size of markers
markeredgecolor or mec: Set the edge color for markers
markeredgewidth or mew: Set the edge width for markers
markerfacecolor or mfc: Set the fill color for markers
alpha: Opacity of the plot
plt.plot(years, apples, marker='s', c='b', ls='-', lw=2, ms=8, mew=2, mec='navy')
plt.plot(years, oranges, marker='o', c='r', ls='--', lw=3, ms=10, alpha=.5)
The fmt argument provides a shorthand for specifying the marker shape, line style, and line color. It can be provided as the third argument to plt.plot.
`fmt` = '[marker][line][color]'
plt.plot(years, oranges, 'o--r') # This is for dashed line, for solid line use 'o-r'
plt.plot(years, oranges, 'or') # Only markers are drawn
Changing the Figure Size
plt.figure(figsize=(12, 6))
Improving Default Styles using Seaborn globally (Full list)
sns.set_style("whitegrid")
sns.set_style("darkgrid")
We can also edit default styles directly by modifying the matplotlib.rcParams dictionary.
import matplotlib
matplotlib.rcParams['font.size'] = 14
matplotlib.rcParams['figure.figsize'] = (9, 5)
matplotlib.rcParams['figure.facecolor'] = '#00000000'
Scatter Plot
The Iris dataset is included with the Seaborn library and can be loaded as a Pandas data frame.
flowers_df = sns.load_dataset("iris")
flowers_df.species.unique()
plt.plot(flowers_df.sepal_length, flowers_df.sepal_width);
sns.scatterplot(x=flowers_df.sepal_length, y=flowers_df.sepal_width);
sns.scatterplot(x=flowers_df.sepal_length, y=flowers_df.sepal_width, hue=flowers_df.species, s=100); # We can color the dots using the flower species as a hue.
plt.figure(figsize=(12, 6)) # Customizing Seaborn Figures
plt.title('Sepal Dimensions')
sns.scatterplot(x='sepal_length', y='sepal_width', hue='species',s=100, data=flowers_df); # Seaborn has in-built support for Pandas data frames
Histogram
plt.title("Distribution of Sepal Width")
plt.hist(flowers_df.sepal_width);
plt.hist(flowers_df.sepal_width, bins=5); # Controlling the size and number of bins
import numpy as np
plt.hist(flowers_df.sepal_width, bins=np.arange(2, 5, 0.25));
plt.hist(flowers_df.sepal_width, bins=[1, 3, 4, 4.5]);
# Multiple Histograms
plt.hist(setosa_df.sepal_width, alpha=0.4, bins=np.arange(2, 5, 0.25))
plt.hist(versicolor_df.sepal_width, alpha=0.4, bins=np.arange(2, 5, 0.25));
plt.hist([setosa_df.sepal_width, versicolor_df.sepal_width, virginica_df.sepal_width], bins=np.arange(2, 5, 0.25), stacked=True)
plt.legend(['Setosa', 'Versicolor', 'Virginica']);
Bar Chart
plt.bar(years, oranges);
plt.bar(years, apples)
plt.bar(years, oranges, bottom=apples);
# Bar Plots with Averages
tips_df = sns.load_dataset("tips"); # Another sample dataset included with Seaborn, called tips
sns.barplot(x='day', y='total_bill', data=tips_df);
sns.barplot(x='day', y='total_bill', hue='sex', data=tips_df); # We can make the bars horizontal simply by switching the axes.
Heatmap
flights_df = sns.load_dataset("flights").pivot("month", "year", "passengers") # Another sample dataset included with Seaborn, called flights
plt.title("No. of Passengers (1000s)")
sns.heatmap(flights_df)
sns.heatmap(flights_df, fmt="d", annot=True, cmap='Blues')
Images
from PIL import Image
img = Image.open('chart.jpg')
An image loaded using PIL is simply a 3-dimensional numpy array containing pixel intensities for the red, green & blue (RGB) channels of the image. We can convert the image into an array using np.array.
img_array = np.array(img)
img_array.shape
# We can display the PIL image using plt.imshow.
plt.imshow(img)
# Turning off the axes & grid lines and showing a title
plt.grid(False)
plt.title('A data science meme')
plt.axis('off')
plt.imshow(img)
# To display a part of the image
plt.grid(False)
plt.axis('off')
plt.imshow(img_array[125:325,105:305])
Plotting multiple charts in a grid
sns.pairplot(flowers_df, hue='species'); # Pair plots with Seaborn
sns.pairplot(tips_df, hue='sex');
fig, axes = plt.subplots(2, 3, figsize=(16, 8))
# Use the axes for plotting
axes[0,0].plot(years, apples, 's-b')
axes[0,0].plot(years, oranges, 'o--r')
axes[0,0].set_xlabel('Year')
axes[0,0].set_ylabel('Yield (tons per hectare)')
axes[0,0].legend(['Apples', 'Oranges']);
axes[0,0].set_title('Crop Yields in Kanto')
# Pass the axes into seaborn
axes[0,1].set_title('Sepal Length vs. Sepal Width')
sns.scatterplot(x=flowers_df.sepal_length,
y=flowers_df.sepal_width,
hue=flowers_df.species,
s=100,
ax=axes[0,1]);
# Use the axes for plotting
axes[0,2].set_title('Distribution of Sepal Width')
axes[0,2].hist([setosa_df.sepal_width, versicolor_df.sepal_width, virginica_df.sepal_width],
bins=np.arange(2, 5, 0.25),
stacked=True);
axes[0,2].legend(['Setosa', 'Versicolor', 'Virginica']);
# Pass the axes into seaborn
axes[1,0].set_title('Restaurant bills')
sns.barplot(x='day', y='total_bill', hue='sex', data=tips_df, ax=axes[1,0]);
# Pass the axes into seaborn
axes[1,1].set_title('Flight traffic')
sns.heatmap(flights_df, cmap='Blues', ax=axes[1,1]);
# Plot an image using the axes
axes[1,2].set_title('Data Science Meme')
axes[1,2].imshow(img)
axes[1,2].grid(False)
axes[1,2].set_xticks([])
axes[1,2].set_yticks([])
plt.tight_layout(pad=2)