Give
loc
an item from the index.
In this
Series,
each item from the index is a two-letter
string.
Give
iloc
the same integer
(non-negative or negative)
that you would use to index a Python
list.
Since a two-letter
string
is a valid Python
identifier
(variable name),
you can abbreviate
series["CA"]
to
series.CA
with the
attribute
reference dot.
import sys
import pandas as pd
data = {
"AL": "Alabama",
"AK": "Alaska",
"AZ": "Arizona",
"AR": "Arkansas",
"CA": "California"
}
series = pd.Series(data = data, name = "state")
series.index.name = "abbr"
print(f'{series.loc["CA"] = }')
print(f'{series.iloc[4] = }')
print(f'{series.iloc[-1] = }')
print(f'{series["CA"] = }') #Same as series.loc["CA"], because "CA" is in the index.
print(f'{series.CA = }') #abbreviation for the above statement.
print(f'{series[4] = }') #Same as series.iloc[4], because 4 is not in the index.
print(f'{series[-1] = }') #Same as series.iloc[-1], because -1 is not in the index.
print()
series["CA"] = "Cali-for-ni-a"
print(series)
sys.exit(0)
series.loc["CA"] = 'California' series.iloc[4] = 'California' series.iloc[-1] = 'California' series["CA"] = 'California' series.CA = 'California' series[4] = 'California' series[-1] = 'California' abbr AL Alabama AK Alaska AZ Arizona AR Arkansas CA Cali-for-ni-a Name: state, dtype: object
import sys
import numpy as np
import pandas as pd
data = [
"Alabama",
"Alaska",
"Arizona",
"Arkansas",
"California"
]
index = pd.RangeIndex(1, len(data) + 1, name = "#")
series = pd.Series(data = data, index = index, name = "state")
print(series)
print()
print(f"{series.dtype.name = }")
print(f"{series.index.dtype.name = }")
print()
print(f"{series.loc[5] = }")
print(f"{series.iloc[4] = }")
print(f"{series.iloc[-1] = }")
print(f'{series[5] = }') #Same as series.loc[5], because 5 is in the index.
#Would raise a KeyError exception because the index is an index of integers:
#print(f'{series.loc[-1] = }')
#print(f'{series[-1] = }')
sys.exit(0)
# 1 Alabama 2 Alaska 3 Arizona 4 Arkansas 5 California Name: state, dtype: object series.dtype.name = 'object' series.index.dtype.name = 'int64' series.loc[5] = 'California' series.iloc[4] = 'California' series.iloc[-1] = 'California' series[5] = 'California'
import sys
import pandas as pd
#Can't store this data into a dict, because there are more than one "LOL".
data = [
["IMHO", "in my humble opinion"],
["LOL", "laughing out loud"],
["LOL", "lots of luck"],
["OMG", "oh my God"],
["ROFL", "rolling on the floor laughing"]
]
index = pd.Index( data = [datum[0] for datum in data], name = "abbr")
series = pd.Series(data = [datum[1] for datum in data], index = index, name = "Texting Abbreviations")
print(series)
print()
print(f'{series["ROFL"] = }') #series["ROFL"] is a str
print()
print(series["LOL"]) #series["LOL"] is a pd.Series
sys.exit(0)
abbr IMHO in my humble opinion LOL laughing out loud LOL lots of luck OMG oh my God ROFL rolling on the floor laughing Name: Texting Abbreviations, dtype: object series["ROFL"] = 'rolling on the floor laughing' LOL laughing out loud LOL lots of luck Name: Texting Abbreviations, dtype: object
Select as many items as you want, in any order.
import sys
import pandas as pd
data = {
"CT": "Connecticut",
"NJ": "New Jersey",
"NY": "New York"
}
series = pd.Series(data = data, name = "state")
series.index.name = "abbr"
print(series)
print()
print(series.loc[["NY", "NJ", "CT"]])
print()
print(series.iloc[[2, 1, 0]])
print()
print(series[["NY", "NJ", "NJ"]]) #same as series.loc[["NY", "NJ", "NJ"]]
print()
sys.exit(0)
abbr CT Connecticut NJ New Jersey NY New York Name: state, dtype: object abbr NY New York NJ New Jersey CT Connecticut Name: state, dtype: object abbr NY New York NJ New Jersey CT Connecticut Name: state, dtype: object abbr NY New York NJ New Jersey NJ New Jersey Name: state, dtype: object
import sys
import pandas as pd
data = {
"AL": "Alabama",
"AK": "Alaska",
"AZ": "Arizona",
"AR": "Arkansas",
"CA": "California"
}
series = pd.Series(data = data, name = "state")
series.index.name = "abbr"
print('series["AK": "AR"]') #Warning: the slice is inclusive. It contains 3 states.
print( series["AK": "AR"]) #could also say print(series.loc["AK": "AR"])
print()
print('series[1:3]') #Warning: the slice is exclusive. It contains only 2 states.
print( series[1:3])
print()
print('series[["AK", "AR"]]') #["AK", "AR"] is a Python list.
print( series[["AK", "AR"]]) #could also say print(series.loc[["AK", "AR"]])
print()
print('series[["AK", "AR", "AZ"]]')
print( series[["AK", "AR", "AZ"]]) #could also say print(series.loc[["AK", "AR", "AZ"]])
print()
print("series.iloc[1:3]") #Warning: the slice is exclusive. It contains only 2 states.
print( series.iloc[1:3])
print()
print("series.iloc[[1, 3]]")
print( series.iloc[[1, 3]])
print()
print("series.iloc[[1, 3, 2]]")
print( series.iloc[[1, 3, 2]])
sys.exit(0)
series["AK": "AR"] abbr AK Alaska AZ Arizona AR Arkansas Name: state, dtype: object series[1:3] abbr AK Alaska AZ Arizona Name: state, dtype: object series[["AK", "AR"]] abbr AK Alaska AR Arkansas Name: state, dtype: object series[["AK", "AR", "AZ"]] abbr AK Alaska AR Arkansas AZ Arizona Name: state, dtype: object series.iloc[1:3] abbr AK Alaska AZ Arizona Name: state, dtype: object series.iloc[[1, 3]] abbr AK Alaska AR Arkansas Name: state, dtype: object series.iloc[[1, 3, 2]] abbr AK Alaska AR Arkansas AZ Arizona Name: state, dtype: object
I didn’t bother to specify an index, so the index defaults to 0, 1, 2, 3, 4, 5, 6, 7.
import sys
import pandas as pd
data = [
"A word that means the world to me.", #0
"O means only that she's growing old.", #1
"T is for the tears she shed to save me.", #2
"E is for her eyes with love-light shining.", #3
"M is for the million things she gave me.", #4
"R means right and right she'll always be.", #5
"Put them all together, they spell mother,", #6
"H is for her heart of purest gold." #7
]
series = pd.Series(data = data, name = "poem")
#To unscramble, you will have to write something like the following.
print(series.iloc[[1, 0, 2, 7, 5, 6, 3, 4]])
print()
#Left-justify the lines.
s = series.to_string(index = False) #s is one big string
lines = [line.lstrip() for line in s.splitlines()] #lines is a list of smaller strings
s = "\n".join(lines) #rejoin them back into one big string
print(s)
sys.exit(0)
import sys
import pandas as pd
data = [
"A word that means the world to me.", #0
"O means only that she's growing old.", #1
"T is for the tears she shed to save me.", #2
"E is for her eyes with love-light shining.", #3
"M is for the million things she gave me.", #4
"R means right and right she'll always be.", #5
"Put them all together, they spell mother,", #6
"H is for her heart of purest gold." #7
]
series = pd.Series(data = data, name = "poem")
print(series)
print()
data = [7, 1, 2, 4, 0, 5, 6, 3]
index = pd.Index(data = data)
series.index = index #Give the Series a new index.
print(series)
print()
series.sort_index(inplace = True) #or series = series.sort_index()
print(series)
sys.exit(0)
0 A word that means the world to me. 1 O means only that she's growing old. 2 T is for the tears she shed to save me. 3 E is for her eyes with love-light shining. 4 M is for the million things she gave me. 5 R means right and right she'll always be. 6 Put them all together, they spell mother, 7 H is for her heart of purest gold. Name: poem, dtype: object 7 A word that means the world to me. 1 O means only that she's growing old. 2 T is for the tears she shed to save me. 4 E is for her eyes with love-light shining. 0 M is for the million things she gave me. 5 R means right and right she'll always be. 6 Put them all together, they spell mother, 3 H is for her heart of purest gold. Name: poem, dtype: object 0 M is for the million things she gave me. 1 O means only that she's growing old. 2 T is for the tears she shed to save me. 3 H is for her heart of purest gold. 4 E is for her eyes with love-light shining. 5 R means right and right she'll always be. 6 Put them all together, they spell mother, 7 A word that means the world to me. Name: poem, dtype: object
"Keep only the rows with a correctly spelled index."
import sys
import pandas as pd
data = [
["Brooklyn", "Brad"],
["Manhattan", "Mark"],
["Queens", "Quentin"],
["Staten Island", "Stanley"],
["Staten Island", "Stanislav"],
["Manhattan", "Mary"],
["Broklyn", "Bill"], #misspelled
["Manhattan", "Manfred"]
]
valids = [
"Bronx",
"Brooklyn",
"Manhattan",
"Queens",
"Staten Island"
]
index = pd.Index( data = [person[0] for person in data], name = "borough")
series = pd.Series(data = [person[1] for person in data], index = index, name = "People")
print(series)
print()
indexOfInvalids = series.index.difference(valids)
print(f"{indexOfInvalids = }")
print()
series.drop(indexOfInvalids, inplace = True)
print(series)
sys.exit(0)
borough Brooklyn Brad Manhattan Mark Queens Quentin Staten Island Stanley Staten Island Stanislav Manhattan Mary Broklyn Bill Manhattan Manfred Name: People, dtype: object indexOfInvalids = Index(['Broklyn'], dtype='object', name='borough') borough Brooklyn Brad Manhattan Mark Queens Quentin Staten Island Stanley Staten Island Stanislav Manhattan Mary Manhattan Manfred Name: People, dtype: object