{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# CS 237 Spring 2021, HW 10\n",
"\n",
"#### Due date: Thursday April 15th at Midnight (1 minute after 11:59pm on 4/8) via Gradescope (with a 6 hour grace period)\n",
"\n",
" Late policy: You may submit the homework up to 24 hours late for a 10% penalty. Hence, the late deadline is Friday 4/16 at Midnight (with a 6 hour grace period). \n",
"\n",
"#### General Instructions\n",
"\n",
"Please complete this notebook by filling in solutions where indicated. \n",
"\n",
"For full credit, please take careful note of the following requirements:\n",
"\n",
"- Do NOT use any HTML tags in your notebook, as Gradescope will ignore them;\n",
"\n",
"- Do NOT answer questions by including images, as Gradescope will ignore them; and \n",
"\n",
"- You MUST \"Restart and Run All\" from the Kernel menu before submitting to Gradescope.\n",
"\n",
"**Any assignments which do not follow these requirements will not receive full credit.** \n",
"\n",
"\n",
"\n",
"There are 8 problems on this homework (each worth 7.5 points), 5 analytical and 3 concerning Pandas, a library for managing data files. Pandas is used extensively in data science as part of the \"data wrangling\" phase of a machine learning project. \n",
"\n",
"The problems for Lab 10 are problems 1 through 3, and the remaining problems are the analytical problems. \n",
"\n",
"NOTE (added Sunday am): I have cancelled problems 8 and 10, mistake in editing, I did not mean to include them.\n"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"# Here are some imports which will be used in code that we write for CS 237\n",
"\n",
"import matplotlib.pyplot as plt # normal plotting\n",
"import numpy as np\n",
"\n",
"from math import log, pi, log, floor, ceil, sqrt # import whatever you want from math\n",
"from random import seed, random\n",
"from collections import Counter\n",
"\n",
"%matplotlib inline\n",
"\n",
"from scipy.special import comb\n",
" \n",
"def C(N,K): \n",
" return comb(N,K,True) # just a wrapper around the scipy function\n",
"\n",
"\n",
"# Here are the basic statistical functions we will use from numpy\n",
"\n",
"from numpy import mean, var, std, median\n",
"\n",
"L = [2,4,3,6,4,5]\n",
"\n",
"# mean value\n",
"\n",
"mean(L) \n",
"\n",
"\n",
"# Variance\n",
"# ddof = delta degrees of freedom, default is 0\n",
"\n",
"# population variance\n",
"var(L) \n",
"\n",
"# sample variance\n",
"var(L,ddof=1)\n",
"\n",
"# Standard deviation\n",
"# ddof = delta degrees of freedom, default is 0\n",
"\n",
"# population standard deviation\n",
"std(L) \n",
"\n",
"# sample standard deviation\n",
"std(L,ddof=1) \n",
"\n",
"# Median\n",
"\n",
"median(L) \n",
"\n",
"# Random sampling of `size` elements from list with or without replacement\n",
"\n",
"np.random.choice(L,size=1,replace=True)\n",
" \n",
"# Scipy statistical functions\n",
"\n",
"from scipy.stats import norm, binom, expon, geom, poisson, gamma, nbinom, bernoulli \n",
"\n",
"# https://docs.scipy.org/doc/scipy/reference/stats.html\n",
"\n",
"#### Normal Distribution #####\n",
"\n",
"###### Note that in this library loc = mean and scale = standard deviation #####\n",
"\n",
"# Examples assume random variable X (e.g., housing prices) normally distributed with mu = 60, sigma = 10\n",
"\n",
"# Probability Density Function (really only useful for drawing the curve)\n",
"# f(x) = P(X == x)\n",
"\n",
"norm.pdf(x=50,loc=60, scale= 10) \n",
"\n",
"# Cumulative Density Function\n",
"# F(x) = P(X < x)\n",
"\n",
"# Example: Percentage of houses less than 50K. \n",
"norm.cdf(x=50,loc=60,scale=10) \n",
"\n",
"# Example: Find P(60 x)\n",
"\n",
"# Example: Percentage of houses more than 50K.\n",
"norm.sf(x=50,loc=60,scale=10) \n",
"\n",
"# Percentage Point Function: Inverse of the CDF:\n",
"# For what is the largest value of k for which P( X < k ) = q ?\n",
"\n",
"# Example: What is the maximum cost of the 5% cheapest houses, \n",
"# i.e., the x such that P(X < x) = 0.05?\n",
"\n",
"norm.ppf(q=0.05,loc=60,scale=40)\n",
"\n",
"# Inverse Survival Function: Inverse (1 - CDF):\n",
"# For what is the smallest value of k for which P( X > k ) = q ?\n",
"\n",
"# Example: What is the minimum cost of the 5% most expensive houses, \n",
"# i.e., the x such that P(X > x) = 0.05?\n",
"\n",
"norm.isf(q=0.05,loc=60,scale=40)\n",
"\n",
"# Give the endpoints of the interval (centered on the mean)\n",
"# which contain alpha/100 percent of the population (alpha is a probability)\n",
"\n",
"# Ex. Give the interval for the middle 75% of the houses\n",
"\n",
"norm.interval(alpha=0.75, loc=60, scale=40)\n",
"\n",
"# generate a random variate\n",
"norm.rvs(loc=60, scale=40)\n",
"\n",
"# generate random variates, returns list of length = size\n",
"norm.rvs(loc=60, scale=40, size=10)\n",
"\n",
"\n",
"\n",
"\n",
"##### Bernoulli Distribution X ~ Bernoulli(p) ####\n",
"\n",
"# p = probability of success for Bernoulli trial\n",
"\n",
"# Generate a random variate\n",
"bernoulli.rvs(p=0.5)\n",
"\n",
"# Generate a list of random variates\n",
"bernoulli.rvs(p=0.5,size=100)\n",
"\n",
"##### Binomial Distribution X ~ B(n,p) ####\n",
"\n",
"# n = number of independent Bernoulli trials\n",
"# p = probability of success for Bernoulli trial\n",
"# k = outcome in range [0 .. n]\n",
"\n",
"# Generate a random variate\n",
"binom.rvs(n=10, p=0.5)\n",
"\n",
"# Generate a list of random variates\n",
"binom.rvs(n=10, p=0.5,size=100)\n",
"\n",
"# Probability mass function.\n",
"binom.pmf(k=4, n=10, p=0.5)\n",
"\n",
"# Cumulative distribution function\n",
"binom.cdf(k=4, n=10, p=0.5)\n",
"\n",
"print()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Pandas Data Management and Analysis Library\n",
"\n",
"In this lab, we will learn about the Pandas library for data analysis. Pandas is part of the Anaconda distribution, and there are decent tutorials on most aspects of Pandas; I would recommend the following:\n",
"\n",
"Pandas Tutorial: http://pandas.pydata.org/pandas-docs/stable/tutorials.html\n",
"\n",
"Basic functionality: http://pandas.pydata.org/pandas-docs/stable/basics.html\n",
"\n",
"Indexing and selecting data: http://pandas.pydata.org/pandas-docs/stable/indexing.html\n",
"\n",
"There are three problems, the first an extended tutorial, and the last two actual problems. You should read through problem 1 carefully and try all the examples. Problems 2 and 3 are more realistic activities based on the material in problem 1.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Problem One: Basic Pandas data manipulation\n",
"\n",
"We will first learn how to read data sets from a text or CSV (\"Comma Separated Values\") file, understand the DataFrame data structure, and learn how to extract data from a dataframe; then we will understand how to select rows and columns from a table, and to apply various functions to tables; finally, we will learn how to display histograms of the data. There is a LOT of complexity in all these various aspects of Pandas, but we will learn a \"novice subset\" of the most important features.\n",
"\n",
"Note: There is a lot of reading and thinking to do in this first problem; when we want you to do something on your own computer, we will indicate it with a **TODO**. All you really have to do for this problem is to try various things\n",
"in Pandas. Do not skip ahead, however, you need to practice these before going on!"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Data as Tables\n",
"\n",
"The basic form of the data sets manipulated in Pandas (and indeed in all modern database systems) is a 2D table of data with rows and columns; for example, here is a data set we will use as an example:"
]
},
{
"attachments": {
"Screen%20Shot%202021-04-08%20at%2011.39.21%20PM.png": {
"image/png": "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"
}
},
"cell_type": "markdown",
"metadata": {},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This is a hypothetical list of 10 students at BU; each row is an individual student, and each column gives a specific piece of information about that student. Note that each column contains the same kind of data (i.e., the first three columns have strings, and the remaining are floats--we will assume that all numberic data is represented as floating-point), and each column has a header giving a description of the information in that column. Column headers are not absolutely necessary, but we will make this assumption for now.\n",
"\n",
"**Note on terminology**: In Pandas, a table is called a **dataframe**; in databases we often call the rows in a table records, the columns are called fields, and the headers are then called field names. This terminology is sometimes used in data analysis."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Comma Separated Value (CSV) Files\n",
"\n",
"A common file format for data stored as text is a CSV file, in which each row is stored on a separate line, with commas between all the fields. The table above would look like this if I opened it with Emacs on my Mac (there would be a newline \\n at the end of each line; on a Windows machine there would be \\r\\n):"
]
},
{
"attachments": {
"Screen%20Shot%202021-04-08%20at%2011.34.40%20PM.png": {
"image/png": "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"
}
},
"cell_type": "markdown",
"metadata": {},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This is a file format supported by Excel as well, and you can import it into an Excel spreadsheet or save a spreadsheet as a CSV file:"
]
},
{
"attachments": {
"Screen%20Shot%202021-04-08%20at%2011.33.56%20PM.png": {
"image/png": "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"
}
},
"cell_type": "markdown",
"metadata": {},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"If you want to create such a file, the easiest way is probably to start with an Excel spreadsheet and simply save it in .csv format!"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Reading and Writing CSV Files in Pandas\n",
"\n",
"To manipulate such data tables in Python, the best library is Pandas, which you should import as follows:"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd # this is already done in the first cell at the top of this notebook"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Here is an example of reading the csv file `lab10.students.csv` from the 237 data repository using a URL\n",
"and storing the data (called a \"data frame\") in a variable `students`. \n",
"\n",
"You can also read a file from your current local directory (which we demonstrate a few cells below). "
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"scrolled": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Userid
\n",
"
Gender
\n",
"
ClassYear
\n",
"
GPA
\n",
"
Credits
\n",
"
Transfer
\n",
"
AP
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
biden
\n",
"
M
\n",
"
U2
\n",
"
4.000
\n",
"
33
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
1
\n",
"
harris
\n",
"
F
\n",
"
U4
\n",
"
3.630
\n",
"
45
\n",
"
71.1
\n",
"
8
\n",
"
\n",
"
\n",
"
2
\n",
"
blinken
\n",
"
M
\n",
"
U4
\n",
"
3.380
\n",
"
113
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
3
\n",
"
emhoff
\n",
"
M
\n",
"
U2
\n",
"
3.210
\n",
"
28
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
4
\n",
"
yellen
\n",
"
F
\n",
"
U2
\n",
"
3.810
\n",
"
33
\n",
"
0.0
\n",
"
4
\n",
"
\n",
"
\n",
"
5
\n",
"
austin
\n",
"
M
\n",
"
U3
\n",
"
3.034
\n",
"
56
\n",
"
28.0
\n",
"
0
\n",
"
\n",
"
\n",
"
6
\n",
"
garland
\n",
"
M
\n",
"
U4
\n",
"
3.200
\n",
"
80
\n",
"
44.0
\n",
"
24
\n",
"
\n",
"
\n",
"
7
\n",
"
haaland
\n",
"
F
\n",
"
U4
\n",
"
3.580
\n",
"
66
\n",
"
30.0
\n",
"
0
\n",
"
\n",
"
\n",
"
8
\n",
"
trump
\n",
"
M
\n",
"
U3
\n",
"
1.230
\n",
"
32
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
9
\n",
"
pence
\n",
"
M
\n",
"
U4
\n",
"
2.040
\n",
"
106
\n",
"
3.0
\n",
"
24
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Userid Gender ClassYear GPA Credits Transfer AP\n",
"0 biden M U2 4.000 33 0.0 12\n",
"1 harris F U4 3.630 45 71.1 8\n",
"2 blinken M U4 3.380 113 0.0 12\n",
"3 emhoff M U2 3.210 28 0.0 0\n",
"4 yellen F U2 3.810 33 0.0 4\n",
"5 austin M U3 3.034 56 28.0 0\n",
"6 garland M U4 3.200 80 44.0 24\n",
"7 haaland F U4 3.580 66 30.0 0\n",
"8 trump M U3 1.230 32 0.0 0\n",
"9 pence M U4 2.040 106 3.0 24"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"students = pd.read_csv(\"https://cs-web.bu.edu/fac/snyder/cs237/Data/lab10.students.csv\")\n",
"students"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"If you ask for the value of the dataframe to be printed out, it will print it out in ASCII:"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" Userid Gender ClassYear GPA Credits Transfer AP\n",
"0 biden M U2 4.000 33 0.0 12\n",
"1 harris F U4 3.630 45 71.1 8\n",
"2 blinken M U4 3.380 113 0.0 12\n",
"3 emhoff M U2 3.210 28 0.0 0\n",
"4 yellen F U2 3.810 33 0.0 4\n",
"5 austin M U3 3.034 56 28.0 0\n",
"6 garland M U4 3.200 80 44.0 24\n",
"7 haaland F U4 3.580 66 30.0 0\n",
"8 trump M U3 1.230 32 0.0 0\n",
"9 pence M U4 2.040 106 3.0 24\n"
]
}
],
"source": [
"print(students)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Although we won't need to do this right now, for reference, you can also write out a dataframe to a csv file; the default is to write out the index numbers on each row--in general you want to avoid this. The following command will create an identical file to the one read in, without the pesky index numbers:"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"students.to_csv('temp.csv', encoding='utf-8', index=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Reading in a local file is the same as reading in a file from a URL:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Userid
\n",
"
Gender
\n",
"
ClassYear
\n",
"
GPA
\n",
"
Credits
\n",
"
Transfer
\n",
"
AP
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
biden
\n",
"
M
\n",
"
U2
\n",
"
4.000
\n",
"
33
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
1
\n",
"
harris
\n",
"
F
\n",
"
U4
\n",
"
3.630
\n",
"
45
\n",
"
71.1
\n",
"
8
\n",
"
\n",
"
\n",
"
2
\n",
"
blinken
\n",
"
M
\n",
"
U4
\n",
"
3.380
\n",
"
113
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
3
\n",
"
emhoff
\n",
"
M
\n",
"
U2
\n",
"
3.210
\n",
"
28
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
4
\n",
"
yellen
\n",
"
F
\n",
"
U2
\n",
"
3.810
\n",
"
33
\n",
"
0.0
\n",
"
4
\n",
"
\n",
"
\n",
"
5
\n",
"
austin
\n",
"
M
\n",
"
U3
\n",
"
3.034
\n",
"
56
\n",
"
28.0
\n",
"
0
\n",
"
\n",
"
\n",
"
6
\n",
"
garland
\n",
"
M
\n",
"
U4
\n",
"
3.200
\n",
"
80
\n",
"
44.0
\n",
"
24
\n",
"
\n",
"
\n",
"
7
\n",
"
haaland
\n",
"
F
\n",
"
U4
\n",
"
3.580
\n",
"
66
\n",
"
30.0
\n",
"
0
\n",
"
\n",
"
\n",
"
8
\n",
"
trump
\n",
"
M
\n",
"
U3
\n",
"
1.230
\n",
"
32
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
9
\n",
"
pence
\n",
"
M
\n",
"
U4
\n",
"
2.040
\n",
"
106
\n",
"
3.0
\n",
"
24
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Userid Gender ClassYear GPA Credits Transfer AP\n",
"0 biden M U2 4.000 33 0.0 12\n",
"1 harris F U4 3.630 45 71.1 8\n",
"2 blinken M U4 3.380 113 0.0 12\n",
"3 emhoff M U2 3.210 28 0.0 0\n",
"4 yellen F U2 3.810 33 0.0 4\n",
"5 austin M U3 3.034 56 28.0 0\n",
"6 garland M U4 3.200 80 44.0 24\n",
"7 haaland F U4 3.580 66 30.0 0\n",
"8 trump M U3 1.230 32 0.0 0\n",
"9 pence M U4 2.040 106 3.0 24"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"pd.read_csv('temp.csv')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Selecting Rows and Columns from the Dataframe using Slices\n",
"\n",
"Pandas gives you a truly bewildering variety of ways to manipulate dataframes, but first we will only need to think about the basics: selecting rows and columns from the table. This amounts to either selecting rows from the table, or selecting columns from the table (or both).\n",
"\n",
"The basic ideas here is that the dataframe is a two-dimensional matrix, where the rows are indexed by numbers and the columns are indexed by column headers, so that\n",
"\n",
"> rows are selecting by using normal Python array slices, e.g., `[0:3]`\n",
"\n",
"and\n",
"\n",
"> columns are selecting by using a list of column headers, e.g., `[['Userid','GPA']]`\n",
"\n",
"The double brackets are not a typo! The outer brackets enclose the parameter, which is a list of header names. If there is only a single header name, then you can use a single set of brackets, e.g., `['Userid']`."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### TODO\n",
"\n",
"For the following, first try to predict what will happen, and then try it to confirm your understanding:\n",
"\n",
"- students[0:3]\n",
"- students[5:]\n",
"- students[-3:]\n",
"- students[:]\n",
"- students[1:7:2]\n",
"- students[::-1]"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Userid
\n",
"
Gender
\n",
"
ClassYear
\n",
"
GPA
\n",
"
Credits
\n",
"
Transfer
\n",
"
AP
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
biden
\n",
"
M
\n",
"
U2
\n",
"
4.00
\n",
"
33
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
1
\n",
"
harris
\n",
"
F
\n",
"
U4
\n",
"
3.63
\n",
"
45
\n",
"
71.1
\n",
"
8
\n",
"
\n",
"
\n",
"
2
\n",
"
blinken
\n",
"
M
\n",
"
U4
\n",
"
3.38
\n",
"
113
\n",
"
0.0
\n",
"
12
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Userid Gender ClassYear GPA Credits Transfer AP\n",
"0 biden M U2 4.00 33 0.0 12\n",
"1 harris F U4 3.63 45 71.1 8\n",
"2 blinken M U4 3.38 113 0.0 12"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"students[0:3]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"So the point here is that selecting rows is really the same as slicing a list.\n",
"\n",
"As explained above, you select columns by giving a list of the column headers in a list (with double square brackets). "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### TODO\n",
"\n",
"For the following, first try to predict what will happen, and then try it to confirm your understanding:\n",
"- students[ ['Userid', 'GPA'] ]\n",
"- students[ ['GPA', 'Gender'] ]\n",
"- students[ ['Credits'] ]\n",
"- students['Credits']\n",
"- students[ ['GPA', 'Userid', 'GPA'] ]"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
GPA
\n",
"
Userid
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
4.000
\n",
"
biden
\n",
"
\n",
"
\n",
"
1
\n",
"
3.630
\n",
"
harris
\n",
"
\n",
"
\n",
"
2
\n",
"
3.380
\n",
"
blinken
\n",
"
\n",
"
\n",
"
3
\n",
"
3.210
\n",
"
emhoff
\n",
"
\n",
"
\n",
"
4
\n",
"
3.810
\n",
"
yellen
\n",
"
\n",
"
\n",
"
5
\n",
"
3.034
\n",
"
austin
\n",
"
\n",
"
\n",
"
6
\n",
"
3.200
\n",
"
garland
\n",
"
\n",
"
\n",
"
7
\n",
"
3.580
\n",
"
haaland
\n",
"
\n",
"
\n",
"
8
\n",
"
1.230
\n",
"
trump
\n",
"
\n",
"
\n",
"
9
\n",
"
2.040
\n",
"
pence
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" GPA Userid\n",
"0 4.000 biden\n",
"1 3.630 harris\n",
"2 3.380 blinken\n",
"3 3.210 emhoff\n",
"4 3.810 yellen\n",
"5 3.034 austin\n",
"6 3.200 garland\n",
"7 3.580 haaland\n",
"8 1.230 trump\n",
"9 2.040 pence"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"students[['GPA','Userid']]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"You can also combine these two, to get a slice of rows and a selection of columns."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### TODO\n",
"\n",
"For the following, first try to predict what will happen, and then try it to confirm your understanding:\n",
"- students[0:3][['GPA','Userid']]\n",
"- students[::-1][['AP','Gender','Credits']]\n",
"- students['GPA'][2:7]\n",
"- students[2:7][['GPA']]\n"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
GPA
\n",
"
Userid
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
4.00
\n",
"
biden
\n",
"
\n",
"
\n",
"
1
\n",
"
3.63
\n",
"
harris
\n",
"
\n",
"
\n",
"
2
\n",
"
3.38
\n",
"
blinken
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" GPA Userid\n",
"0 4.00 biden\n",
"1 3.63 harris\n",
"2 3.38 blinken"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"students[0:3][['GPA','Userid']]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"Note carefully the last two examples: it does not matter what order you put the selectors in.\n",
"\n",
"Finally, you would expect that you could do a slice on column names, like this:\n",
"\n",
" students[ ['GPA' : 'AP'] ] # error!\n",
"\n",
"Nope! In order to do this, you need to use the `loc` function, and give it two slices separated by a comma.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### TODO\n",
"\n",
"For the following, first try to predict what will happen, and then try it to confirm your understanding:\n",
"\n",
"- students.loc[ 0:3 , 'GPA':'AP' ]\n",
"- students.loc[:, 'Userid':'GPA']\n",
"- students.loc[:5, 'ClassYear': ]\n",
"- students.loc[2:7, 'Userid':'Transfer':2]\n",
"- students.loc[::-1, ::-1]\n"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"a = students.loc[ 0:3 , 'GPA':'AP' ]"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
],
"text/plain": [
" AP Transfer Credits GPA ClassYear Gender Userid\n",
"9 24 3.0 106 2.040 U4 M pence\n",
"8 0 0.0 32 1.230 U3 M trump\n",
"7 0 30.0 66 3.580 U4 F haaland\n",
"6 24 44.0 80 3.200 U4 M garland\n",
"5 0 28.0 56 3.034 U3 M austin\n",
"4 4 0.0 33 3.810 U2 F yellen\n",
"3 0 0.0 28 3.210 U2 M emhoff\n",
"2 12 0.0 113 3.380 U4 M blinken\n",
"1 8 71.1 45 3.630 U4 F harris\n",
"0 12 0.0 33 4.000 U2 M biden"
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"students.loc[::-1, ::-1]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Selecting Rows and Columns from the Dataframe using Boolean Expressions\n",
"\n",
"Pandas gives you lots of ways of selecting data, and a particularly useful way of selecting rows is to specify a boolean expression that the row values must satisfy. For example, to select only those rows representing men, we could do this:"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"scrolled": true
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Userid
\n",
"
Gender
\n",
"
ClassYear
\n",
"
GPA
\n",
"
Credits
\n",
"
Transfer
\n",
"
AP
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
biden
\n",
"
M
\n",
"
U2
\n",
"
4.000
\n",
"
33
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
2
\n",
"
blinken
\n",
"
M
\n",
"
U4
\n",
"
3.380
\n",
"
113
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
3
\n",
"
emhoff
\n",
"
M
\n",
"
U2
\n",
"
3.210
\n",
"
28
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
5
\n",
"
austin
\n",
"
M
\n",
"
U3
\n",
"
3.034
\n",
"
56
\n",
"
28.0
\n",
"
0
\n",
"
\n",
"
\n",
"
6
\n",
"
garland
\n",
"
M
\n",
"
U4
\n",
"
3.200
\n",
"
80
\n",
"
44.0
\n",
"
24
\n",
"
\n",
"
\n",
"
8
\n",
"
trump
\n",
"
M
\n",
"
U3
\n",
"
1.230
\n",
"
32
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
9
\n",
"
pence
\n",
"
M
\n",
"
U4
\n",
"
2.040
\n",
"
106
\n",
"
3.0
\n",
"
24
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Userid Gender ClassYear GPA Credits Transfer AP\n",
"0 biden M U2 4.000 33 0.0 12\n",
"2 blinken M U4 3.380 113 0.0 12\n",
"3 emhoff M U2 3.210 28 0.0 0\n",
"5 austin M U3 3.034 56 28.0 0\n",
"6 garland M U4 3.200 80 44.0 24\n",
"8 trump M U3 1.230 32 0.0 0\n",
"9 pence M U4 2.040 106 3.0 24"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"men = students[ students['Gender'] == 'M' ]\n",
"men"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"and here we select the rows where the Credits are less than the Transfer:"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"scrolled": true
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Userid
\n",
"
Gender
\n",
"
ClassYear
\n",
"
GPA
\n",
"
Credits
\n",
"
Transfer
\n",
"
AP
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
harris
\n",
"
F
\n",
"
U4
\n",
"
3.63
\n",
"
45
\n",
"
71.1
\n",
"
8
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Userid Gender ClassYear GPA Credits Transfer AP\n",
"1 harris F U4 3.63 45 71.1 8"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"lessCredits = students[students['Credits'] < students['Transfer']]\n",
"lessCredits"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### TODO\n",
"\n",
"For the following, first try to predict what will happen, and then try it to confirm your understanding:\n",
"- students[ students['Userid'] >= 'haaland' ]\n",
"- students[ students['AP'] == 0 ]\n",
"- students[ students['GPA'] < 3.5 ]"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Userid
\n",
"
Gender
\n",
"
ClassYear
\n",
"
GPA
\n",
"
Credits
\n",
"
Transfer
\n",
"
AP
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1
\n",
"
harris
\n",
"
F
\n",
"
U4
\n",
"
3.63
\n",
"
45
\n",
"
71.1
\n",
"
8
\n",
"
\n",
"
\n",
"
4
\n",
"
yellen
\n",
"
F
\n",
"
U2
\n",
"
3.81
\n",
"
33
\n",
"
0.0
\n",
"
4
\n",
"
\n",
"
\n",
"
7
\n",
"
haaland
\n",
"
F
\n",
"
U4
\n",
"
3.58
\n",
"
66
\n",
"
30.0
\n",
"
0
\n",
"
\n",
"
\n",
"
8
\n",
"
trump
\n",
"
M
\n",
"
U3
\n",
"
1.23
\n",
"
32
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
9
\n",
"
pence
\n",
"
M
\n",
"
U4
\n",
"
2.04
\n",
"
106
\n",
"
3.0
\n",
"
24
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Userid Gender ClassYear GPA Credits Transfer AP\n",
"1 harris F U4 3.63 45 71.1 8\n",
"4 yellen F U2 3.81 33 0.0 4\n",
"7 haaland F U4 3.58 66 30.0 0\n",
"8 trump M U3 1.23 32 0.0 0\n",
"9 pence M U4 2.04 106 3.0 24"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"students[ students['Userid'] >= 'haaland' ]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"To do compound boolean expressions when selecting rows, you have to enclose the expressions in parentheses and use the \"bitwise\" boolean operations `~`, `&`, `|` (instead of the normal Python `not`, `and`, `or`); here is an example:"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"scrolled": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Userid
\n",
"
Gender
\n",
"
ClassYear
\n",
"
GPA
\n",
"
Credits
\n",
"
Transfer
\n",
"
AP
\n",
"
\n",
" \n",
" \n",
"
\n",
"
2
\n",
"
blinken
\n",
"
M
\n",
"
U4
\n",
"
3.380
\n",
"
113
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
3
\n",
"
emhoff
\n",
"
M
\n",
"
U2
\n",
"
3.210
\n",
"
28
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
5
\n",
"
austin
\n",
"
M
\n",
"
U3
\n",
"
3.034
\n",
"
56
\n",
"
28.0
\n",
"
0
\n",
"
\n",
"
\n",
"
6
\n",
"
garland
\n",
"
M
\n",
"
U4
\n",
"
3.200
\n",
"
80
\n",
"
44.0
\n",
"
24
\n",
"
\n",
"
\n",
"
8
\n",
"
trump
\n",
"
M
\n",
"
U3
\n",
"
1.230
\n",
"
32
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
9
\n",
"
pence
\n",
"
M
\n",
"
U4
\n",
"
2.040
\n",
"
106
\n",
"
3.0
\n",
"
24
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Userid Gender ClassYear GPA Credits Transfer AP\n",
"2 blinken M U4 3.380 113 0.0 12\n",
"3 emhoff M U2 3.210 28 0.0 0\n",
"5 austin M U3 3.034 56 28.0 0\n",
"6 garland M U4 3.200 80 44.0 24\n",
"8 trump M U3 1.230 32 0.0 0\n",
"9 pence M U4 2.040 106 3.0 24"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"menGPA = students[(students['Gender'] == 'M') & (students['GPA'] < 3.5 )]\n",
"menGPA"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Turning a column into a list\n",
"\n",
"Sometimes you simply want to grab one particular column (say the GPA) into a list, so that you\n",
"can manipulate it in Python. This is easy, as you just have to convert a single-column frame into a list:"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[4.0, 3.63, 3.38, 3.21, 3.81, 3.034, 3.2, 3.58, 1.23, 2.04]"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"G = list(students['GPA'])\n",
"G"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Converting a whole data frame into a list of lists\n",
"\n",
"You can convert the entire data set into Python lists as follows. "
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[['biden', 'M', 'U2', 4.0, 33, 0.0, 12],\n",
" ['harris', 'F', 'U4', 3.63, 45, 71.1, 8],\n",
" ['blinken', 'M', 'U4', 3.38, 113, 0.0, 12],\n",
" ['emhoff', 'M', 'U2', 3.21, 28, 0.0, 0],\n",
" ['yellen', 'F', 'U2', 3.81, 33, 0.0, 4],\n",
" ['austin', 'M', 'U3', 3.034, 56, 28.0, 0],\n",
" ['garland', 'M', 'U4', 3.2, 80, 44.0, 24],\n",
" ['haaland', 'F', 'U4', 3.58, 66, 30.0, 0],\n",
" ['trump', 'M', 'U3', 1.23, 32, 0.0, 0],\n",
" ['pence', 'M', 'U4', 2.04, 106, 3.0, 24]]"
]
},
"execution_count": 18,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"students.values.tolist()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"You can slice this to get only particular rows, or a single row:"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['harris', 'F', 'U4', 3.63, 45, 71.1, 8]"
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"(students[1:2].values.tolist())[0]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Saving your work to a file\n",
"\n",
"Note that each of these expressions we have explored returns a new dataframe, so that if you wanted to create a new data set derived from an existing set, you could assign an expression to a variable and write it out, e.g.,"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [],
"source": [
"studrev = students.loc[::-1, ::-1]\n",
"studrev.to_csv(\"studentrev.csv\", index=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"or just combine without bothering with the variable:"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
"students.loc[::-1, ::-1].to_csv(\"studentrev.csv\", index=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Both of these would put the file studentrev.csv into my work directory, which I could then import into Excel"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Functions on DataFrames\n",
"\n",
"Having read in a dataframe, or created a new one using one of the expressions just shown, we can use a variety of functions to explore and organize the data. Most of these are very intuitive, so we will mostly try a bunch of examples, you can easily explore these in the link Basic Fuctionality: http://pandas.pydata.org/pandas-docs/stable/basics.html"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### TODO\n",
"\n",
"For the following, which sort the dataframe, first try to predict what will happen, and then try it to confirm your understanding:\n",
"\n",
"- students.sort_values('Userid')\n",
"- students.sort_values('GPA', ascending=False)\n",
"- students.sort_values( ['Gender', 'Userid'] )\n",
"- students.sort_values( ['Gender', 'Userid'], ascending=[False,True] )"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Userid
\n",
"
Gender
\n",
"
ClassYear
\n",
"
GPA
\n",
"
Credits
\n",
"
Transfer
\n",
"
AP
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
biden
\n",
"
M
\n",
"
U2
\n",
"
4.000
\n",
"
33
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
4
\n",
"
yellen
\n",
"
F
\n",
"
U2
\n",
"
3.810
\n",
"
33
\n",
"
0.0
\n",
"
4
\n",
"
\n",
"
\n",
"
1
\n",
"
harris
\n",
"
F
\n",
"
U4
\n",
"
3.630
\n",
"
45
\n",
"
71.1
\n",
"
8
\n",
"
\n",
"
\n",
"
7
\n",
"
haaland
\n",
"
F
\n",
"
U4
\n",
"
3.580
\n",
"
66
\n",
"
30.0
\n",
"
0
\n",
"
\n",
"
\n",
"
2
\n",
"
blinken
\n",
"
M
\n",
"
U4
\n",
"
3.380
\n",
"
113
\n",
"
0.0
\n",
"
12
\n",
"
\n",
"
\n",
"
3
\n",
"
emhoff
\n",
"
M
\n",
"
U2
\n",
"
3.210
\n",
"
28
\n",
"
0.0
\n",
"
0
\n",
"
\n",
"
\n",
"
6
\n",
"
garland
\n",
"
M
\n",
"
U4
\n",
"
3.200
\n",
"
80
\n",
"
44.0
\n",
"
24
\n",
"
\n",
"
\n",
"
5
\n",
"
austin
\n",
"
M
\n",
"
U3
\n",
"
3.034
\n",
"
56
\n",
"
28.0
\n",
"
0
\n",
"
\n",
"
\n",
"
9
\n",
"
pence
\n",
"
M
\n",
"
U4
\n",
"
2.040
\n",
"
106
\n",
"
3.0
\n",
"
24
\n",
"
\n",
"
\n",
"
8
\n",
"
trump
\n",
"
M
\n",
"
U3
\n",
"
1.230
\n",
"
32
\n",
"
0.0
\n",
"
0
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Userid Gender ClassYear GPA Credits Transfer AP\n",
"0 biden M U2 4.000 33 0.0 12\n",
"4 yellen F U2 3.810 33 0.0 4\n",
"1 harris F U4 3.630 45 71.1 8\n",
"7 haaland F U4 3.580 66 30.0 0\n",
"2 blinken M U4 3.380 113 0.0 12\n",
"3 emhoff M U2 3.210 28 0.0 0\n",
"6 garland M U4 3.200 80 44.0 24\n",
"5 austin M U3 3.034 56 28.0 0\n",
"9 pence M U4 2.040 106 3.0 24\n",
"8 trump M U3 1.230 32 0.0 0"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"students.sort_values('GPA',ascending=False)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"There are also many statistical functions which operate mostly on individual columns:"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### TODO\n",
"\n",
"Let us assume that we have created a new dataframe\n",
"\n",
"> `st = students['GPA']`\n",
"\n",
"First try to predict what will happen, and then try it to confirm your understanding.\n",
"\n",
"- `st.max()`\n",
"- `st.min()`\n",
"- `st.mean()`\n",
"- `st.median()`\n",
"- `students['ClassYear'].mode() # the mode is the most frequent value in the list`\n",
"- `students['Credits'].sum()`\n",
"- `st.count() * 2 + st.max() # weird, just to show that values can be used any way you want!`\n",
"- `students[ ['GPA','Credits']].max() # ok, another weird one, it returns the max in each column!`"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"1.23"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"students['GPA'].min()\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Displaying Distributions from DataFrames\n",
"\n",
"A final function that we should look at is the `hist(...)` function, which will (as you might expect), produce a graphical display of the histogram for a column of values. It is precisely the same function that we studied in the first lab, except that it takes as its values the data in the column(s) of the dataframe. If you give it a whole dataframe, it will give you histograms of all the numeric columns:"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"scrolled": false
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"students.hist()\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"It is usually more useful to consider a single column at a time:"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"students['GPA'].hist()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"You can set some of the parameters that work with the `hist(...)` function, such as setting the bin boundaries, colors, and so on:"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"scrolled": true
},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"students['GPA'].hist(bins=[0.0,0.5,1.0,1.5,2.0,2.5,3.0,3.5,4.0],color='r',edgecolor='k',grid=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"You can also use most of the standard matplotlib functions to add titles, change the size, etc.:"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"plt.figure(figsize=(10,5))\n",
"plt.title(\"GPA Data\")\n",
"students['GPA'].hist(bins=[0.0,0.5,1.0,1.5,2.0,2.5,3.0,3.5,4.0],color='g',edgecolor='k')\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Problem Two\n",
"\n",
"Download the following file of student data for 4287 students at a modern university: \n",
"\n",
"> https://www.cs.bu.edu/fac/snyder/cs237/Data/studentdata.csv \n",
"\n",
"and do the following:\n",
"\n",
"(A) Print out the mean GPA for men;\n",
"\n",
"(B) Print out the mean GPA for seniors (U4);\n",
"\n",
"(C) Display the 10 individuals with the largest number of credits earned, sorted in descending order by GPA;\n",
"\n",
"(D) Display a histogram of the GPA of all individuals, with bins for each letter grade, i.e., 0.0, 1.0, 2.0, 2.33, 2.67, 3.0, 3.33, 3.67, and 4.0.\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [],
"source": [
"# Your solution here\n",
"# Make sure to run it before submitting it"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Problem Three\n",
"\n",
"Download the following file of heights and weights of 25,000 individuals: \n",
"\n",
"> https://cs-web.bu.edu/fac/snyder/cs237/Data/biometricdata.csv \n",
"\n",
"and do the following:\n",
"\n",
"(A) Print out the maximum, minimum, mean, and (unbiased) standard deviation (all functions listed above) for the heights of all individuals;\n",
"\n",
"(B) Print out the mean height for all individuals weighing more than 130 pounds;\n",
"\n",
"(C) Print out how many individuals have a height >= 65 inches and <= 70 inches;\n",
"\n",
"(D) Display a histogram of the heights of all individuals from the minimum to the maximum, where each bin represents 1 inch. Calculate the bin boundaries so that the bins are centered on the height, that is, the boundaries are\n",
"half way between each inch measurement:\n",
"\n",
"> [ .... 64.5, 65.5, 66.5, 67.5, ..... ]\n",
"\n",
"The left edge of the lowest bin would be the minimum value (rounded to inches) - 0.5, and the right edge of\n",
"the highest bin would be the maximum (rounded to inches) + 0.5. The height values themselves are not rounded, just\n",
"the maximum and minimum, to get the appropriate bin edges. \n",
"\n",
"Also, provide an appropriate title, and make the histogram larger using an appropriate figsize, as shown above. \n",
"\n",
"Hint: For (D), create the bin edges using the function `np.arange(...)` (Google it!). \n"
]
},
{
"cell_type": "markdown",
"metadata": {
"scrolled": false
},
"source": [
"**Solution:**\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
" ## Problem Four (Normal Distribution)\n",
" \n",
"The lifetime of backup battery systems made by a \n",
"Company A has a mean of\n",
"5 years and a standard deviation of 2 years. Those made by Company B\n",
"have a mean of 4 years and a standard deviation of 18 months. \n",
"Suppose Wayne buys one backup system from Company A, and also one from\n",
"Company B. The one from Company A lasts 4 years and 3 months, and the one\n",
"from Company B lasts 3 years and 9 months. \n",
"\n",
"(A) Which of these backup battery systems performed relatively better, compared\n",
"with other systems from the same company? \n",
"\n",
"(B) If the backup system from Company A lasts 5 years, how long would the system from Company B last if it lasted the same amount of time relative to the performance from each company? \n",
"\n",
"\n",
"Hint: Standardize the two normal distributions and compare!\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Solution**\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Problem Five (Normal Approximation to the Binomial)\n",
"\n",
"This problem concerns the normal approximation to the binomial. The "continuity correction" (sometimes called Yates's Continuity Correction) was shown in lecture. \n",
"\n",
"In this problem we will measure the accuracy of the approximations by using\n",
"the percent error, as in previous homeworks. \n",
"\n",
"\n",
"
(A) Suppose of all the kids that show up on Halloween night, 58% are dressed in Spiderman costumes. If 60 kids show up, what is the probability that between 33 and 38 (inclusive) kids will be dressed in Spiderman costumes? (Use the binomial.)
\n",
"
(B) Repeat the previous question, but using the normal approximation to the binomial, without using the continuity correction, and express the accuracy of your approximation using the percentage error.
\n",
"
(C) Repeat the previous question, but now using the continuity correction, again showing the accuracy using the percentage error.
Suppose the heights of 3400 male students at a university are normally distributed with mean 68 inches and standard deviation 3 inches. That is, you have a random variable $X$ which uniformly at random chooses a male student and returns his height, where $E(X) = 68.0$ and $\\sigma_X = 3.0$.
\n",
"
(A) Supposing sample groups of 25 men are taken from this population (with replacement)\n",
" and the average height of the group calculated, i.e., you are investigating the random variable $\\overline{X}_{25}$. What would be the expected value and standard deviation for $\\overline{X}_{25}$.?
\n",
"
(b) Supposing we wanted to get more accuracy in our sampling procedure, so that we wanted the standard deviation of the result to be at most 0.25 inches. What is the smallest sample size we could use to insure this? (Formally, what is the smallest $n$ for which $\\sigma_{\\overline{X}_n} \\le 0.25$?)
\n",
"
(c) Supposing you take 80 samples of size 25 (80 \"pokes\" of $\\overline{X}_{25}$), in how many samples would you *expect* to find the output from $\\overline{X}_{25}$ between 66.8 and 68.3 inches? The expected value may be a floating-point number.
\n",
"\n",
"\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Solution:**\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Problem Seven (Sampling Theory) \n",
"\n",
"This problem considers three different ways of answering a question about samples from an infinite population. Suppose you flip a fair coin 120 times. What is the probability that 75 or more of the flips will be heads?\n",
"\n",
"(A) First solve this problem precisely using the binomial, showing your formula (you'll need to use Python -- look at the end of the previous cell above to see some useful functions). \n",
"\n",
"(B) Next, solve the problem by using the normal approximation to the binomial (using the continuity correction). \n",
"\n",
"Finally, we will solve this as a problem in sampling: let $X$ be a Bernoulli (each coin flip) and let \n",
"\n",
"$$\\overline{X} = {X_1 + \\cdots + X_{120})\\over 120},$$ so that by the CLT, $\\overline{X}\\sim N(\\mu,\\sigma^2)$\n",
"for some mean $\\mu$ and variance $\\sigma^2$. \n",
"\n",
"(C) Give the mean $\\mu$ and variance $\\sigma^2$.\n",
"\n",
"(D) Now calculate the answer using the CLT, using the continuity correction\n",
"(where effectively the bins have width 1/120, so you should adjust the boundary by\n",
"one half the width of the bin), and give the percentage error. \n",
"\n",
"Hint: You should get the same answer for (B) and (D); they will be close to, but not the same as, the \"ground truth\" answer in (A). "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Solution:**\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Problem Eight \n",
"\n",
"CANCELLED!"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Problem Nine (CLT and Sampling) \n",
"\n",
"Suppose you know that when there was a vote for the major of Cambridge, MA, 54.8% of the people voted for candidate A. \n",
"\n",
"(a) Supposing samples of size 30 are taken, what would you expect to be the mean and standard deviation of the sampling distribution of proportions? Give your result in terms of percentages.\n",
"\n",
"(b) Supposing we wanted to get more accuracy in our sampling procedure, so that we wanted the standard deviation of the sample distribution of proportions to be at most 5%. What is the smallest sample size we could use to insure this?\n",
"\n",
"(c) Supposing you take 100 samples of size 30, in how many samples would you expect to find the proportion accurate to 1%, i.e., between 53.8% and 55.8% ? (Don't worry about using the continuity correction here.)\n",
"\n",
"Hint: This is the same as the last problem, but where the population represents the outcomes of a Bernoulli\n",
"experiment with p = 0.548. In such cases, we do not need to be given the population standard deviation, because it is determined by a formula involving p; (get out\n",
"your \"cheatsheet\" from the midterm!). This is a special case called \"sampling with proportions.\" "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Solution:**\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Problem Ten \n",
"\n",
"CANCELLED! "
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.7.6"
}
},
"nbformat": 4,
"nbformat_minor": 2
}