Welcome to Class!

Gratitude Journal

Announcements

The data science lab opens this week!

Project 1

Import packages

import ??? as ???

Load the names data

my_data = pd.read_csv()

Understanding your data

You should be able to introduce your data sets to people, the same way you introduce a friend!

• What does each row represent? If you don’t know, then you don’t understand what groups you can analyze.
• What does each column represent? If you don’t know, then you don’t understand what information you can evaluate for each group.

Being able to explain your data out loud to someone else follows the same principles as rubber duck debugging.

Introduction to pandas “DataFrame”

What is a pandas DataFrame? We can read the official documentation. I also like the video in this tutorial.

DataFrames come with attributes and built-in functions that can help us get a feel for our data.

Run the code below one line at a time (or use other functions of your choice) to explore the names data. What do you learn?

my_data.columns
my_data.shape
my_data.size
my_data.head()
my_data.describe()

Let’s practice!

1. How many unique names does the names dataframe contain? Work with a partner to find the answer. You might want to look at this pandas cheat sheet.

2. What is the range of years in the names dataframe? Again, work with a partner and use the pandas cheat sheet.

Extra Practice

# Pause: can you explain what this code is doing?
df = pd.DataFrame(
{"a" : [5, 4, 6, 2, 3],
"b" : [7, 8, 9, 10, 11],
"c" : [10, 11, 12, 101, 0]})

# read in data
flights_url = "https://github.com/byuidatascience/data4python4ds/raw/master/data-raw/flights/flights.csv"
flights = pd.read_csv(flights_url)
flights['time_hour'] = pd.to_datetime(flights.time_hour, format = "%Y-%m-%d %H:%M:%S")

# without method chaining
flights = flights.filter(['dep_time'])
flights = flights.assign(hour = lambda x: x.dep_time // 100)
flights = flights.assign(minute = lambda x: x.dep_time % 100)
flights.head(5)

# with method chaining
(flights
    .filter(['dep_time'])
    .assign(
      hour = lambda x: x.dep_time // 100,
      minute = lambda x: x.dep_time % 100
      )
    .head(5)
)

df = pd.DataFrame(
{"a" : [5, 4, 6, 2, 3],
"b" : [7, 8, 9, 10, 11],
"c" : [10, 11, 12, 101, 0]})