Contents

4.0 Reading in Data

Pandas, dataframes, and seaborn 🐼

In the previous lessons, we have been focusing on learning the nuts and bolts of Python: lists, matplotlib, and loops. We used these to make models of phyllotaxy and growing sunflowers.

Now, we shift our focus from formal coding and modeling to data analysis. Data analysis means dataframes, pandas, and seaborn!

Often, your data will be a lot like an Excel spreadsheet, with rows, columns, and column headers. But we need more functionality than Excel can provide. That’s where pandas comes in. Pandas is a lot like the Python version of R, which means that it is built to use dataframes and has a lot of built-in statistical capabilities. For plotting, Seaborn is a Python graphing module that, although lacking the versatiity of matplotlib, is built for graphing with statistical needs in mind.

We can’t cover all that pandas does in a few notebooks. But as you continue your studies and need to do something specific with respect to data wrangling or analysis, there are many tutorials that will certainly have what you need. And of course, google your question!

The best way to use Seaborn is to gaze the gallery of plots and select what you need! The documentation and examples will be there.

Watch the video below to get started using pandas and seaborn. We will be analzying a dataset examining harvest dates in European grapevines with respect to a warming climate later in the notebook. So, in this video, we will be looking at measurements of CO2 levels in ppm from the Mauna Loa Observatory in Hawaii. You can download the dataset used in the video tutorial here.

Pay attention to the following points:

🐍 Python learning objectives

  • How to read in a dataframe using pandas (1:32)

  • How to quickly assess dataframe properties, using head and tail (2:28)

  • How to select columns of a dataframe (4:10)

  • How to perform descriptive statistics using pandas (.min, .max, .std, .value_counts) (6:09)

  • How to mask data (a Boolean statement to fish out data that you want, square brackets after a dataframe) (8:00)

  • How to use iloc to isolate data within a dataframe (rows first, columns second, start:end, and square brackets) (12:13)

  • Plotting with seaborn (specialized plotting for statistics) (16:32)

🌻 Plant learning objectives

  • CO2 levels are increasing, which increases global temperatures

  • Plants respond to increases in CO2 and temperature

  • In the next activity, we’ll examine the effects of global warming on crop plants

  • Statistics and data visualization are powerful techniques to explore global trends

How to read in a dataframe using pandas

Watch this video from 0:00 to 2:25

# To load the video, execute this cell by pressing shift + enter

from IPython.display import YouTubeVideo
from datetime import timedelta
start=int(timedelta(hours=0, minutes=0, seconds=0).total_seconds())
end=int(timedelta(hours=0, minutes=2, seconds=25).total_seconds())

YouTubeVideo("jEQRU55x0e4",start=start,end=end,width=640,height=360)

The following is a transcript of the video.

Let’s learn how to read in a dataframe using the pandas module.

Pandas is usually imported as pd so you press shift + enter to import pandas and the most useful function in pandas is probably  read_csv(). This is a function that will take .csv data and read it in as a dataframe. Dataframes have rows and columns and the columns have names. By default, the files should be in your home directory but you can also provide a path to it.

So for example I’ll be loading in this data from my desktop so I added “./Desktop” and then the name of the file. So we hit shift + enter and now we’ve created a variable, data, that contains our dataframe.

NOTE: In the video the file is read in locally from the Desktop, but here we will load in the data using a link from GiHub where the data is posted.

# Pandas is usually imported in as "pd"

import pandas as pd

# A useful function is pd.read_csv()
# which reads in a CSV as a dataframe
# Dataframes have rows and columns with names

# By default the the file should be in your home directoty
# but you can also provide a path
# For example, I'll load it in from my Desktop

# data = pd.read_csv("./Desktop/co2_mlo_weekly.csv")

# NOTE: in the video the file is read in locally from the Desktop
# but here we will load in the data using a link from GitHub 
# where the data is posted

data = pd.read_csv('https://raw.githubusercontent.com/DanChitwood/PlantsAndPython/master/co2_mlo_weekly.csv')
Welcome / Bienvenid@ 4.1 Head and Tail