How to plot two different dataframe columns at time based on the same datetime x-axis

Question

Hi I have a dataframe like this:

        Date  Influenza[it]  Febbre[it]  Cefalea[it]  Paracetamolo[it]  \
0    2008-01            989        2395         1291              2933   
1    2008-02            962        2553         1360              2547   
2    2008-03           1029        2309         1401              2735   
3    2008-04           1031        2399         1137              2296    

     Unnamed: 6 tot_incidence  
0           NaN          4.56  
1           NaN          5.98  
2           NaN          6.54  
3           NaN          6.95

I'd like to plot different figures with on x-axis the Date column and the y-axis the Influenza[it] column and another column like Febbre[it]. Then again x-axis the Date column, y-axis Influenza[it] column and another column (ex. Paracetamolo[it]) and so on. I'm trying to figure out if there is a fast way to make it without completely manipulate the dataframes.

ImportanceOfBeingErnest · Accepted Answer · 2016-12-17 20:38:51Z

3

You can simply plot 3 different subplots.

import pandas as pd
import matplotlib.pyplot as plt

dic = {"Date" : ["2008-01","2008-02", "2008-03", "2008-04"],
       "Influenza[it]" : [989,962,1029,1031],
        "Febbre[it]" : [2395,2553,2309,2399],
        "Cefalea[it]" : [1291,1360,1401,1137],
        "Paracetamolo[it]" : [2933,2547,2735,2296]}

df = pd.DataFrame(dic)
#optionally convert to datetime
df['Date'] = pd.to_datetime(df['Date'])

fig, ax = plt.subplots(1,3, figsize=(13,7))
df.plot(x="Date", y=["Influenza[it]","Febbre[it]" ], ax=ax[0])
df.plot(x="Date", y=["Influenza[it]","Cefalea[it]" ], ax=ax[1])
df.plot(x="Date", y=["Influenza[it]","Paracetamolo[it]" ], ax=ax[2])

#optionally equalize yaxis limits
for a in ax:
    a.set_ylim([800, 3000])

plt.show()

If you want to plot each plot separately in a jupyter notebook, the following might do what you want.
Additionally we convert the dates from format year-week to a datetime to be able to plot them with matplotlib.

%matplotlib inline
import pandas as pd
import matplotlib.pyplot as plt

dic = {"Date" : ["2008-01","2008-02", "2008-03", "2008-04"],
       "Influenza[it]" : [989,962,1029,1031],
        "Febbre[it]" : [2395,2553,2309,2399],
        "Cefalea[it]" : [1291,1360,1401,1137],
        "Paracetamolo[it]" : [2933,2547,2735,2296]}

df = pd.DataFrame(dic)
#convert to datetime, format year-week -> date (monday of that week)
df['Date'] = [ date + "-1" for date in df['Date']] # add "-1" indicating monday of that week
df['Date'] = pd.to_datetime(df['Date'], format="%Y-%W-%w")

cols = ["Febbre[it]", "Cefalea[it]", "Paracetamolo[it]"]
for col in cols:
    plt.close()
    fig, ax = plt.subplots(1,1)
    ax.set_ylim([800, 3000])
    ax.plot(df.Date, df["Influenza[it]"], label="Influenza[it]")
    ax.plot(df.Date, df[col], label=col)
    ax.legend()
    plt.show()

edited Dec 17, 2016 at 20:38

answered Dec 17, 2016 at 14:18

ImportanceOfBeingErnest

342k61 gold badges737 silver badges771 bronze badges

Sign up to request clarification or add additional context in comments.

8 Comments

Alex Punkallo Over a year ago

Nice! I'd like to make it without writing everytime a line because i have to use many columns but it's good anyway! I'm plotting in the jupiter notebook and i ask you if there is a way to make the subplots one after the other and not side by side (i mean figures with the largest x-axis and small y-axis)

ImportanceOfBeingErnest Over a year ago

This is an important piece of information that should be part of the question. I updated the answer.

Alex Punkallo Over a year ago

I tried with edited code but i receive an error to this line 'ax.plot(df.Date, df["Influenza[it]"], label="Influenza[it]")' : 'ValueError: invalid literal for float(): 2016-16'. In case i run 'df['Date'] = pd.to_datetime(df['Date'])' i receive this error 'ValueError: month must be in 1..12'

ImportanceOfBeingErnest Over a year ago

Sorry, I forgot to remove the comment: Using matplotib you need to use datetimes - it's not optional anymore. The error that you get, makes sense, since "2016-16" is no valid date. Are you expecting that matplotlib invents 4 new months in the year or did I misunderstand the datetime format you are using?

ImportanceOfBeingErnest Over a year ago

Well that is.. another piece of information that should be part of the question. Is your data covering several years? Are dates equally spaced, i.e. is every week between the first and last entry present? Would you rather have the date axes in weeks or in normal dates (like every monday of that specific week)?

|

Collectives™ on Stack Overflow

How to plot two different dataframe columns at time based on the same datetime x-axis

1 Answer 1

8 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

8 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related