Count frequency of entries in a pandas column, then plot them in plotly with X axis string lablel

Question

I have the following pandas column:

FuncGroup
ABC
ABC
ABC
ABC
BCD
BCD
BCD
SDS
SDS
ABC
BCD
SDS
BCD

and I want to get this expected output in pandas dataframe:

pd['FunctionGroup','FunctionCount']
ABC  4
BCD  5
SDS  3

How to do this, it's needed for graph purpose.

edit 1: by referring to the below answers i did some modification of the original code to plot using plotly . Now all the counts are plotted but the X axis label is not coming using this method this is the reason i want the label and the count to be stored in a pd.

reference code

otrace1 =go.Bar(
    #x=stock_opt_pe.index
    x=datalist['Function group'].nunique(),
    y=datalist['Function group'].value_counts(),
    text=datalistFg, # dont know what to give here to get a X axis label
    textposition = 'auto',
    #xaxis-type (enumerated: “-” | “linear” | “log” | “date” | “category” )
    #xaxis-type (enumerated: “-” | “linear” | “log” | “date” | “category” )
    #name='Function Group Vx RespPerson',
    #orientation = 'v',
    #marker = dict(
        #color = 'rgba(224, 224, 224, 0.6)',
        #line = dict(
            #color = 'rgba(246, 250, 206, 1.0)',
            #color = 'rgb(60, 60, 60)',
            #width = 0)
    #)
)

This is simply "Count entries in a column". Then barplot them. I can't see why you'd say "using unique reference of another column". — smci
– smci, Commented Oct 19, 2018 at 9:46
@smci As I updated below in plotly the x-axis prints number 1 2 3 4 5 ...25 which is a count of each unique label. What is want is a print of X-axis label as ABC, BCD, SDS so in this method the count not work. — Marx Babu
– Marx Babu, Commented Oct 19, 2018 at 9:50
Ok so your new question is how to use the labels as the X-axis. With plotly. That's pretty easy. But it's also a new separate question. In general on SO, the etiquette is not to ask multiple questions in one, especially tacking follow-on questions on. But for this time, please just edit your question title and body. I tried to help you with "Count frequency of entries in a pandas column, then barplot them". By the way the term for ""Count entries in a column using unique reference of another column" is "Count frequency(/ies) in column" — smci
– smci, Commented Oct 19, 2018 at 23:17

Anton vBR · Accepted Answer · 2018-10-19 09:30:19Z

4

You might be looking for value counts, which is similar to collections counter.

df['FuncGroup'].value_counts()

For plotting, look at this example:

import pandas as pd

df = pd.DataFrame({
    'FuncGroup': ['ABC','ABC','BCD']
})

s = df['FuncGroup'].value_counts()
s.plot(kind='bar')

dfout = df['FuncGroup'].value_counts().reset_index()
print(dfout)

#  index  FuncGroup
#0   ABC          2
#1   BCD          1

Returns:

edited Oct 19, 2018 at 9:30

answered Oct 17, 2018 at 10:07

Anton vBR

19k6 gold badges47 silver badges47 bronze badges

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

8 Comments

Marx Babu Over a year ago

i need a one to one mapping so that [ABC, 4] [BCD,5][SDS,3] comes and i can graph it ABC BCD SDS as X axis and value 4 5 3 as Y axis.

Anton vBR Over a year ago

@MarxBabu Yes sure. But you don't need that. just use .plot()

jezrael Over a year ago

@AntonvBR - it is value_counts + ploting, so reopening ;)

Anton vBR Over a year ago

@jezrael I dunno man. I think this might be a dupe anyway. Sometimes it is hard to ask the right question right away.

jezrael Over a year ago

@AntonvBR - Ya, you are right. Btw, I like more s.plot.bar() ;)

|

Rahul Agarwal · Accepted Answer · 2018-10-22 09:04:00Z

3

Check if this works for you:

import pandas as pd
import plotly.plotly as py

Sample df:

raw =pd.DataFrame({'FuncGroup':[
'ABC',
'ABC',
'ABC',
'ABC',
'BCD',
'BCD',
'BCD',
'SDS',
'SDS',
'ABC',
'BCD',
'SDS',
'BCD']})

Create new df with counts:

s = raw['FuncGroup'].value_counts() ## Counts the occurrence of unqiue elements and stores in a variable called "s" which is series type
new = pd.DataFrame({'FuncGroup':s.index, 'Count':s.values})  ## Converting series type to pandas df as plotly accepts dataframe as input. The two columns of df is FuncGroup which is being made by index of series and new variable called count which is made by values of series s.

Create plotly bar graph:

py.iplot(new, filename='basic-bar')

edited Oct 22, 2018 at 9:04

answered Oct 17, 2018 at 10:02

Rahul Agarwal

4,1168 gold badges33 silver badges56 bronze badges

4 Comments

Marx Babu Over a year ago

i have used this and passed this to plotly it prints the value as 1 2 3 5 ...25 it is not useful to print the label in the group.

Rahul Agarwal Over a year ago

Your original question didn't ask for ploting. It stopped to get only the count of unique values, which is what the above solution does!!

Marx Babu Over a year ago

Later i added the plot details . Plot has no interactive chart so plotly is my need . How this can be solved. it is going crazy still.

Marx Babu Over a year ago

This works fine . Thank you for your answer.It is a bit not straightforward still it is the right answer for me . Thank you.If you could just explain new=pd... line of code bit in details it will be helpful learning.

Collectives™ on Stack Overflow

Count frequency of entries in a pandas column, then plot them in plotly with X axis string lablel

2 Answers 2

8 Comments

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

8 Comments

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related