Plot the count of a Pandas df column using Plotly

Question

I have a df column where its data are only categorical (e.g., [a, b, c, a, a, d, c, b, etc]). I want to plot the count of these data using plotly count bar (bar chart).

I have calculated the count of the data using df.groupby('<col_name>')['<col_name>'].count(), but this returns a series data structure so I will only have the count data (1-D).

How can I get the count result and the corresponding data item in the resulting output efficiently?

I want to get this output and plot the bar chart using Plotly:

import plotly.express as px

fig = px.bar(count_df, x="<col_name>", y="count", color="count", title="----------")
fig.show()

I think you need this.

r-beginners
– r-beginners

2021-03-16 10:22:17 +00:00
Commented Mar 16, 2021 at 10:22 — r-beginners
– r-beginners, Commented Mar 16, 2021 at 10:22

vestland · Accepted Answer · 2021-03-16 22:14:54Z

4

It doesn't get more efficient than this:

df['x'].plot(kind = 'hist')

And NO, this isn't matplotlib, but rather a figure constructed directly from a pandas dataframe using plotly as a backend. And yes, it's awesome!

Complete code:

import random
import pandas as pd
pd.options.plotting.backend = "plotly"
random.seed(7)

df = pd.DataFrame({'x':[random.choice(list('abcde')) for i in range(25)]})
fig = df['x'].plot(kind = 'hist')
fig.layout.bargap = 0
fig.show()

answered Mar 16, 2021 at 22:14

vestland

62.1k41 gold badges220 silver badges343 bronze badges

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

1 Comment

mj_whales Over a year ago

While this is efficient, the x-axis values for a histogram need to be at least ordered, if not numeric. Also, the y-axis values need to be continuous. Here, the x-axis is categorical and the y-axis is discrete. So we need a bar or column chart, not a histogram.

Ali H. Kudeir · Accepted Answer · 2021-03-16 10:34:03Z

3

Answering my own question.

I found a solution by converting the result of value_counts(returns a Series) to a pd DataFrame. Ref : SO question and answers

import plotly.express as px

new_df = df['<col_name>'].value_counts().rename_axis('<col_name>').reset_index(name='counts')

fig = px.bar(new_df, x="<col_name>", y="counts", color="counts", title="----------")
fig.show()

answered Mar 16, 2021 at 10:34

Ali H. Kudeir

1,1043 gold badges12 silver badges27 bronze badges

Comments

Khyber Thrall · Accepted Answer · 2021-03-16 10:04:39Z

1

Should be able to use .index of the result to give you the values for x-axis and the series itself for the y-axis.

Also, I think using df[‘col_name’].value_counts() is probably what you want to use here.

answered Mar 16, 2021 at 10:04

Khyber Thrall

513 bronze badges

Comments

Andrea NR · Accepted Answer · 2021-03-16 10:06:13Z

0

You can get it easily by using hist(), for example:

df['<col_name>'].hist()

and also you can see the abs frequency.

Another way to do the same is:

df['<col_name>'].value_counts().plot(kind='bar')

answered Mar 16, 2021 at 10:06

Andrea NR

1,8171 gold badge8 silver badges16 bronze badges

1 Comment

Ali H. Kudeir Over a year ago

Thank you but I want to use Plotly and not the Pandas builtin stuff

Collectives™ on Stack Overflow

Plot the count of a Pandas df column using Plotly

4 Answers 4

Complete code:

1 Comment

Comments

Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Complete code:

1 Comment

Comments

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related