pivot table in proper order for the heatmap

Question

while creating a heatmap I have following syntax:

import matplotlib.pyplot as plt
import seaborn as sns
sns.set_theme()
data=df.rename(columns={0:'Year', 1:'Month', 2:'Count'})
data= pd.pivot_table(data, values='Count', index='Year', columns='Month')
f, ax = plt.subplots(figsize=(15, 6))
sns.heatmap(data, annot=True, fmt="d", linewidths=.5, ax=ax)

and that generates following heatmap:

What I want is ascending or descending order of month along x-axis. That is Jan, Feb, Mar, etc. How do I accomplish like this?

Sample data here:

    0        1      2
0   2005    Jan     84482
1   2011    Apr     28243
2   2007    Apr     64992
3   2013    Feb     46542
4   2016    Sept    24445
5   2011    July    23346
6   2019    Dec     28251
7   2015    Jan     34505
8   2007    June    72561
9   2015    Apr     26973
10  2006    May     102896
11  2006    Jan     88664
12  2012    Nov     32046
13  2005    Sept    65498
14  2014    Sept    24856

Answer 1

If your data frame is sorted:

import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd
import numpy as np

dates = pd.date_range("2000-01-01", periods=48, freq="M")
df = pd.DataFrame({"Year":dates.year,
                   "Month":dates.month_name().str.slice(stop=3),
                  "Count":np.random.randint(0,100,48)})

    Year    Month   Count
0   2000    Jan 96
1   2000    Feb 5
2   2000    Mar 97
3   2000    Apr 40
4   2000    May 55
5   2000    Jun 16

Then:

df['Month'] = pd.Categorical(df['Month'],categories=df['Month'].unique())

Else create one list of the order:

month_order = pd.date_range("2000-01-01", periods=12, freq="M").month_name().str.slice(stop=3)
df['Month'] = pd.Categorical(df['Month'],categories=month_order)

And the plot will work:

data= pd.pivot_table(df, values='Count', index='Year', columns='Month')
f, ax = plt.subplots(figsize=(15, 6))
sns.heatmap(data, annot=True, fmt="d", linewidths=.5, ax=ax)