CODEWITHSUNDEEP
python-3.xpandasmatplotlibseaborn
gm-123
gm-123

Reputation: 238

Plot columns of data-frame based on conditions of other columns in Python

Below is my dataset:

    City    Product Spend_Year  Total_Spend
0   BANGALORE   Gold    2004    248006886
1   BANGALORE   Gold    2005    357076946
2   BANGALORE   Gold    2006    329375735
3   BANGALORE   Platimum    2004    138238363
4   BANGALORE   Platimum    2005    167115119
... ... ... ... ...
67  TRIVANDRUM  Platimum    2005    222789632
68  TRIVANDRUM  Platimum    2006    129408676
69  TRIVANDRUM  Silver  2004    14957733
70  TRIVANDRUM  Silver  2005    27465271
71  TRIVANDRUM  Silver  2006    31632500

I need to plot City vs Total_Spend while selecting one product and Spend_Year for all possible combinations.

Example: Plot of City vs Total_Spend for Gold Product in the Year 2004.

P.S.: In the data set I have 3 product categories and data is available for 3 years.

Any help would be highly appreciated.

Upvotes: 0

Views: 253

Answers (2)

Quang Hoang
Quang Hoang

Reputation: 150745

You can use seaborn.FaceGrid:

g = sns.FacetGrid(data=df, row='Spend_Year', col='Product')
g = g.map(sns.lineplot, 'City', 'Total_Spend')

Or you can use groupby():

fig, axes = plt.subplots(3,3)

for (k,d), ax in zip(df.groupby(['City','Spend_Year']), axes.ravel()):
    city, year = k
    d.plot(x='City', y='Total_Spend', ax=ax)

    # extra format with ax if needed
    ax.set_title(f'{city} - {year}')

Upvotes: 1

bjonen
bjonen

Reputation: 1563

The code below will make one plot for each combination. It should be easy to use matplotlib subplots to move them all into one figure. 

import matplotlib.pyplot as plt
for product in df['Product'].unique():
    for syear in df['Spend_Year'].unique():
        plt.figure()
        ax = df.query('Product==@product and Spend_year==@syear').set_index('City')['Total Spend'].plot()
        ax.set_title(f'Product {product} year {syear}')

Upvotes: 0

Related Questions