Seaborn Lineplot displays Index error, pandas plot fine

Question

I am having an interesting issue with displaying data with Seaborn lineplot.

I have item sales for 5 items over a period of time. I would like to see the sales after each product's introduction.

Here is my code:

items  = ['Item 1', 'Item 2', 'Item 3', 'Item 4', 'Item 5']

fig, ax = plt.subplots(squeeze=False)
ax[0] = sns.lineplot(x=item_sales.index, y='Item 1', data=item_sales, alpha=0.2) 
ax[1] = sns.lineplot(x=item_sales.index, y='Item 2', data=item_sales, alpha=0.2) 
ax[2] = sns.lineplot(x=item_sales.index, y='Item 3', data=item_sales, alpha=0.2) 
ax[3] = sns.lineplot(x=item_sales.index, y='Item 4', data=item_sales, alpha=0.4)
ax[4] = sns.lineplot(x=item_sales.index, y='Item 5', data=item_sales, alpha=0.2)
ax.set_ylabel('')  
ax.set_yticks([])
plt.title('Timeline of item sales')
plt.show()

This code errors out with the following line but draws 2 lines:

ax[1] = sns.lineplot(x=item_sales.index, y='Item 2', data=item_sales, alpha=0.2) 

IndexError: index 1 is out of bounds for axis 0 with size 1

However, following line displays the plot perfectly without any errors:

item_sales.plot()

What could be the reason for the above error - the data is pretty clean and there are no missing values:

item_sales.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 36 entries, 0 to 35
Data columns (total 6 columns):
 #   Column        Non-Null Count  Dtype 
---  ------        --------------  ----- 
 0   Date Created  36 non-null     object
 1   Item 1        36 non-null     int64 
 2   Item 2        36 non-null     int64 
 3   Item 3        36 non-null     int64 
 4   Item 4        36 non-null     int64 
 5   Item 5        36 non-null     int64 
dtypes: int64(5), object(1)
memory usage: 1.8+ KB

Thank you.

You don't have multiple axes, try using just sns.lineplot(data=item_sales) — eemilk
– eemilk, Commented Sep 29, 2020 at 10:13
Thank you, but this didn't work ValueError: A wide-form input must have only numeric values. Plus, this is not what I need, as I will highlight an Item by changing its alpha value, therefore I have to be able to manipulate the lines individually. — tbalci
– tbalci, Commented Sep 29, 2020 at 10:33
Your column zero is object type, so you have to choose just columns 1-5 to plot them as they are numeric. Take a look from here stackoverflow.com/questions/45556499/… , maybe it's helpful — eemilk
– eemilk, Commented Sep 29, 2020 at 11:28

gherka · Accepted Answer · 2020-09-29 11:46:35Z

1

The reason you're getting the IndexError is because your ax object is a 2-dimensional array and you're indexing on the first (length=1) dimension:

squeeze bool, default: True
If False, no squeezing at all is done: the returned Axes object is always a 2D array containing Axes instances, even if it ends up being 1x1.

If you want to plot multiple lines on the same plot, you can just have them share the same ax by passing it to seaborn like so:

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

# prepare sample data
items  = ['Item 1', 'Item 2', 'Item 3', 'Item 4', 'Item 5']
sales_data = dict(zip(items, np.random.randint(0, 25, (5, 30))))
item_sales = pd.DataFrame(sales_data)

fig, ax = plt.subplots(figsize=(8,4))

sns.set_palette("tab10", n_colors=5)
sns.lineplot(x=item_sales.index, y='Item 1', data=item_sales, alpha=0.3, ax=ax) 
sns.lineplot(x=item_sales.index, y='Item 2', data=item_sales, alpha=0.3, ax=ax) 
sns.lineplot(x=item_sales.index, y='Item 3', data=item_sales, alpha=0.3, ax=ax) 
sns.lineplot(x=item_sales.index, y='Item 4', data=item_sales, alpha=1, ax=ax)
sns.lineplot(x=item_sales.index, y='Item 5', data=item_sales, alpha=0.3, ax=ax)
ax.set_ylabel('')  
ax.set_yticks([])
plt.title('Timeline of item sales')
plt.show()

answered Sep 29, 2020 at 11:46

gherka

1,45610 silver badges18 bronze badges

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

1 Comment

tbalci Over a year ago

Thanks very much @gherka, your reply was exactly where things were failing. I made the changes you mentioned and now everything is in place.

Collectives™ on Stack Overflow

Seaborn Lineplot displays Index error, pandas plot fine

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related