Plotting the most recent data points with Seaborn scatterplot

Question

I'm trying to plot the predicted vs actual of a stock using Seaborn's scatterplot, I can plot the scatter fine, but what I want to do is also visualise where today's data is, in a different colour.

I tried doing:

current_x = df['Prediction'].iloc[-1]
current_y = df['Actual'].iloc[-1]

and plotting that but got this error message :

ValueError: If using all scalar values, you must pass an index.

Any help would be really appreciated cheers.

EDIT:

So I have a df containing a df['Prediction'] and df['Actual'] columns of price data, the code to print the scatterplot I have used so far is very imple:

sns.scatterplot(x='Predicted', y='Actual, data=data)

What I'm looking for is just to also plot, on top of this original scatter graph, todays most recent data for x and y, so .iloc[-1] for each, if you will.

Please post [sample data](https://stackoverflow.com/questions/20109391/how-to-make-good-reproducible-pandas-examples) even plotting code so we understand your process. — Parfait, May 08 '19 at 12:38
take a look at matplotlib's [animation](https://matplotlib.org/api/animation_api.html). — Quang Hoang, May 08 '19 at 12:45

score 3 · Answer 1 · answered May 08 '19 at 13:12

3

Use the argument hue when creating the scatterplot. If you don't have a hue, you can create one easily for your simple case:

data['hue']=[0]*(len(df)-1)+[1]
sns.scatterplot(x='Predicted', y='Actual', hue='hue', data=data)

answered May 08 '19 at 13:12

CAPSLOCK

6,243
3
33
56

score 2 · Accepted Answer · answered May 08 '19 at 13:04

Bit hacky, but this works.

data = pd.DataFrame({'Prediction': np.random.rand(10), 
                     'Actual': np.random.rand(10)})

sns.scatterplot(x='Prediction', y='Actual', data=data)
sns.scatterplot(x='Prediction', y='Actual', data=data.iloc[-1].to_frame().T)

score 2 · Answer 3 · answered May 08 '19 at 14:26

In case you want to use matplotlib directly instead of seaborn, it's very straight forward to plot all but the last point and the last point separately.

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

data = pd.DataFrame({'Prediction': np.random.rand(10), 
                     'Actual': np.random.rand(10)})

# Plot all but the last point
plt.scatter(x='Prediction', y='Actual', data=data.iloc[:-1])
# Plot only the last point
plt.scatter(x='Prediction', y='Actual', data=data.iloc[-1])

plt.show()

Plotting the most recent data points with Seaborn scatterplot

3 Answers3