Python: maximum displayable points on chart

[Anonymous]

Hey guys!

I have created a scatter plot that has the date on the X-axis and various values on the Y-axis (Columns: maxCurrentCutter & maxCurrentNormal — see screenshot). Since I wanted a slicer for differentiation, I transformed the table into a Unipivot table. However, when I want to visualize this, only the data points from 2020 to 2022 are visible on the visual, not 2020-2024. I'll show you a screenshot with the original table (maxCurrentCutter & Normal as two separate columns) and one with the Unipivot table.

In the Unipivot table, there are naturally twice as many rows as in the original table (about 250K vs. 125K), and even when I filter the Unipivot visualization so that about 140K data points are again present, it works as usual. But the original Visual has also 250K values and there it works to show all the data.

Therefore, my question is: Is there a maximum number of displayable points in Python visuals, or why else are not all datasets displayed?

That doesn't make any sense to me 

--------------------------------------------------------------------------------------------

Unipivot-visualisation:

import pandas as pd

import numpy as np

import matplotlib.pyplot as plt

import matplotlib.dates as mdates

# Datetype of 'Date'

dataset['Date'] = pd.to_datetime(dataset['startDate'])

# Define colors based on the 'Attribute' column.

colors = np.where(dataset['Attribute'] == 'maxCurrentCutter', 'green', 'orange')  # 'green' is for cutter, 'orange' is for normal

#Plot

plt.scatter(dataset['Date'], dataset['Value'], s=10, c = colors)

plt.title('Max Thread Cut Current')

plt.xlabel('DateTime')

plt.ylabel('[Ampere]]')

plt.grid(True)

# vertical Grid

#plt.gca().xaxis.grid(False)

plt.xticks(rotation=30)

plt.tight_layout()

plt.legend(handles=[matplotlib.patches.Patch(color='green', label='maxCurrentCutter'),

                    matplotlib.patches.Patch(color='orange', label='maxCurrentNormal')],

           loc='upper left')

plt.show()

Original visual:

import pandas as pd

import numpy as np

import matplotlib.pyplot as plt

# Datetype of 'Date'

dataset['Date'] = pd.to_datetime(dataset['startDate'])

# Plot  

plt.scatter(dataset['Date'], dataset['maxCurrentCutter'], color='green', s=10, label='maxCurrentCutter')

plt.scatter(dataset['Date'], dataset['maxCurrentNormal'], color='orange', s=10, label='maxCurrentNormal')

plt.title('Max Thread Cut Current')

plt.xlabel('Date')

plt.ylabel('[Ampere]]')

plt.grid(True)

# vertical Grid

#plt.gca().xaxis.grid(False)

plt.xticks(rotation=30)

plt.tight_layout()

plt.legend(loc='upper left')

plt.show()

[Anonymous · Accepted solution]

Hi @Anonymous ,

This does not match what the official documentation says, my guess is that the Unipivot table has an increased amount of data and therefore only shows the first part of the data.

Hope it helps!

Best regards,
Community Support Team_ Scott Chang

If this post helps then please consider Accept it as the solution to help the other members find it more quickly.

[Anonymous · Accepted solution]

The solution is simple: The data the Python visual uses for plotting is limited to 150,000 rows. If more than 150,000 rows are selected, only the top 150,000 rows are used, and a message appears on the image. The input data also has a limit of 250 MB.

More for this: https://learn.microsoft.com/en-us/power-bi/connect-data/desktop-python-visuals#limitations

Hope it helps somebody

[Anonymous · Accepted solution]

Hi @Anonymous ,

This does not match what the official documentation says, my guess is that the Unipivot table has an increased amount of data and therefore only shows the first part of the data.

Hope it helps!

Best regards,
Community Support Team_ Scott Chang

If this post helps then please consider Accept it as the solution to help the other members find it more quickly.