Power BI Desktop with Python slow, once published to the Power BI Service nearly unworkable

[lutho]

Hi All, I am needing the flexibility of python to produce visuals not available. I tried various simple scenarios where I simply graph a grid or basic plot with a dataframe generated within the script (no data feed from Power BI to Python). Here the refresh is 2-3 seconds but when published to the Power BI Service this becomes a 25-40 second delay. So to re-iterate - the visual generates its own data and does not have to fetch any data.

This poor performance is unworkable as my audience need to click through and they are simply not going to wait that long per page. My data source settings are direct import - but should not be relevant to this scenario given the data is generated within the visual?

Am I doing something wrong to cause this poor performance? I could not image this to be the norm.

Thanks in advance,

Lex

The scipt is:

# The following code to create a dataframe and remove duplicated rows is always executed and acts as a preamble for your script:

# dataset = pandas.DataFrame()

# dataset = dataset.drop_duplicates()

# Paste or type your script code here:

from matplotlib import pyplot as plt

import numpy as np

randn = np.random.randn

from pandas import *

idx = Index(np.arange(1,11))

df = DataFrame(randn(10, 5), index=idx, columns=['A', 'B', 'C', 'D', 'E'])

vals = np.around(df.values,2)

norm = plt.Normalize(vals.min()-1, vals.max()+1)

colours = plt.cm.hot(norm(vals))

fig = plt.figure()

ax = fig.add_subplot(111, frameon=True, xticks=[], yticks=[])

the_table=plt.table(cellText=vals, rowLabels=df.index, colLabels=df.columns,

colWidths = [0.03]*vals.shape[1], loc='center',

cellColours=colours)

plt.show()

[lbendlin]

All this just to see if the RNG has enough entropy? Maybe try the performance of a real dataset?