0845 643 64 63

Jon Fletcher

How to delay a Python loop

In this blog post, I will run through 3 different ways to execute a delayed Python loop. In these examples, we will aim to run the loop once every minute.
To show the delay, we will print out the current datetime using the datetime module.

1 – Sleep
The sleep function from Python’s time module pauses the Python execution by the number of seconds inputted. The example below pauses the script for 60 seconds.

The above script has a 60 second delay between the end of a loop and the start of the next. However, what if we wanted to execute the next loop 60 seconds after the start of the previous loop.
In other words, how do we start the loop at the same time every minute.

2 – IF Statement
To do this we will switch to a WHILE loop and use an IF statement to see how many seconds have passed in the current minute. A WHILE loop is used because it will execute until i = 10, until the datetime has been printed 10 times. The IF statement checks the time to see when the print statement should be executed.
We will print the datetime every time the minutes changes. The Sleep function will continue to be used as the loop would execute multiple times in one second if we didn’t.

date_time[-2:] – is the last two characters of our datetime variable which is the number of seconds that have passed this minute.

3 – Task Scheduler
The previous two options are good for executing a loop a few times, ten in our case. If we wanted to execute a python script continuously without expiring, we could use the above examples with an infinite appending loop.
However, if one loop errors the script will stop. Therefore, we want to execute the entire Python script once a minute using an external trigger. This is where we can use Task Scheduler.

Task Scheduler can execute a python script from source but it is often easier to use a batch file. The batch file includes the location of the python application (python.exe) and the location of the python script (.py). For more detail on using Task Scheduler and batch files to run Python scripts, please see the following datatofish post – https://datatofish.com/python-script-windows-scheduler

Our batch file is:

To demonstrate Task Scheduler, I’m going to run the following Python code every minute.
This code uses Pandas to produce a blank CSV file, but the name of CSV file is the datetime the script was run.

These following screenshots show the triggers and actions used.

This produced the following CSV files, we can see that the files takes 1 – 4 seconds to create.

In summary we have seen three different ways to delay a Python loop, two using loops inside Python and one using Task Scheduler. All can be used depending on what kind of delay is best.

Sorting a Power BI table by multiple columns

A common request that is raised by clients is how to sort a table in Power BI by multiple columns, in the same way you can in Excel.
For a while, there was no way (at least no easy way) to do this until the Power BI March 2020 update.

I learnt this tip from the following YouTube video:
https://www.youtube.com/watch?v=ik0K1H9j2Uc
Full credit to Dhruvin Shah, check his video out.

Below I have a Power BI table displaying fruit sales, currently unsorted.

To sort the table by Fruit click on the column header Fruit.

The table is now sorted by Fruit in alphabetical order.
To add a secondary sort on Sales, hold the Shift key and click on the column header Sales.

The table is now sorted by
– Fruit name in alphabetical order
– Sales in descending order

Some extras to note:
– There is no limit on the number of columns that can be used to sort a table. Just hold the shift key and keep choosing columns.
– This feature is not available for matrices.
– To switch the sorting from ascending to descending or vice-versa continue to hold shift and click on the column header again.

Creating a Python function to calculate Pi

Pi is 3.14159 to 5 decimal places.
To work out Pi, we will be using Leibniz’s formula:
X = 4 – 4/3 + 4/5 – 4/7 + 4/9 – …
This series converges to Pi, the more terms that are added to the series, the closer the value is to Pi.
For the proof on why this series converges to Pi – https://proofwiki.org/wiki/Leibniz’s_Formula_for_Pi
There are several points to note about the series:

  • It’s infinite, we need to find a way to continue adding term after term.
  • The denominator of the fraction increases by 2 every term.
  • The terms alternate between positive and negative.

Firstly, let’s create a function called pi.

To continue adding terms, let’s use a for loop. Every time the loop executes another term is added.

range(1,10) will produce the numbers 1, 2, … 9, 10.
However, before the loops starts, our variables’ initial values need to be set.

The pi series will start from 0.
n represents the numerator of our fractions which is the constant 4.
d represents the denominator of our fractions which starts as 1.

d needs to increase by 2 every loop, let’s use sum equals to do this.
Pi will also use a sum equals, a denotes our positive/negative function which we will get on to:

The only problem left is how to get a to alternate between 1 and -1.
This is where we introduce modulo.
Modulo outputs the remainder of a division and is denoted by %.

This example shows from numbers 1 to 4, modulo 2 alternates between 0 and 1.
If we multiple by 2 and minus 1 it will alternate between 1 and -1 which is what we require.

Putting everything together gives:

This isn’t close to 3.14159 at all.


However, we are only executing the loop 10 times, hence only 10 terms are being used to calculate Pi.

Increasing the number of loops to a million will change this:

There it is! A function written from scratch to calculate Pi.
To get a value even closer to Pi just increase the number of loops.

Finally, if you do wish to use Pi in python the easiest way is to use the numpy library, which has a pi constant stored.

Creating Power BI Layouts using PowerPoint

First question, why bother with layouts?
Using layouts in Power BI allows a user to make their visuals stand out better, the page looks professional and more appealing to its audience.

Second question, why PowerPoint?
The default page size in Power BI desktop is 16:9, (this trick doesn’t work for other Power BI page sizes), which is identical to a PowerPoint slide.
Therefore whatever is designed in PowerPoint will fit onto a Power BI page perfectly. Also PowerPoint is very easy to use; most people are familiar with it.

You’ll need a Power BI dashboard that you want to add a layout to. I’ll be using a dashboard displaying data on the Premier League top 50 goals scorers.

To start the creation of a layout, open PowerPoint with a blank side and add a rectangle.

Next, add glow to the rectangle, glow is found in the drawing ribbon under shape effects.

Repeat the process of adding rectangles with glow and lay them out so the visuals will fit.

If there is an important visual that needs to be a certain size you can find the size of the visual in Power BI in pixels and tailor the rectangle size to your visual.
Unfortunately PowerPoint displays sizes in cm so the numbers will need converting.
I use the following – https://www.unitconverters.net/typography/pixel-x-to-centimeter.htm

Once happy with the layout, save the slide as a PNG file. Upload this as the background in Power BI by going to format then page background, then selecting your created PNG file.

There you go!! A Power BI dashboard with its own custom layout.

Power BI Sentinel
The Frog Blog

Team Purple Frog specialise in designing and implementing Microsoft Data Analytics solutions, including Data Warehouses, Cubes, SQL Server, SSIS, ADF, SSAS, Power BI, MDX, DAX, Machine Learning and more.

This is a collection of thoughts, ramblings and ideas that we think would be useful to share.

Authors:

Alex Whittles
(MVP)
Reiss McSporran
Jeet Kainth
Jon Fletcher
Nick Edwards

Data Platform MVP

Power BI Sentinel
Frog Blog Out
twitter
rssicon