power bi cool visuals

Synoptic visual

Calendar heatmap

Heatmap

https://powerbidocs.com/2020/12/04/creating-a-table-heatmap-in-power-bi/

The custom visuals 'table heatmap' and 'heat streams' are not really good heatmap. They have to choose a category variable and a fixed number of Y columns. The category can change, e.g. Shear start time. The Y columns are fixed, e.g. 0-149 supports. However, the category variable is always the X axis, and seems no way to swap the X and Y. Also you need to pivot the dataset to have all the 0-149 supports as columns. Each of the columns needs to be drags in to the visual manually. Painful.

Matrix visual

Matrix visual provides a better option for heatmap.

It doesn't even need to pivot data. Simply specify the rows, columns and the values. The X, Y axes are done.

About the color, click at the drop down arrow of the values variable. Go to conditional formattting for background color.

Choose the color for the lowest, center and highest values (selected diverging).

Do the same for the font color, so the text will be hidden, and only the color shows.

Then turn off the subtotals for rows and columns, see screenshot below.

However still need to adjust the cell size, the labels, etc.

Trouble is the columns are sorted alphabetically... so may need to manipulate a Sort column

Python Visual

This basically receives the data table as a dataframe, and you can run your own python script to generate visualisation using e.g. matplotlib, so plt.show() will display the image onto the visual.

Just drag the required columns into the Values list. The values will be automatically added into a dataframe called 'dataset', and then just go from there.

Note only a limited set of python libraries are supported in Power BI, check microsoft power bi website for details.

Note the number of rows supported is about 2millions in every table, so the dataframe inherits the same limit.