Skip to main content
Solved

How to include an image created in the Machine Learning Workbench into a Studio sheet

  • July 28, 2023
  • 6 replies
  • 23 views
A
Level 2

As the title says I want to create a plot that cannot bed created in Celonis (as far as I have researched and tried). The workaround was to create a script on Machine Learning Workbench that uses the information from the KPI's but I don't know how to include the plot created with matplotlib into the sheet devilsquadrangleFORUM 

And then to appear somewhere in my dashboard

 

dashboard 

Thanks!

Best answer by janpeter.van.d

HI @alvaro.miran,

 

In studio analyses, this is hard to establish. These images are hosted in a back-end storage and can be obtained using PyCelonis, but not published (as far as I know).

In studio views however, this would be possible in two steps:

  1. Publish the image on a hosting site with a public link, for instance Dropbox. This can be done using the Dropbox API. Be sure to overwrite the same image every time an update is available.
  2. Link to this image in your view. Every time the view loads the latest version is obtained.

 

Cheers,

Jan-peter

 

6 replies

S
Level 8
  • sunny.thind
  • Level 8
  • July 31, 2023

Hi , If you are trying to embed the chart am not sure if possible, but maybe you could use the _ Run ML Notebook.

https://docs.celonis.com/en/machine-learning-components.html#UUID-f050a189-0bad-0494-4c07-2bee70d853b2_id_MachineLearningComponents-RunMLNotebookComponent

 

If you would just like to embed the image then use design component of "image".

https://docs.celonis.com/en/image.html

 

G
Level 12

I wonder if with pycelonis you can manipulate a component of an analysis. That way you could, theoretically, create or modify the url associated with the embedded image in the analysis from the same notebook that creates the image.

Maybe you will need also to export the image to jpg and upload to some repo that is accesible from the studio

 

 

janpeter.van.d
Level 12
Forum|alt.badge.img+26

HI @alvaro.miran,

 

In studio analyses, this is hard to establish. These images are hosted in a back-end storage and can be obtained using PyCelonis, but not published (as far as I know).

In studio views however, this would be possible in two steps:

  1. Publish the image on a hosting site with a public link, for instance Dropbox. This can be done using the Dropbox API. Be sure to overwrite the same image every time an update is available.
  2. Link to this image in your view. Every time the view loads the latest version is obtained.

 

Cheers,

Jan-peter

 

A
Level 2
  • alvaro.miran
  • Author
  • Level 2
  • August 1, 2023

I wonder if with pycelonis you can manipulate a component of an analysis. That way you could, theoretically, create or modify the url associated with the embedded image in the analysis from the same notebook that creates the image.

Maybe you will need also to export the image to jpg and upload to some repo that is accesible from the studio

 

 

Thanks, thats what I'll try to do with a service that allows me to connect through their API.

A
Level 2
  • alvaro.miran
  • Author
  • Level 2
  • August 1, 2023

HI @alvaro.miran,

 

In studio analyses, this is hard to establish. These images are hosted in a back-end storage and can be obtained using PyCelonis, but not published (as far as I know).

In studio views however, this would be possible in two steps:

  1. Publish the image on a hosting site with a public link, for instance Dropbox. This can be done using the Dropbox API. Be sure to overwrite the same image every time an update is available.
  2. Link to this image in your view. Every time the view loads the latest version is obtained.

 

Cheers,

Jan-peter

 

Thanks @janpeter.van.d 

 

I'll try this approach. Thanks

B
Celonaut
  • bryce.paque
  • Celonaut
  • August 3, 2023
Hi Alvaro, Thank you for your inquiry. After review, there is no ability to use PyCelonis and the ML Workbench to upload a new image to an image asset within a studio analysis. We highly suggest reaching out to your Celonis Value Engineer to open a feature request for this chart type and our development team will take this into consideration. If you need assistance identifying your value engineer, please reach out to anyone on your account team or open a support case via CeloPeers and a support engineer can provide this contact information. Please be sure to include your team URL in the case description. We look forward to hearing from you soon! Have a great weekend ahead.
Community Code of ConductAccessibility statement