Black Lives Matter. Please consider donating to Black Girls Code today.

Question about a hosting a Dashboard

Hi guys,

I am working with Python and Matplotlib, Pandas, Plotly etc since some month. For my bachelorthesis i would like to create a poject with plotly and dash. Later then the documentation about it.


A guy has written a script with an knn algorithmus which is taking some values, with these values the algorithm is creating some csv files and from these csv files the plotly charts have to appear in different charts (plotly charts).

My Question:

I want to build a website which is hosted by my university, i wantet to put the algorithm script and the input csv files on this webserver, in the website i thought of a value section where i can set the input values for the algorithm (logic and script of algorithm is not my scope) with the output (csv files from algorithm) i wanted to display the charts from plotly in an iframe on the right site

like this example here
(but on right site 4-8 different charts (bar, gantt, line …)

I have 9 weeks from middle of february and some experiences with plotly, so the charts should not be the problme, i have concerns about the left part, input values and things like that.

Is this possible to implement these things in some weeks? I have to do a proof of concept, so would be nice if you let me know what you think about it, which librarys i need (maybe django) and if someone has a website with a similar or nearly similar example for me (which is not hosted by dash or plotly website) because i found nothing.

thanks for your help

have a nice week

Hi @0000,

  • for writing the code of the Dash app, the Dash documentation is quite pedagogical and reading through the Dash tutorial on should get you started quickly, all the more if you have some experience with the plotly graphing library
  • as for deployment, you could first write a minimal app and deploy it to heroku to understand how deployment works (see On your university machine, you would need to install the required dependencies and configure the Flask server
  • Dash is enough to write a full app, so in principle you don’t need to include it in another application like Django, but some documentation can be found here , plus some other discussions on this forum (Google search “embedding plotly dash application” for example or “plotly dash django”)

So my advice would be to start small and build your project gradually (local app, then heroku, then configure the deployment at your university, then refine the app).

So it sounds possible, thank you very much. I will read into the documentation.