I don’t have to convince you why we need an interactive dashboard. But most people don’t know that they don’t have to buy expensive licenses of Tableau<\/i> or PowerBI<\/i>. You don’t have to enroll in a JavaScript<\/i> course, either.<\/p>\n\n\n\n
Dash<\/i> apps allow you to build interactive dashboards purely in Python. Interestingly, it could reach heights that popular BI platforms can not.<\/p>\n\n\n\n
Unlike other tools to create dashboard web applications, you can host Dash apps on your servers and terms. You can use this tool for various use cases, from a key performance indicator dashboard for your employees that changes once a month to a real-time network monitoring dashboard.<\/p>\n\n\n\n
\nI strongly recommend the book, Interactive Dashboards and Data Apps with Plotly and Dash<\/b><\/a> because it gives an in-depth understanding of interactive dashboard design.<\/p>\n\n\n\n
Other resources about Dashboards and Data Visualization:
The Big Book of Dashboards<\/b><\/a>
<\/b>Storytelling with Data<\/b><\/a><\/p>\n<\/blockquote>\n\n\n\n\n\n\nGrab your aromatic coffee <\/a>(or tea<\/a>) and get ready…!<\/p>\n<\/div>\n<\/div>\n<\/div>\n\n\n\n
Why Dash? Why not Tableau, Power BI, or some JavaScript library?<\/h2>\n\n\n\n
BI Platforms such as Tableau<\/a> and PowerBI<\/a> do a fantastic job. It allows even non-technical managers to do data exploration themselves<\/a>. I don’t have complaints about them.<\/p>\n\n\n\n
They are excellent tools for performing analysis on read-only datasets. But in a large data science project<\/a>, you’ll have to perform complex actions. For instance, you must trigger a backend function and start the model retraining.<\/p>\n\n\n\n
In such cases, my best solution was to build a web app from scratch<\/a>. JavaScript data visualization libraries such as HighCharts<\/a> are excellent tools for this. They have callbacks for almost every possible user action. I use them to send data back to the server and control it better.<\/p>\n\n\n\n
Related:<\/b> How to Make a PDF Text-to-Speech Reader in Python<\/i><\/b><\/a><\/p>\n\n\n\n
But this wasn’t a walk in the park. My data science team is exceptional in Python and R but not JavaScript<\/i><\/a>. N<\/i>ot on web frameworks such as Django<\/i><\/a> either. And that’s not enough; to build modern web apps<\/a>, you need frontend<\/i> web frameworks such as React <\/i><\/i>.<\/p>\n\n\n\n
As we progressed, we realized the harsh truth. Every new technology in our stack inflates the difficulty exponentially.<\/p>\n\n\n\n
And we were fortunate to find Dash<\/a>\u2014a pure Python open-source web dashboard framework. If you’re comparing Plotly dash vs. React, Dash is built using React, so you don’t have to do the heavy lifting for yourself.<\/p>\n\n\n\n
Now, this is how I create my own web dashboards in Python. If you’re looking for a lightweight alternative, check out Streamlit<\/i><\/b><\/a>.<\/p>\n\n\n\n
Read along if you need flexible, custom dashboards entirely in Python. There’s also an amazing extension to this package that lets you create Dash apps inside Jupyter Notebooks<\/a>. See its repository for more Jupiter dashboard examples<\/a>.<\/p>\n\n\n\n
Streamlit vs. Dash vs. Gradio<\/h3>\n\n\n\n
Both tools are great for building a dashboard web application. But streamlit is more focused on building quick control panels for ML projects. Plotly is excellent for building cool dashboards at ease.<\/p>\n\n\n\n
Another excellent alternative to creating a dashboard web application is Gradio<\/a>. With gradio, you can even create a dashboard in Python Jupiter Notebook.<\/p>\n\n\n\n