[qwert-e]

> I think the development of Python and Jupyter and other less known things like Vega are much more interesting.

In that case you may be interested in Dash (dash.plot.ly). It’s a free and open source library that you can use to create dashboards online with Python only.

[ZeroCool2u]

We love Dash on our team, anything more than a Tableau dashboard goes into Dash. You can basically just treat it as a Flask app.

We write our back ends with FastAPI[1], which is usually just a wrapper around our ML models. Then serve both Dash and FastAPI with gunicorn. The backend is provided the uvicorn[2] worker class with the gunicorn -k arg[3] to greatly increase the speed as well.

For personal projects you can use this stack in GCP's AppEngine standard environment to basically host your (relatively low traffic) apps for free.

1. https://fastapi.tiangolo.com/ 2. https://www.uvicorn.org/ 3. https://docs.gunicorn.org/en/stable/run.html#commonly-used-a...

[geophph]

Another +1 for Dash!

[jbny_pi]

+1 for Dash. It's very good.

[ruoyucad]

plotly aint exactly free, you get 25ish free plot and then you gotta pay.

[nicolaskruchten]

That's not totally accurate... If you use our SaaS Chart Studio product then yes, but otherwise (i.e. in 'offline' mode) you can use our Python, R and Javascript libraries as much as you like: they're MIT licensed.

[nicolaskruchten]

PS: and Dash, which is what the GP was talking about, is totally free and unrelated to Chart Studio :)

[stonecharioteer]

I'd like to add that Dash helped me grok React far quicker than I'd expected. I code in a lot of React now and I'm only here because of Dash. Thank you for your hard work!