Hacker News Reader

Comment by hprotagonist | Hacker News Reader

Of the "we really don't like matplotlib syntax, really do want to live in dataframes all the time, so we wrapped mpl and pandas" plotting libraries, plotnine has impressed me the most by far:

https://plotnine.readthedocs.io/en/stable/

plotnine is an implementation of a grammar of graphics in Python, it is based on ggplot2. The grammar allows users to compose plots by explicitly mapping data to the visual objects that make up the plot.

View on HN

I'm really fond of https://altair-viz.github.io/ which wraps and translates into https://vega.github.io/vega-lite/.

The interactivity and cross-filtering possibilities is really good when you have many datapoints.

From vega-lite I miss that they did not develop object constancy (https://bost.ocks.org/mike/constancy/) animation possibilities since they're building on top of d3.js.

I'm a big fan of Altair and have used it for demos in the past. However, restrictions such as a 5,000-row limit on dataframes and usage only inside notebooks pushes me to other solutions like Plotly.

BTW, the 5000 restriction is more a "guide" for structuring your data - you can turn it off.

Also, Altair plots can certainly be used outside of notebooks! Happy to share some examples..

Please, I couldn't find a straightforward configuration to do that

When I was in the business of making plots, ggplot2 was by far my favourite. It seemed odd at first, but eventually it clicked and it felt great. Once you grok the grammar of graphics you'll find you can really quickly express all kinds of plots from any given dataset. I would routinely impress people with how quickly I could visualise data.

It looks like this one is designed specifically to be easy to learn and remember. It looks very useful to me for people who make plots less frequently.

What makes you choose this over something like seaborn?

Not parent, obv. But the syntax (or grammar if you like) although jarring at first, is starting to grow on me. I am not fully onboard yet, but it is an interesting idea.

Reply on news.ycombinator.com