Any 'pretty' data visualization libraries

问题:

There are plenty of 'pretty-printing' visualization libraries for Javascript. E.g. those listed here.

Googling for 'python visualization libraries' only turns up stuff like VTK and mayavi, which are primarily more for no-nonsense scientific use.

So, do you know of any Python libraries similar to those Javascript ones in the above link? I particularly like the Javascript Infovis Toolkit.

回答1:

For Python there really isn't "one viz library to rule them all". There are different libraries and toolkits for different purposes. For graphs in Python you may find igraph useful. For other types of scientific or data visualizations matplotlib is also good.

回答2:

Here's a new port of R's ggplot2 over to python. Looks very slick!

https://github.com/yhat/ggplot/

More info here: http://blog.yhathq.com/posts/ggplot-for-python.html

---

Also have a look at Seaborn, described as "Improved matplotlib for statistical data visualization": https://github.com/mwaskom/seaborn

The Seaborn examples are pretty slick:

Plotting Complex Linear Models

Visualization Distributions

Time Series Visualizations

回答3:

Check out Bokeh at pydata

source code here

回答4:

There's PyCha for charts.

回答5:

are you looking for graph software?

Checkout http://www.graphviz.org/Gallery.php. it has python bindings.

回答6:

I second matplotlib. Also Chaco. It's not Python, but for really quick heatmaps, go to OpenHeatMap.com

回答7:

Networkx looks to be the best-documented of the graph libraries I've seen. I think it interfaces with Matplotlib for visualization.