graph_objs.Scatter slow for large-ish data sets

[sbowman-mitre]

Hi, I'm using Scatter to plot ~300K data points and finding slow rendering performance. Is this directly related to Plotly.js #741 or is this independent?

[chriddyp]

Try go.Scattergl instead: it's the WebGL version (instead of SVG). There are some more links and discussion the second part of the "Performance" chapter about graphs: https://plot.ly/dash/performance

[sbowman-mitre]

Yes! That is much improved performance! Thanks.

[thusithaC]

Are there 3D alternatives to scattergl and pointcloud?

[chriddyp]

@thusithaC - Scatter3d (https://plot.ly/python/3d-scatter-plots/) is the plotly WebGL enabled 3D scatter plot

[thusithaC]

@chriddyp Thanks. In that case I guess no easy way of solving the issue i have raised in plotly/plotly.py#893 .

[chriddyp]

@thusithaC - There isn't. I'd also recommend checking out the issues or reposting your issue in https://github.com/plotly/plotly.js - that's the project for the underlying graphing library that Dash and Plotly.py use.

We'd definitely like to improve the performance of all of these chart types but they are big projects and we're looking for institutional or corporate sponsors to fund the work (See https://plot.ly/products/consulting-and-oem/). Thanks to one commercial sponsor, we're in the process of revamping the 2D scattergl plot in plotly/plotly.js#1869

[jvdd]

If you are working in Python, it might be interesting to consider plotly-resampler when you want to visualize larger datasets.

This extension adds resampling functionality to Plotly figures (by running an under the hood dash app), allowing to visualize tons of datapoints while being responsive.

Hope this might help!