Plotting a very long time-series data

Is there a way we can plot a very long time-series data which is more than the RAM of 16GB I have. This time series is worth 30 GB.

Of course, I used go.scattergl to plot a large data via WebGL. But I am not sure there is a way I can handle data which is more than RAM?

I thought there is way I can update the already plotted layout on .html file corresponding to plot of few initial points of this time-series I have but I think it overwrites it rather than update it.

Hi @abinashsinha330,

Iā€™d highly recommend using Datashader for this purpose. See http://datashader.org/user_guide/3_Timeseries.html for the datashader documentation for time-series data.

You can then display the datashader output in plotly by following along with https://plot.ly/python/change-callbacks-datashader/. Note that this uses FigureWidget and is intended for use in the Jupyter notebook.

Hope that helps!
-Jon

1 Like