Datashader is a graphics pipeline system for creating meaningful representations of large amounts of data. It breaks the creation of images into 3 steps:
-
Projection
Each record is projected into zero or more bins, based on a specified glyph.
-
Aggregation
Reductions are computed for each bin, compressing the potentially large dataset into a much smaller aggregate.
-
Transformation
These aggregates are then further processed to create an image.
Using this very general pipeline, many interesting data visualizations can be created in a performant and scalable way. Datashader contains tools for easily creating these pipelines in a composable manner, using only a few lines of code.
The project is under active development, and all the code and documentation is subject to frequent changes.
Datashader is available on most platforms using the conda package manager,
from the bokeh channel:
conda install -c bokeh datashader
Alternatively, you can manually install from the repository:
git clone https://github.com/bokeh/datashader.git
cd datashader
conda install -c bokeh --file requirements.txt
python setup.py install
Several examples can be found in the examples directory.
Additional resources are linked from the [datashader documentation] (http://datashader.readthedocs.org), including papers and talks about the approach.
