Datashader Examples

A variety of example notebooks and applications are maintained in the examples/ subdirectory of the git repository, and these will be installed somewhere on your local system when you install datashader. To get a copy of the examples in your own directory so that you can run and edit them, you can run these commands in your terminal:

1. cd ~
2. python -c 'from datashader import examples ; examples("datashader-examples")'
3. cd datashader-examples
4. python download_sample_data.py

Steps 1-3 steps will give you a copy of the notebooks and apps, and step 4 will download various datasets used by them. The total download size is currently about 2.5GB to transfer, requiring about 7.5GB on disk when unpacked, which can take some time depending on the speed of your connection. The files involved are specified in the text file datasets.yml that was copied to your directory in step 3, and instead of step 4 you can download each file individually if you prefer.

The "Census" example data is the largest file and should be the last thing to be downloaded, so you should be able to start running all of the other examples while that one completes.

Datashader itself is independent of other plotting libraries, but most of the examples also require bokeh (version 0.12 or higher) to be installed. Bokeh is available through either conda or pip:

conda install bokeh

or

pip install bokeh

Specific examples may have other dependencies as listed below.

Dashboard

An example interactive dashboard using bokeh server integrated with a datashading pipeline. Requires webargs and (on Python2) fastcache:

pip install webargs
conda install fastcache

To start, launch it with one of the supported datasets specified:

python dashboard/dashboard.py -c dashboard/nyc_taxi.yml
python dashboard/dashboard.py -c dashboard/census.yml

The '.yml' configuration file sets up the dashboard to use one of the datasets downloaded above. You can write similar configuration files for working with other datasets of your own, while adding features to dashboard.py itself if needed.

If you have less than 16GB of RAM on your machine, you will want to add the "-o" option before "-c" to tell it to work out of core instead of loading all data into memory, though doing so will make interactive use substantially slower than if sufficient memory were available.

To launch multiple dashboards at once, you'll need to add "-p 5001" (etc.) to select a unique port number for the web page to use for communicating with the Bokeh server.

Notebooks

Most of the examples are in the form of runnable Jupyter notebooks. Copies of these with all the images and output included are hosted at Anaconda Cloud. To run these notebooks on your own system, install and start up a Jupyter notebook server:

conda install jupyter

or

pip install jupyter

To start:

jupyter notebook

plotting_pitfalls

Motivation for the ideas behind datashader. Shows perceptual problems that plotting in a conventional way can lead to. Re-running it locally is usually not required, since the filled out version at the link above has the full data. If you do wish to re-run it, you will need to install the HoloViews package:

conda install -c ioam holoviews

pipeline

Step-by-step documentation for each of the stages in the datashader pipeline, giving an overview of how to configure and use each component provided. Most useful when you have looked at the other example dashboards and the notebooks below, and are ready to start working with your own data.

conda install -c ioam holoviews

nyc_taxi

Making geographical plots, with and without datashader, using trip data from the NYC Taxi dataset.

census

Plotting the 2010 US Census data, both to show population density and to show racial categories. Requires pytables:

conda install pytables

nyc_taxi-nongeo

Scatterplots for non-geographic variables in the taxi dataset.

tseries

Plotting large or multiple plots of time series (curve) data.

trajectory

Plotting a 2D trajectory.

census

Combining raster data with scatterpoint data, using the census data on race along with gridded elevation data for Austin, TX. Requires rasterio (conda install rasterio krb5).

osm

Plotting the 2.7 billion gps coordinates made available by open street map. This dataset is not provided by the download script, and the notebook is only included to demonstrate working with a large dataset. Requires dask for out-of-core operation and castra for the fast file format. The run notebook can be viewed at anaconda.org.

Amazon.com center distance

Cities in the USA colored by their distance to the nearest Amazon.com distribution center.