These are the slides and notebooks for a guest lecture about Dask given to Harvard CS207 on 11/30/16.
Slides can be viewed here.
The dask_array
and dask_dataframe
notebooks should be runnable locally,
while the dask_dataframe_cluster
notebook relies on some local data that
isn't public.
- Clone the repo
- Install
conda
$ conda create -n harvard_dask python=3.5
$ source activate harvard_dask
$ conda install dask distributed bokeh jupyter -c conda-forge
For a more information, please check out the following resources:
If you have any questions or wish to contribute (we love new contributors), please feel free to reach out via github issues or via gitter.