We will use NYC taxi dataset
, data from January, 2015. This is modified
dataset from https://s3.amazonaws.com/datashader-data/nyc_taxi_wide.parq
used in holoviz
ecosystem https://holoviz.org/. Column names are shortened.
For workshop we will need only the sample of 310000 records
(data/nyc_taxi_sample.parq
). Full modified dataset is available at https://feelmath.eu/Download/nyc_taxi.parq.
For running the notebooks, modules panel, ipyleaflet, polars, pandas, pyarrow, fastparquet, plotly
need to be installed.
Docker image with all workshop files can be pulled by
docker pull misolietavec/daworkshop
. The Dockerfile
in this directory is
for creating such (or similar) images. Running with
docker run -p 3500:8888 misolietavec/daworkshop
will give you Jupyterlab application at http://127.0.0.1:3500/lab
.