Metadata-Version: 2.1
Name: person_linkage_case_study
Version: 0.1.0
Summary: Person linkage case study for PyPI.
Home-page: https://github.com/ihmeuw/person_linkage_case_study
Author: IHME Simulation Science Team
Author-email: zmbc@uw.edu
License: BSD-3-Clause
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: pandas
Requires-Dist: pyarrow
Requires-Dist: numpy
Requires-Dist: matplotlib
Requires-Dist: pseudopeople
Requires-Dist: splink
Requires-Dist: jellyfish
Requires-Dist: snakemake
Requires-Dist: papermill
Requires-Dist: ipython
Requires-Dist: ipykernel
Requires-Dist: pulp<2.8
Provides-Extra: dask
Requires-Dist: pseudopeople[dask]; extra == "dask"
Requires-Dist: dask_jobqueue; extra == "dask"
Requires-Dist: bokeh!=3.0.*,>=2.4.2; extra == "dask"
Provides-Extra: spark
Requires-Dist: pyspark==3.4.1; extra == "spark"
Provides-Extra: dev
Requires-Dist: jupyterlab; extra == "dev"
Requires-Dist: nbdime; extra == "dev"
Requires-Dist: black[jupyter]; extra == "dev"
Requires-Dist: pseudopeople[dask]; extra == "dev"
Requires-Dist: dask_jobqueue; extra == "dev"
Requires-Dist: bokeh!=3.0.*,>=2.4.2; extra == "dev"
Requires-Dist: pyspark==3.4.1; extra == "dev"

# Person linkage case study (for PyPI)

**If you have access to GitHub, we recommend installing from https://github.com/ihmeuw/person_linkage_case_study.**
**This package is provided only for the convenience of users who cannot access GitHub.**

## Getting started

Before you start, you'll need to have Python installed.
If you intend to use Spark, you'll also need Spark **3.4** to be installed
on your system.

First, make a directory to use for development.
Inside this directory, create and enter a virtual environment for this project:

```console
$ python -m venv .venv
$ source .venv/bin/activate
```

and install this package:

```console
$ pip install person_linkage_case_study
```

Your new working directory for interacting with the case study will
be *inside this venv*. Specifically:

```console
$ cd .venv/lib/python3.*/site-packages/person_linkage_case_study
```

To test that things are set up correctly, try running:

```console
$ snakemake --forceall
```

## Installing extras

You can install dependencies for using Spark and Dask by installing with those extras:

```console
$ pip install person_linkage_case_study[spark,dask]
```

## Spark without Singularity

The case study typically runs Spark in a Singularity container. If you cannot use Singularity,
edit the file at `person_linkage_case_study/profiles/default/config.yaml` to say
`use-singularity: false` instead of `use-singularity: true`.
As mentioned above, this will rely on Spark **3.4** being installed on your system.
Specifically, it must be installed at `/opt/spark`.

## What's next?

For further instructions, see the README.md bundled within the package (at the top level
of your working directory as described previously).
Aside from how to use Spark, and installing with `pip install person_linkage_case_study`
instead of `pip install -e .`, all the instructions in that README apply to a PyPI installation.
