Exploratory analysis of Bayesian models with dashboards.
This project brings dashboards to ArviZ enabling users to compare different visualizations in the same view and interact with them.
We will use mamba
to create a virtual environment where we will install a development version of
ArviZ Dashboard. The first step is to follow the instructions here
https://github.com/conda-forge/miniforge#mambaforge
to install the correct version of mamba
for your operating system. Note that if you have
conda
installed already, you can exchange the command mamba
for conda
with the same results.
mamba create --name arviz-dashboard pip python
mamba activate arviz-dashboard
Once the virtual environment has been created, and you have activated it with the above commands,
you can install the development requirements for arviz_dashboard
with the following commands.
git clone https://github.com/arviz-devs/arviz_dashboard
cd arviz_dashboard
pip install --editable '.[dev,examples]'
Once the package has been installed, we need to install the pre-commit
hooks used for maintaining
code hygiene. Run the following commands to set up the required pre-commit
hooks for development.
pre-commit install
When you commit your changes to your branch, pre-commit
will install the tools defined in the
config file, and check give feedback about required changes in order for the push to pass linting
and formatting tests.
If you add a new hook to the .pre-commit-config.yaml
file, run the following command in order to
check if your hook is working against all the files.
pre-commit run --all-files
TBD
pytest
Dashboard usually includes multiple visualizations with different purposes. If you have any problems with certain visualizations, you can find more explanations for the ArviZ visualizations in ArviZ and Bayesian Modeling and Computation in Python