Reusing the RVAT Reynolds number dependence experiment

This repo is a Calkit project that reuses the RVAT Reynolds number dependence experimental results. The goal here is to explore ideas to make reuse easier.

How this was setup

To be able to run these steps, you'll need to install Calkit, create a token for your local machine, and create a free account on calkit.io.

1. Created a new project on calkit.io.
2. Cloned project locally with calkit clone {git repo url}.
3. Ran git submodule add https://github.com/UNH-CORE/RVAT-Re-dep rvat-re-dep.
4. Ran

   calkit new conda-env \
       -n reuse-rvat-re-dep \
       python \
       matplotlib \
       jupyter \
       pandas \
       numpy \
       seaborn \
       pip \
       h5py \
       scipy \
       --pip pxl==0.0.10 \
       --pip progressbar \
       --pip calkit-python==0.8.5 \
       --stage check-conda-env

   to create a Conda environment and pipeline stage to check it's up-to-date.
5. Added a pipeline stage to install the Python package from rvat-re-dep in local mode. In dvc.yaml this looks like:

   install-rvat-re-dep:
     cmd: calkit runenv pip install -q -e ./rvat-re-dep
     always_changed: true

6. Added a script to simulate running a CFD simulation, and added a stage to run it and generate a dataset.
7. Added a pipeline stage notebook cell to notebook.ipynb to plot the simulated versus experimental power coefficient.

Getting started

After installing Calkit and setting your token in its config, clone this project repo with:

calkit clone --recursive https://github.com/petebachant/reuse-rvat-re-dep

Then run the pipeline with:

calkit run

If you'd like to generate more figures, data processing, etc., add stages to the pipeline in dvc.yaml.

If you'd like figures to be visible on calkit.io, add them to calkit.yaml.