NREL / OpenOA

This library provides a framework for assessing wind plant performance using operational assessment (OA) methodologies that consume time series data from wind plants. The goal of the project is to provide an open source implementation of common data structures, analysis methods, and utility functions relevant to wind plant OA.
https://openoa.readthedocs.io/
BSD 3-Clause "New" or "Revised" License
194 stars 63 forks source link

Feature/149 windfarm map #153

Closed charlie9578 closed 3 years ago

charlie9578 commented 3 years ago

Adding a wind farm plot as described in issue #149. I've also updated the sphinx documentation and requirements to facilitate bokeh plots, and included the plot as an example in the 00_toolkit_example notebook.

codecov-commenter commented 3 years ago

Codecov Report

Merging #153 (9717359) into develop (5449fcc) will not change coverage. The diff coverage is n/a.

Impacted file tree graph

@@           Coverage Diff            @@
##           develop     #153   +/-   ##
========================================
  Coverage    70.78%   70.78%           
========================================
  Files           23       23           
  Lines         1595     1595           
========================================
  Hits          1129     1129           
  Misses         466      466           

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 2fd44bf...9717359. Read the comment docs.

charlie9578 commented 3 years ago

I seem to sometimes fail the "docs/readthedocs.org:openoa" tests when creating a pull request, but they all pass locally. In this case I think its due to the packages, and potetially the docs test only installing packages in the sphinx/requirements.txt, rather than those in the main requirements.

Is there a way to also run the test on my forked github to check it passes before pushing to the NREL repo?

jordanperr commented 3 years ago

Hi Charlie, you can make an account with readthedocs.org and point it to your own fork of OpenOA. I'm happy to walk you through the configuration. Readthedocs installs only a subset of the dependencies which are required to build the documentation. Historically, a full installation of OpenOA was not needed for the documentation.

CLAassistant commented 3 years ago

CLA assistant check
All committers have signed the CLA.

jordanperr commented 3 years ago

Hi Charlie, In addition to rob's suggestion, I added some more documentation about the "fill_color" option to the plotting function. I also loosened the bokeh version to 2.3.*, since fixing version numbers precisely can make the installation too brittle. I also reverted some changes in the tags-to-pypi and paper/* files. These must have been remnants of a hotfix that didn't make it all the way into your branch.

The tests are passing for me locally, but for some reason they are failing through gh actions. I am not sure why this is happening and am investigating now. Sorry to break your branch!

Update 1: I was able to reproduce the gh action test failure locally by re-installing OpenOA in a new environment. So there may be an incompatibility with an update to a dependency. This is likely unrelated to the current pull request. Don't worry about the gh-actions test failure for now.

Update 2: Looks like the test failure was related to Panda's newest version, 1.3.0, which was released just last week. I set the maximum pandas version to <1.3 in requirements.txt and created issue #167 to fix this.

charlie9578 commented 3 years ago

Hi Charlie,

It looks like there is no need for the bokeh_plot.py module because you enabled the bokeh_plot extension functionality in conf.py, so it would probably be best to remove that file. Otherwise, this is a great addition, especially with the docstring plotting demonstrations.

Thanks!

Hi @RHammond2, I've just tested removing the bokeh_plot.py module and like you say it seems to work fine. I've pushed the update and removed a couple of debugging print statements to clear things up as well. Hopefully all good now. Thanks, Charlie