-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Inference verification plot #14
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks! This will be very useful when plotting/evaluating various input formats in a standalone script. The two comments wrt 23
and 42
should be adressed before merge.
I think this PR would benefit from an additional script (slurm, CLI) for the user to interact with the standalone plotting. Some guidance on how to use it should then be added to the README. Would you have time to add such functionality?
- Arguments of function were in wrong order - No need to return the last plot but free up memory
- slight changes to the save_path
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good! I made some smaller changes, as the code was not running on Balfrin initially. Please review my changes and if you agree squash commits and merge PR.
Thanks 🙏
Purpose
Implements an extra-feature of the vis module to perform standalone plotting of the predictions. This allows for the verification of the inference output which outputs a Numpy array (see PR#13). Users can specify file path for saving and the channel (variable and level) to be plotted over the time range available in the inference output.
Code changes:
vis.py
: Addition of the standalone functionverify_inference
, described in the purpose section of this descriptionweather_dataset.py
: Addition of a split and specific datatype called "verification" which handles a numpy array instead of .zarr files, the current datatype supported by the model. The file is loaded as such, and, since it is backtransformed before storage, no specific transformation of the file is needed to pre-process it before plotting.cli_plotting.py
: Creation of a command line entry point to run the plotting on inference data as a standaloneREADME.md
: adapted the readme according to this new functionalityChecklist
Before submitting this PR, please make sure:
Review
For the review process follow the guidelines at Checklist