Issue/228/calpreddistr #291
Conversation
… utils files from Bitrateep in condition_pit_utils
…8/calpreddistr merging main branch into issue/228/calpreddistr branch as of 20220819
…putation of Bitrateep instead of that from class UnconditionPIT, addition of utils.py file from Bitrateep work
merge with main changes as of 2022.11.07
…ated predictive distribution method, test file to run the latter added
torluca
added
documentation
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## main #291 +/- ##
===========================================
- Coverage 100.00% 98.43% -1.57%
===========================================
Files 77 38 -39
Lines 5548 2559 -2989
===========================================
- Hits 5548 2519 -3029
- Misses 0 40 +40
Flags with carried forward coverage won't be shown. Click here to find out more.
☔ View full report in Codecov by Sentry. |
Merge with main branch as of 10th January 2023
…t least the class initialization is covered by pycoverage
…RAIL into issue/228/calpreddistr Merging Sam's changes
|
@sschmidt23 Biprateep is wrapping up his code into a package. This way we can get rid of hard coded stuff in RAIL and just import his external package. I will keep you updated on when I'm implementing these new changes. For the moment, I would pause the merging with the main branch. |
|
Ok, just message me again when Biprateep's code is in place and you think the code is ready for review. |
|
Hi the package is more or less ready and available here: https://github.com/lee-group-cmu/Cal-PIT , will update you in a week or so when the final version of the package is public and pip installable. |
This is a pull request to merge the implementation of the calibrated predictive distribution into the main branch.
The file that has been added is in the src/rail/evaluation folder.
I added a demo notebook on how to run DSPS (/examples/creation/calibrated_predictive_distributions_demo.ipynb).
Unit tests are still missing and taking now most of the time.
There were suggestions of modifying the code to have an estimation and an evaluation part.