Thomas Schill
Thomas Schill
- kdqtree and HDDDM have different default behaviors: static window for the former, updating window for the latter. This choice was because kdqtree's transform is expensive to run. - It...
- Confirm that current pipeline runs the script versions to check for no error - Freeze and/or do not execute very expensive examples (LFR, wilds) - Should we use small,...
Years 2010 and 2013 technically contain drift relative to the prior years, since they revert to the original distribution. The `drift` column has those years marked as `False`, though. This...
This is a placeholder task for the implementation of DetectA. - [ ] Commit code to new branch - [ ] Make a PR - [ ] Before merge @tms-bananaquit...
**Overview**: Currently, `KDQTreePartittioner` behavior on datasets with columns containing categorical/n-hot encoded/ordinal data will be volatile. Fixing this will generalize `KDQTreePartitioner` to mixed-type datasets. **Details**: For example, if one column in...
- `datasets` includes a couple of csv files that have to be included with the library to work properly right now, rather than being downloading as needed/being generated via script....
- In the Maciel 2015-style ensemble, some detectors have their output ignored while they "wait" and learn a potential new concept. - In MD3, the `waiting_for_oracle` status also looks like...
MD3's API differs from the other detectors, because it is intended for a semi-supervised context. It may be desirable to address some or all of these: - Currently, MD3 has...