Warn when there is no data in a split
Open
trivial
- Dominant language
- Jupyter Notebook
- Stars
- 201
- Forks
- 62
- PR merge metrics
- No merged PRs in 30d
Description
_From @thcrock on July 25, 2017 18:44_
I believe there are still no checks for this. When catwalk trains the models, it does check and skip these but this should really be flagged when we're building the matrices.
https://github.com/dssg/timechop/issues/15
_Copied from original issue: dssg/architect#20_
Contributor guide
Research direction
Read the linked timechop issue and trace how catwalk checks splits and how matrices are built. Identify the matrix-building entry point and existing tests or checks around empty splits. Done means a split with no data is flagged while matrices are being built, rather than only skipped during model training.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100