ModelOriented / ModelOriented/DALEX
Wish: allow model_profile to specify only some variable_splits (python)
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 1.5k
- Forks
- 172
- PR merge metrics
- No merged PRs in 30d
Description
I am using dalex to plot multiple partial dependence plots in Python. It is a very useful package!
For some numeric features (especially those with outliers), I would like to specify the evaluation grid using option variable_splits in model_profile. However, if I do this, I need to specify a grid for all other features as well. Ideally, the function would use the keys in variable_splits only to overwrite the default. For those features without specified variable_splits key, it would resort to the default.
In the following case: the left-most feature should stop at a maximum of 10. The range for the other feature is fine though:
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at the model_profile entry point and trace how variable_splits is applied to feature grids. Verify the partial dependence plot case where one feature has a custom maximum of 10 while unspecified features retain their defaults; the work is done when partial grids no longer need to be supplied for every feature.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- data-visualization, machine-learning
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100