Skip to content
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

Rank Widget Improvements #1547

Closed
BlazZupan opened this issue Sep 7, 2016 · 1 comment · Fixed by #1560
Closed

Rank Widget Improvements #1547

BlazZupan opened this issue Sep 7, 2016 · 1 comment · Fixed by #1560
Assignees

Comments

@BlazZupan
Copy link
Contributor

Orange version

3.3

Expected behavior
  1. Add “Scoring” box. Shows scoring methods, full name, with checkboxes to decide if the column associated with the scoring method is displayed. Like in Orange2, but plain, with no parameters (always use defaults). Content (classification or regression) based on type of input data set (default, when no data, shows for classification). By default, following methods are checked: Classification: Gain Ratio, Gini Index, Regression: Linear regression, RReliefF (there are no others for now).
  2. Scores should have a full name (shown with the checkboxes) and shorter name (column headers).
  3. Scores should compute when the scoring method is enabled (checked). If the user unchecks a method, keep the scoring method in cache.
  4. Selection of attributes is context dependent (applicable only if attributes are selected manually, not with None, All or Best ranked).
  5. Adding scoring methods to the input of the widget only adds a new scoring column (does not alter the content of the box with checkboxes). These columns are always displayed.
  6. Implement progress bar (with fat data sets, feature scoring takes quite some time).
  7. Consider implementing the score in the background, check how this is done in Test & Score (this feature is optional, but may take substantial time if we postpone it after other issues with this widget are fixed).
Actual behavior

Currently the widget displays scores for all the measures. For classification, there are many, and can be quite confusing for the user. Plus computing them all can take substantial time.

@ajdapretnar
Copy link
Contributor

Please see also #1495 for this issue. Tnx!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants