Skip to content

Commit

Permalink
Update collections/_posts/2024-10-20-tabular-explainability.md
Browse files Browse the repository at this point in the history
Co-authored-by: Nathan Painchaud <[email protected]>
  • Loading branch information
olivier-bernard-creatis and nathanpainchaud authored Oct 29, 2024
1 parent b0198fc commit a40f2f8
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion collections/_posts/2024-10-20-tabular-explainability.md
Original file line number Diff line number Diff line change
Expand Up @@ -143,7 +143,7 @@ prediction was defined as that which corresponds to the highest attention value
<img src="/collections/images/tabular_explainability/explanation_distribution.jpg" width=700></div>
<p style="text-align: center;font-style:italic">Figure 4 - Best concept group distribution per method</p>

* For each method, the proportion of samples who considered each concept group to be the best is indicated
* For each method, the proportion of samples that considered each concept group to be the best is indicated
* For each dataset, the number of incorrectly classified samples is less than 5%, which has no impact in the overall distributions
* MLA appears to take more groups than SA and SH into account when identifying differences among samples

Expand Down

0 comments on commit a40f2f8

Please sign in to comment.