Adding feature importance visualizations #48
Merged
+654
−0
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
jenna-tomkinson
approved these changes
Nov 13, 2023
There was a problem hiding this comment.
Some comments (applies to both plots):
- The legends are a bit small and very close to overlapping on the neighboring plot.
- Can you add the plot labels (e.g., A, B, C)?
- Just for my own knowledge, are these heatmaps what we expect to see? What makes it different than the shuffled baseline which doesn't look all that from the final model plots?
- I would recommend increasing the font size for these plots if possible. It might not be possible, but I think given the amount of whitespace that you couldn't remove, then might as well increase the font. No worries if that isn't possible.
There was a problem hiding this comment.
Thanks for the feedback Jenna! (sorry for my delays in response)
- Good point. I've decided to update how I generate this plot, given it's lesser importance in the overall story now. I will no longer compile within R, but will choose to compile within the supplementary figure document itself. This is going to make the plot look nicer and will be easier :)
- There are certain conditions where panel labels are necessary, but I do not believe this is one of them. I think it is worth adding panel labels if individual plot components need to be referred to individually. We won't do this in the text - all we say is "In general, most phenotypes use different features to make predictions indicating that each phenotype can be explained by a unique set of nuclei measurements (Supplementary Figure 4). "
- This is a great question! The plots are actually pretty different shuffled vs. real. The big difference here is the color scale! I need to make a change so that the scale is consistent across heatmaps. This will drive home the difference :)
- Yes, totally agree. Will work on it. Thanks!
|
All set! Sup figure updated - I think this looks much better (see sup fig document), thanks again! Will merge now. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Heatmaps of multiclass model coefficients. Note that I could not figure out how to remove the extra whitespace margins around the figures after two hours of trying. I think it is good enough as is (especially for a sup figure)!