-
Notifications
You must be signed in to change notification settings - Fork 11
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
Plot Median AE instead of mean, add mean to metrics #439
Conversation
This will shift the points to the left, so we need to re-upload them all. We (I) did this once. I could do it again if needed. |
@mlocardpaulet We must wait with the merge, until we set up a way to recompute all the points? |
Yes. I think that this is necessary. |
Would like to get this up and going ASAP. Should be done before Christmas. |
In the end, the idea is to:
Now the mean AND median are calculated at each submission, but we do not have the median values for points that were uploaded before now. There is an issue opened for setting up the automatic re-running of all points already submitted (while keeping their unique ID and associated comments): #470. Since the "main" plot will not change, maybe we do not have to wait for the issue to be fixed. The only thing is that until we are able to re-run everything there will have points missing in the plot with the median? |
Mentioned in original issue, but relevant here too: |
Saving of the raw data is implemented. Only thing that remains is a notebook that allows for automatic reuploading. But I think we should do that manually for now. Are you OK with that? Automatic reupload should be done from: https://proteobench.cubimed.rub.de/datasets/ As this is currently not supported (most saved submissions are wrong; or incomplete). Only in the latest version everything is correct. |
@@ -205,7 +205,7 @@ def plot_metric( | |||
width=700, | |||
height=700, | |||
xaxis=dict( | |||
title="Mean absolute difference between measured and expected log2-transformed fold change", | |||
title="Median absolute difference between measured and expected log2-transformed fold change", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Shouldnt the title be changed based on what is plotted?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, and that's what's currently implemented on https://github.com/Proteobench/ProteoBench/tree/mean_median_plot
I'll make a new PR from that branch, then I think this one should be closed without merging
Yes, thanks. Manual re-upload with a Jupiter notebook is great. |
Fixes #436 according to @mlocardpaulet's suggestion
Important: the changes made here do mean that all the datapoints currently plotted are plotted incorrectly, because metrics for these datapoints cannot be retroactively recalculated (I think)