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

stylo.rolling_delta() analysis issue - unable to run on more than 12 documents and unable to customize colors/shapes with gui=True #54

Open
corvusMidnight opened this issue Mar 1, 2023 · 3 comments

Comments

@corvusMidnight
Copy link

When using the stylo() package in R, we've encountered issues with the stylo.rolling_delta() analysis function. Specifically, it is not possible to run this analysis on more than 12 documents: when the attribute gui=True is used, the user interface does not allow the user to customize the colors or shapes used in the output graph beyond those that are provided in the interface to fit more 12 documents.

We have attempted to find a way to customize the colors and shapes used in the graph, even when not using the GUI, but have not been successful.

These issues are hindering the usability of the package, and we would appreciate any guidance or assistance in resolving them.

Steps to reproduce:

Install the stylo() package in R
Use stylo.rolling_delta() analysis function on more than 12 documents
Try to customize colors or shapes in output graph when gui=True or not using GUI

@perechen
Copy link

perechen commented Mar 2, 2023

I'm not aware of any limitation, but (if I remember correctly) you had a setup where each document formed its own class. Having more than 12 classes, indeed, might set things off, since it is rarely practical to do a sequential classification on that many. Can you maybe provide the console output when you run >12 docs? Also, a screenshot of the training set corpus would be helpful.

@corvusMidnight
Copy link
Author

This is what my folder looks like:

issue

This is the output graph:

Good_Omens_NOVEL001

The console output does not present any irregularities.

@perechen
Copy link

perechen commented Mar 8, 2023

Thanks! I see the legend has all documents; the problem is that the lines on plot correspond only to 12? It might be some color assignment weirdness (and data is actually there). Btw, rolling_classify() should treat documents by the default stylo procedure, which is treating the string before the first underscore as a class.

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

No branches or pull requests

2 participants