Alejandro de la Vega

I think the issue is that `descriptions` should not be a list, but an object. I think the assumption here is that across all the files the columns should be...

Relevant line: https://github.com/neuroscout/neuroscout/blob/ed79e9cf4b1ee1320a2d43c72e95f3fd3619c9b7/neuroscout/tasks/upload.py#L71 Might be a good idea to validate inputs

Idea: Instead of replicating git push, simply add a script that can be run via ssh which pull the `latest` image and does down/up

Awesome, thanks for looking into it. Seems like the simplest starting point is to make a script to cross reference the subject event file and generate new event files with...

- Let's strip punctuation for now to be consistent. But in the future we can think of a way to ingest punctuation / sentence structure. We need to think about...

I think the breaks looks short enough that collapsing would be fair. We could also set a threshold of maybe 1s. If there is a longer break, we can split...

About the breaks between words, we could also model it both ways, although I just don't think it'll make too much of a difference if the breaks are so short....

Scale is buggy if you switch Transformations

I think is related to #854

I also keep getting a "Ooops something went wrong" error message. Confirm this is still an issue