Closed creuzige closed 2 years ago
One option that I sometimes use is having the full DataFrame available for computations, but then round whichever columns as you are plotting it..
Open to that, but it seemed like what gets returned to be plotted isn't the dataframe anymore, it's a dictionary. Maybe there are other ways to return values/tables we should look into? I felt that a bit when trying to add additional terms to plot...
Yeah, it's kind of annoying that dash data_table can't just take in a pandas DataFrame.. perhaps it'd be better to keep the table as a dataframe and to write a short function that converts the DF to a dictionary only when needed for plotting (and we can do the same for rounding)
I like that plan. Then we probably could just return dataframes from the compute* modules. And maybe then just output the dataframes to a file...
To your last sentence, do you think we should have a 'download as .csv' button for the user to download the tables as a csv?
I think Max started on something like this, i.e. being able to download a 'report'. But the content and format hasn't gotten ironed out yet.
I've tried a few times to truncate the precision of the items in the normalized intensity table with no success. I added and commented out what I've tried. Most of the suggestions from Dash suggest modifying the DataFrame or data directly, which I'd prefer not to do.