In wsgi_app error in app, loc doesnt work well with dropdownmenus multi=True


I have a dropdown menu with multi for a callback

but this gives errors

X= dff3.loc[:,[drop1]]

what could I use instead? (when I add more than 1 drop, the app stops …

Does your drop downs consist of values to search or column names?

Your loc syntax is incorrect also, looks like you are trying iloc.

Depending on your above answer, I would use isin.

Df.loc[df.isin(multi list)]

Hi, the first dropdown consist of column names.

1q: the multi list is a list of all the column names?

I’ve tried Df.loc[df.isin(multi list)] and the app breaks …

Can you share the error?

Can you also share what you are expecting in the output? Do you just want the specified columns of data?

line 1192, in _getitem_axis
    raise ValueError("Cannot index with multidimensional key")
ValueError: Cannot index with multidimensional key

Can you also share what you are expecting in the output? Do you just want the specified columns of data?

this is the case of study:

I have some data

Ihave to clean that data befero doing operations

to select the data, I have to choose the columns with the multi dropdown1 (and dont know how to do this)

after this I need to clean data with dropna

then comes all the operations (where the other dropdown comes to action)

and after this, it should give a table of results

that table of results its of width dropdown 2 and hight dropdown 1

and I should show the table in the dashboard

If all you want is the data and selected columns, then df[drop1] would do that. No need for loc.

If you still get the error, then it would appear you have a 2D dataframe.

I would need a sample of the data to help further.

drop1 is a multi … it can be 2d or 90d (here’s where I struggle)

Sharing your code would be helpful then.

cant do that … sorry … I’ve tried to be as explicit as I could …

my biggest problem is with the multi dropdown selection

dont know how to loc it?

Hi, I’ve managed to print a the data in the dashboard … its not in a table … but at least its the correct data … do you know how I could do a table for this (the data is show very ugly)

the table must be arranged by the drop2 == columns and drop1 == rows
(the solution was to not use the multi= true in the dropbox … a cheap solution but works)