CODEWITHSUNDEEP
pythonpandaslistdataframecomputer-vision
Engr Ali
Engr Ali

Reputation: 369

how to split multiple lists in a column and make new dataframe and map for specific value in python?

I am working on a data frame containing multiple columns. One of the column having multiple lists like: [[[0.0, 0.0, 1024.0, 1024.0], [0.0, 0.0, 1024.0, 1024.0], [0.0, 0.0, 1024.0, 1024.0], [0.0, 0.0, 1024.0, 1024.0]], ['Effusion', 'Emphysema', 'Infiltration', 'Pneumothorax']]. If you don't understand the data shape then please find the file shared in drive link

Each row having a single or multiple lists of lists and the first list pointing to the first disease which is mentioned in a string format in the second list and so on.

What is am Looking for:

I want to split this column to make new columns for each disease to make a data frame like this:

['Effusion'][[[0.0, 0.0, 1024.0, 1024.0] for each disease mentioned. OR want like this from the uploaded data data frame like: img_name, class, xmin = 0.0 , ymin= 0.0, xmax = 1024, ymax = 1024

If anyone is possible then iterates over all the rows and for all images and print, all Bounding Box coordinates on the image like below. OR

I want to make a data frame like this from the uploaded file link as you can see at the end:

img_name  label           xmin        ymin       xmax      ymax
001       Effusion        0.0         0.0        1024      1024
001       Emphysema       0.0         0.0        1024      1024
001       Infiltration    0.0         0.0        1024      1024
001       Effusion        595.97      601.39     776.93    817.03
002       Mass            0.0         0.0        1024      1024
002       Cardiomagly     0.0         0.0        1024      1024
002       any disease     395.97      351.39     866.93    917.03
002       any disease     0.0         0.0        1024      1024

Here the above data frame style is change as I want to use explode function of pandas data.

you can also use your expertise to just get those coordinated with the image name mentioned and print on the image. I am uploading a sample data frame here in this link: Dataframe

Thanks

Upvotes: 0

Views: 279

Answers (1)

mozway
mozway

Reputation: 261860

It is difficult to give you a precise code to run without a minimal reproducible example, but assuming this input:

list_of_list = [[[0.0, 0.0, 1024.0, 1024.0],
                 [0.0, 0.0, 1024.0, 1024.0],
                 [0.0, 0.0, 1024.0, 1024.0],
                 [0.0, 0.0, 1024.0, 1024.0]],
                ['Effusion', 'Emphysema', 'Infiltration', 'Pneumothorax']
               ]

You can use zip to reshape your list:

list(zip(list_of_list[1], list_of_list[0]))

or

list(zip(*list_of_list[::-1]))

output:

[('Effusion', [0.0, 0.0, 1024.0, 1024.0]),
 ('Emphysema', [0.0, 0.0, 1024.0, 1024.0]),
 ('Infiltration', [0.0, 0.0, 1024.0, 1024.0]),
 ('Pneumothorax', [0.0, 0.0, 1024.0, 1024.0])]

Upvotes: 1

Related Questions