concat 3 columns of dataframe to one column

[flash77]

Hi,
I've got a dataframe with 100 rows and 3 columns (please see attachment).

How can I concate the 3 columns to one column?

Per example the 3 elements of the first row (255 255 255) should be 1 element (255255255) and so on.

1 2 3 4 5 6 7 8 9 10 11 12

import numpy as np import pandas as pd from PIL import Image   colorImage = Image.open("p2.bmp") colorPixels = colorImage.convert('RGB') # Make into array colorArray = np.array(colorPixels.getdata()) df = pd.DataFrame(colorArray) print(df) df2 = pd.concat(df["0"], df["1"], df["2"]) print(df2)

Thanks a lot for your help!!

[deanhystad]

You cannot put a list, array or tuple in a DataFrame column. Do you want to convert to a string?

1 2 3 4 5 6 7 8

import numpy as np import pandas as pd from PIL import Image   colorImage = Image.open("test.png").convert("RGB") colorArray = np.array([f"({r:03}{g:03}{b:03})" for r, g, b in colorImage.getdata()]) df = pd.DataFrame(colorArray) print(df)

Output:
               0
0    (034177076)
1    (237028036)
2    (237028036)
. . .

I don't see any value in that, but maybe there is for you.

Are you trying to get a 24 bit color number, or color string?

1 2 3 4 5 6 7 8 9 10 11 12

import numpy as np import pandas as pd from PIL import Image   colorImage = Image.open("test.png").convert("RGB") df = pd.DataFrame(np.array(colorImage.getdata()), columns=["R", "G", "B"]) df["24bit"] = df["R"] * 65536 + df["G"] * 256 + df["B"] df["Red"] = df["R"].apply(hex) df["Green"] = df["G"].apply(hex) df["Blue"] = df["B"].apply(hex) df["RGB"] = df["24bit"].apply(hex) print(df)

Output:
       R    G   B     24bit   Red Green  Blue       RGB
0     34  177  76   2273612  0x22  0xb1  0x4c  0x22b14c
1    237   28  36  15539236  0xed  0x1c  0x24  0xed1c24
2    237   28  36  15539236  0xed  0x1c  0x24  0xed1c24

If you're only using pandas to try to make a 24bit pixel you can do that in numpy.

1 2 3 4 5 6

import numpy as np from PIL import Image   rgb = np.array(Image.open("test.png").convert("RGB").getdata()) b24 = np.apply_along_axis(lambda p: p[0] * 65536 + p[1] * 256 + p[2], 1, rgb) print([hex(pixel) for pixel in b24])

Output:
['0x22b14c', '0xed1c24', '0xed1c24', '0xed1c24', '0xed1c24', '0xed1c24', '0xed1c24', '0xed1c24',...

Or if you just one those goofy strings from the first example, you don't need pandas or numpy.

1 2 3 4 5 6 7

from PIL import Image   b24 = [     f"({r:03}{g:03}{b:03})"     for r, g, b in Image.open("test.png").convert("RGB").getdata() ] print(b24)

Output:
['(034177076)', '(237028036)', '(237028036)', '(237028036)', '(237028036)', '(237028036)',...

[flash77]

Hi,

Thanks a lot for your answer!

I will describe in detail what I want to achieve tomorrow.

Greetings...