Importing other Python scripts as Modules

Question

I am currently trying to program an application that splits a video into its individual frames and then finds the faces on the video and extracts them as .jpg. I split the Project into multiple files, app.py which is responsible for GUI and so on and extractor.py which does the work.

I thought you would be able to import a file with:

import extractor

And then run it like this:

extractor()

Apparently, this doesnt seem to work. I also tried making the entire extractor script a function and then calling the function, but this also doesnt work.

app.py:

import extractor
extractor()

extractor.py:

import cv2
import os
import face_recognition
from PIL import Image
import multiprocessing

try:
    if not os.path.exists('frames'):
        os.makedirs('frames')
except OSError:
    print('Error: Creating directory of frames')

try:
    if not os.path.exists('faces'):
        os.makedirs('faces')
except OSError:
    print('Error: Creating directory of faces')

def extract_frames(video_file_path):
    currentFrame_extract = 1
    video_capture = cv2.VideoCapture(video_file_path)

    while(True):
        ret, frame = video_capture.read()
        if ret == False:
            break
        name = 'frames/frame_' + str(currentFrame_extract) + '.jpg'
        print(f"Extracting Frame {currentFrame_extract}, saving it as Frame_{currentFrame_extract}.jpg")
        cv2.imwrite(name, frame)
        currentFrame_extract += 1

    video_capture.release()
    cv2.destroyAllWindows()
    return currentFrame_extract

def find_faces_a(a):
    i = 0
    currentFrame = 1

    while (True):

        if a > currentFrame:

            image = face_recognition.load_image_file(f"data/frame_{currentFrame}.jpg")
            face_locations = face_recognition.face_locations(image)

            if len(face_locations) >= 1:
                top, right, bottom, left = face_locations[0]

                face_image = image[top:bottom, left:right]
                pil_image = Image.fromarray(face_image)
                pil_image.save(f"faces/face_{currentFrame}.jpg".format(i))
                print(f"Found a face at Frame_{currentFrame}, exporting it as Face_{currentFrame}.jpg")

            currentFrame += 4
        else:
            break

def find_faces_b(a):
    i = 0
    currentFrame = 2

    while (True):

        if a > currentFrame:

            image = face_recognition.load_image_file(f"data/frame_{currentFrame}.jpg")
            face_locations = face_recognition.face_locations(image)

            if len(face_locations) >= 1:
                top, right, bottom, left = face_locations[0]

                face_image = image[top:bottom, left:right]
                pil_image = Image.fromarray(face_image)
                pil_image.save(f"faces/face_{currentFrame}.jpg".format(i))
                print(f"Found a face at Frame_{currentFrame}, exporting it as Face_{currentFrame}.jpg")

            currentFrame += 4
        else:
            break

def find_faces_c(a):
    i = 0
    currentFrame = 3

    while (True):

        if a > currentFrame:

            image = face_recognition.load_image_file(f"data/frame_{currentFrame}.jpg")
            face_locations = face_recognition.face_locations(image)

            if len(face_locations) >= 1:
                top, right, bottom, left = face_locations[0]

                face_image = image[top:bottom, left:right]
                pil_image = Image.fromarray(face_image)
                pil_image.save(f"faces/face_{currentFrame}.jpg".format(i))
                print(f"Found a face at Frame_{currentFrame}, exporting it as Face_{currentFrame}.jpg")

            currentFrame += 4
        else:
            break

def find_faces_d(a):
    i = 0
    currentFrame = 4

    while (True):

        if a > currentFrame:

            image = face_recognition.load_image_file(f"data/frame_{currentFrame}.jpg")
            face_locations = face_recognition.face_locations(image)

            if len(face_locations) >= 1:
                top, right, bottom, left = face_locations[0]

                face_image = image[top:bottom, left:right]
                pil_image = Image.fromarray(face_image)
                pil_image.save(f"faces/face_{currentFrame}.jpg".format(i))
                print(f"Found a face at Frame_{currentFrame}, exporting it as Face_{currentFrame}.jpg")

            currentFrame += 4
        else:
            break

if __name__ == "__main__":

    video_file_path = "Video_3.mp4"
    currentFrame_extract = extract_frames(video_file_path)

    currentFrame_extract = [currentFrame_extract]
    p1 = multiprocessing.Process(target=find_faces_a, args=(currentFrame_extract))
    p2 = multiprocessing.Process(target=find_faces_b, args=(currentFrame_extract))
    p3 = multiprocessing.Process(target=find_faces_c, args=(currentFrame_extract))
    p4 = multiprocessing.Process(target=find_faces_d, args=(currentFrame_extract))

    p1.start()
    p2.start()
    p3.start()
    p4.start()

    p1.join()
    p2.join()
    p3.join()
    p4.join()

    print("Frame extraction and alignment finished successfully.")

I receive the error: TypeError: 'module' object is not callable. If I do it like some of you have suggested or as in the question marked as "similar", the script will start but it still wont work and only create the folders.

When you import the module, everything before the `if __name__ == '__main__'` block will be evaluated. So all those functions will be defined and added to the `extractor` namespace. Try `print(dir(namespace))` to see what I mean. You could call `extractor.find_faces_a(some_object_here)` after you import the module... — duhaime, Jul 09 '19 at 14:50
Hi duhaime, thank you for your answer. I'm aware that I have all these individual functions, but how can I combine them into one, so that when I call extractor() it justs executes the entire script? The problem is that the "if __name__ == "__main__" is essential for the script and it should all run just as if I would run the file manually. — marcr, Jul 09 '19 at 14:51
Possible duplicate of [In Python, can I call the main() of an imported module?](https://stackoverflow.com/questions/14500183/in-python-can-i-call-the-main-of-an-imported-module) — Corentin Limier, Jul 09 '19 at 14:58

score 0 · Answer 1 · answered Jul 09 '19 at 15:05

0

You can run extractor.py by converting if __name__ == "__main__": to a new function def extractor() and importing the module:

import extractor;
extractor.extractor();

You can also only import a specific name (in our case extractor() function) by using the following variant of import:

from extractor import extractor;
extractor();

Check out this link (https://repl.it/repls/MeaslyMerrySymbol) where I have done a sample import similar to your files.

answered Jul 09 '19 at 15:05

Mayur Dhurpate

1,112
4
16
34

Hi, I tried doing that. It does create the folders and execute the start of the extractor script, however it doesnt extract the images neither search for faces. – marcr Jul 09 '19 at 15:17
So it runs without throwing any error per se? If it throws an error, pasting the full error message (5-10 red lines) (also called traceback) would give us more insights on what might be happening. – Mayur Dhurpate Jul 09 '19 at 15:25
Yep, just runs through and say "Process finished with exit code 0". However it doesnt really work and I think the all the functions dont get executed. It gets as far as creating the folders, nothing more happens. – marcr Jul 09 '19 at 15:28
Any Idea? I think the problem is that the __name__ == __main__ doesnt start and with that the functions on top dont get started either. I dont really understand what __name__ == __main__ even does, but when I leave it away the script doesnt work. – marcr Jul 09 '19 at 15:43

score 0 · Answer 2 · answered Jul 09 '19 at 15:06

Encapsulate your extractor functionality in another file, like extractor_impl. Then put everything in a function in this file:

def extract(video_file_path)
    currentFrame_extract = extract_frames(video_file_path)

    currentFrame_extract = [currentFrame_extract]
    p1 = multiprocessing.Process(target=find_faces_a, args=(currentFrame_extract))
    p2 = multiprocessing.Process(target=find_faces_b, args=(currentFrame_extract))
    p3 = multiprocessing.Process(target=find_faces_c, args=(currentFrame_extract))
    p4 = multiprocessing.Process(target=find_faces_d, args=(currentFrame_extract))

    p1.start()
    p2.start()
    p3.start()
    p4.start()

    p1.join()
    p2.join()
    p3.join()
    p4.join()

    print("Frame extraction and alignment finished successfully.")

You can then import the extractor_impl file from your extractor file and just call this function, but you can import and call it from other files as well.

I tried that, but unfortunatly some of the parts of the script are just left out when I do it this way. It creates the folders, but doesnt extract any images. — marcr, Jul 09 '19 at 15:17

Importing other Python scripts as Modules

2 Answers2