AttributeError: partially initialized module 'cv2' has no attribute 'VideoCapture'

[emanuelsc3s]

I have the error when running:

used Delphi 10.4.2 and Python 3.9.2 64bit

Traceback (most recent call last):
File "", line 51, in remover_frame_a_frame_e_fazer_predicoes_e_plotar
AttributeError: partially initialized module 'cv2' has no attribute 'VideoCapture' (most likely due to a circular import)

Help please...

def retirar_frame_a_frame_e_fazer_predicoes_e_plotar(path_video, y_true):

  violent_count = 0
  non_violent_count = 0
  contador = 0
  
  vetor_probs = []
  vetor_labels = []
  vetor_imagens = []
  
  cap = cv2.VideoCapture(path_video)

  while (cap.isOpened()):
    ret, frame = cap.read()

    if (ret == False):
      break
    
    img = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB) 

    img_pil = Image.fromarray(img)
    vetor_imagens.append(img_pil)

    img_tensor = transformer(img_pil)
    
    img_tensor.unsqueeze_(0)

    resultado = model.forward(img_tensor)

    probs = torch.exp(resultado)

    probs = probs.max(1)

    probabilidade = probs[0].detach().numpy()[0]
    vetor_probs.append(probabilidade)

    label_predita = probs[1].detach().numpy()[0]
    vetor_labels.append(label_predita)

    contador += 1

    if (label_predita == 1):
      violent_count += 1
    else:
      non_violent_count += 1
    
    taxa_frames_violentos = 100*violent_count/contador
    taxa_frames_nao_violentos = 100*non_violent_count/contador
  
  plt.figure(figsize=(15, 10))
  for k in range(9):
    plt.subplot(3, 3, k+1)
    n = random.randint(0, len(vetor_probs)-1)
    img = vetor_imagens[n]
    label = vetor_labels[n]
    plt.imshow(img)
    plt.xticks([])
    plt.yticks([])
    plt.title('Real: {0}\nPredito: {1}'.format(vetor_classes[y_true], vetor_classes[label]))
  
  plt.show()

  return np.array(vetor_probs), np.array(vetor_labels), taxa_frames_violentos, taxa_frames_nao_violentos

[emanuelsc3s]

I used

import cv2
import pathlib
import os
import numpy as np
import matplotlib.pyplot as plt
from PIL import Image, ImageDraw
import random
import shutil
import zipfile
from tqdm import tqdm

import torch
import torchvision
from torch.utils.data import DataLoader, Dataset
from torchvision.datasets import ImageFolder
from torchvision import transforms
import random
from time import time
from torch.utils.data import random_split

[pyscripter]

Why do you believe this is P4D issue?

[pyscripter]

@alperyazgan Please address your question to the support forum (with some more details).

[emanuelsc3s]

Por que você acredita que este é um problema de P4D?

because I ran the same code in Visual Studio Code and it worked

[emanuelsc3s]

Why do you believe this is P4D issue?

the error happens in the function call:

vetor_probs, vetor_labels, taxa_frames_violentos, taxa_frames_nao_violentos = retirar_frame_a_frame_e_fazer_predicoes_e_plotar(r'C:\cam2\11.mp4', 1)

[pyscripter]

because I ran the same code in Visual Studio Code and it worked

This does not show that it is P4D issue. VS Code may be running a different version of Python with a different sys.path and different versions of the libraries. You can print sys.version, sys.path sys.executable etc. and compare.
It is python that is running your code and not P4D. You need to figure out what goes wrong.

[SwiftExpat]

I solved a smiliar situation:

I used demo01 to call each of my imports individually to eventually trace it down to C++ runtime not being installed. For me the offending package was skimage. It is possible to see what is missing when you trace the dll load with sysinternals procmon.

I can best illustrate it as:
VS Code --> C++ [python.exe] --> loads python.dll --> imports skimage --> requires C++ runtime (provided by python)

  Delphi[Demo01.exe] --> loads python.dll --> imports skimage --> requires C++ runtime (provided by OS in path)

So when it runs in VS Code, the runtime requirement is fulfilled. (at least my best guess)

[emanuelsc3s]

Thank you very much for your attention ...
It worked with MaskFPUExceptions in link:

https://github.com/pyscripter/python4delphi/wiki/MaskFPUExceptions

@pyscripter very much