Python/Keras: -1073740791 (0xC0000409) Error when calling model.predict()

I am working on the Computer Vision Deep Learning Program and I have encountered what looks to me a very unspecific error message that does not give any hints whatsoever.

Process finished with exit code -1073740791 (0xC0000409)

In search of the solution I found out that the code works perfectly fine with the CPU, although it takes much longer to process.
When debugging, I found out that the error is thrown at the return of the model.predict(). Searching the web I couldn't find a relevant problem and the solution that would work.

Code used:

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout
from tensorflow.keras.preprocessing import image
from tensorflow.keras.applications.vgg16 import VGG16
from sklearn.metrics import accuracy_score
import numpy as np
import pandas as pd
from tqdm import tqdm
import cv2
import math
import os
from glob import glob
from scipy import stats as s


def dl_training():
    base_model = VGG16(weights='imagenet', include_top=False)

    # defining the model architecture
    model = Sequential()
    model.add(Dense(1024, activation='relu', input_shape=(25088,)))
    model.add(Dropout(0.5))
    model.add(Dense(512, activation='relu'))
    model.add(Dropout(0.5))
    model.add(Dense(256, activation='relu'))
    model.add(Dropout(0.5))
    model.add(Dense(128, activation='relu'))
    model.add(Dropout(0.5))
    model.add(Dense(64, activation='relu'))
    model.add(Dropout(0.5))
    model.add(Dense(32, activation='relu'))
    model.add(Dropout(0.5))
    model.add(Dense(2, activation='softmax'))

    # loading the trained weights
    model.load_weights("weights_test.hdf5")

    # compiling the model
    model.compile(loss='categorical_crossentropy', optimizer='Adam', metrics=['accuracy'])

    # getting the test list
    f = open("Anomaly_Test_2class.txt", "r")
    temp = f.read()
    videos = temp.split('\n')
    del temp

    # creating the dataframe
    test = pd.DataFrame()
    test['video_name'] = videos
    test = test[:-1]
    test_videos = test['video_name']
    test.head()

    # creating the tags
    train = pd.read_csv('train_new_2class.csv')
    y = pd.get_dummies(train['class'])

    predict = []
    actual = []

    # for loop to extract frames from each test video
    for i in tqdm(range(test_videos.shape[0])):
        count = 0
        videoFile = test_videos[i]
        cap = cv2.VideoCapture(
            'Videos/' + videoFile)  # capturing the video from the given path
        # print('Videos/' + videoFile)
        frameRate = cap.get(5)  # frame rate

        # removing all other files from the temp folder
        files = glob('temp/*')
        for f in files:
            os.remove(f)
        while cap.isOpened():
            frameId = cap.get(1)  # current frame number
            ret, frame = cap.read()
            if not ret:
                break
            if frameId % math.floor(frameRate) == 0:
                # storing the frames of this particular video in temp folder
                filename = 'temp/' + "_frame%d.jpg" % count
                count += 1
                cv2.imwrite(filename, frame)
        cap.release()

        # reading all the frames from temp folder
        images = glob("temp/*.jpg")

        prediction_images = []
        for j in range(len(images)):
            img = image.load_img(images[j], target_size=(224, 224, 3))
            img = image.img_to_array(img)
            img = img / 255
            prediction_images.append(img)

        # converting all the frames for a test video into numpy array
        prediction_images = np.array(prediction_images)

        # extracting features using pre-trained model
        prediction_images = base_model.predict(prediction_images)

        # converting features in one dimensional array
        prediction_images = prediction_images.reshape(prediction_images.shape[0], 7 * 7 * 512)

        # predicting tags for each array
        prediction = model.predict_classes(prediction_images)

        # appending the mode of predictions in predict list to assign the tag to the video
        predict.append(y.columns.values[s.mode(prediction)[0][0]])
        # print("\n" + "Prediction: " + y.columns.values[s.mode(prediction)[0][0]])

        # appending the actual tag of the video
        if videoFile.split('/')[0].split('_')[0] == "Testing":
            actual.append("Normal_Videos_event")
            # print("Actual Class: Normal_Videos_event")
        else:
            actual.append(videoFile.split('/')[0])
            # print("Actual Class: " + videoFile.split('/')[0])

    # checking the accuracy of the predicted tags
    accuracy = accuracy_score(predict, actual) * 100
    print("Accuracy: %f" % accuracy)

This error also occurs when executing any other code that has a predict function in it. I should clarify that the GPU is connected and when testing using tf.test.is_built_with_cuda() and tf.test.is_gpu_available() is positive.

Also, one more thing worth mentioning is that I cannot fetch the documentation of the predict function.

Is this intended?

Studying this code from: https://www.analyticsvidhya.com/blog/2019/09/step-by-step-deep-learning-tutorial-video-classification-python/

Using Python 3.7, PyCharm, tensorflow-gpu 2.1.0, keras 2.3.1, numpy 1.20.1, scikit-learn 0.24.1, pandas 1.2.2, tqdm 4.56.2, scipy 1.4.1, opencv 4.5.1.48, pip 21.0.1, setuptools 53.0.0, six 1.15.0