# loop over the input image paths
for (i, imagePath) in enumerate(imagePaths):

    print ("[INFO] predicting on image {} of {}".format(i+1, len(imagePaths)))

    # Create the filename to store the predictions in and then open it in write mode
    filename = (imagePath.split(os.path.sep)[-1]).split('.')[0]
    output_file = os.path.sep.join([args["output"], '{}.txt'.format(filename)])
    file = open(output_file, 'w')

    #load the input image (BGR), clone it, and preprocess it
    image = read_image_bgr(imagePath)
    image = preprocess_image(image)
    (image, scale) = resize_image(image)
    image = np.expand_dims(image, axis=0)

    # detect objects in the input image and correct for the image scale
    (boxes, scores, labels) = model.predict_on_batch(image)
    boxes /= scale