How to load a .pt pretained model using scipy.misc

hi there,
I am trying to load a pretrained VGG16 model through the code below:

import os

import torch

import misc
import numpy as np
import pdb
from config import EasyDict
import tfutil
import argparse
import csv
import tensorflow as tf
import tensorflow_hub as hub
import PIL
from PIL import Image
import matplotlib.pyplot as plt

# initialize parser arguments
parser = argparse.ArgumentParser()
parser.add_argument('--results_dir', '-results_dir', help='name of training experiment folder', default='C:/Users/kaisios/Desktop/AI-plastic-sergery-work-space/pretrained_model/', type=str)
parser.add_argument('--labels_size', '-labels_size', help='size of labels vector', default=60, type=int)
parser.add_argument('--iters', '-iters', help='learning rate of algorithm', default=100000, type=int)
parser.add_argument('--lr', '-lr', help='learning rate of algorithm', default=0.1, type=float)
parser.add_argument('--alpha', '-alpha', help='weight of normal loss in relation to vgg loss', default=0.7, type=float)
parser.add_argument('--gpu', '-gpu', help='gpu index for the algorithm to run on', default='0', type=str)
parser.add_argument('--image_path', '-image_path', help='full path to image', default='../datasets/CelebA-HQ/img/03134.png', type=str)
parser.add_argument('--resolution', '-resolution', help='resolution of the generated image', default=256, type=int)

args = parser.parse_args()

# manual parameters
result_subdir = misc.create_result_subdir('results', 'inference_test')

# initialize TensorFlow
print('Initializing TensorFlow...')
env = EasyDict()  # Environment variables, set by the main program in
env.TF_CPP_MIN_LOG_LEVEL = '1'  # Print warnings and errors, but disable debug info.
env.CUDA_VISIBLE_DEVICES = args.gpu  # Unspecified (default) = Use all available GPUs. List of ints = CUDA device numbers to use. change to '0' if first GPU is better
tf_config = EasyDict()  # TensorFlow session config, set by tfutil.init_tf().
tf_config['graph_options.place_pruned_graph'] = True  # False (default) = Check that all ops are available on the designated device.
tf_config['gpu_options.allow_growth'] = True

# load network
network_pkl = misc.locate_network_pkl(args.results_dir)
print('Loading network from "%s"...' % network_pkl)
G, D, Gs = misc.load_network_pkl(args.results_dir, None)
#Gs = torch.load(network_pkl)
# initiate random input
latents = misc.random_latents(1, Gs, random_state=np.random.RandomState(800))
labels = np.random.rand(1, args.labels_size)

# upload image and convert to input tensor
img =
img = img.resize((args.resolution,args.resolution), Image.ANTIALIAS)'/')[-1]) # save image for debug purposes
img = np.asarray(img)
img = img.transpose(2, 0, 1)
img = np.expand_dims(img, axis=0)
img = (img / 127.5) - 1.0 # normalization

# execute algorithm
history = Gs.reverse_gan_for_etalons(latents, labels, img, results_dir=args.results_dir, dest_dir=result_subdir, iters=args.iters,, alpha=args.alpha)

# save history of latents
with open(result_subdir+'/history_of_latents.txt', 'w') as f:
    for item in history:

but when it gets to this line “G, D, Gs = misc.load_network_pkl(args.results_dir, None)”
it return an error " TypeError: ‘int’ object is not iterable"
it seems like it load the model as an integer
any help pleas ??


I’m not familiar with a misc package? Is that a custom thing? I can’t find any doc online for misc.load_network_pkl do you have a reference?

to be honest I am not pretty sure if it something stander I’ve downloaded the app from gethub and i want to run it I am gonna refer to it

I 'm trying to run