What is this?

This document shows you how to install Colorful Image Colorization on Windows 10 and Python 3.x (3.8.6).

Steps

Step 1. Install Python on Windows

Version 3.8 of Windows x86-64 executable installer seems better:
https://www.python.org/downloads/windows/

Step 2. Install richzhang/colorization

Clone the repository

git clone https://github.com/richzhang/colorization.git

CUDA

Download and install CUDA 10.2:
https://developer.nvidia.com/cuda-downloads

PyTorch (torch)

pip install torch===1.6.0 torchvision===0.7.0 -f https://download.pytorch.org/whl/torch_stable.html

# For more details see below
# https://stackoverflow.com/questions/56859803/modulenotfounderror-no-module-named-tools-nnwrap

Others

Please note that I chose scikit-image instead of skimage.

pip install wheel scikit-image matplotlib argparse
pip install ipython

Step 3. Run

cd colorization
python demo_release.py -i imgs/ansel_adams3.jpg

Congratulations!

Logs

PyTorch (torch)

Click here to see command-line logs

C:\Users\AAA>pip install torch===1.6.0 torchvision===0.7.0 -f https://download.pytorch.org/whl/torch_stable.html
Looking in links: https://download.pytorch.org/whl/torch_stable.html
Collecting torch===1.6.0
  Downloading https://download.pytorch.org/whl/cu102/torch-1.6.0-cp38-cp38-win_amd64.whl (1077.4 MB)
     |████████████████████████████████| 1077.4 MB 833 bytes/s
Collecting torchvision===0.7.0
  Downloading https://download.pytorch.org/whl/cu102/torchvision-0.7.0-cp38-cp38-win_amd64.whl (1.1 MB)
     |████████████████████████████████| 1.1 MB 6.8 MB/s
Collecting future
  Downloading future-0.18.2.tar.gz (829 kB)
     |████████████████████████████████| 829 kB 3.2 MB/s
Collecting numpy
  Downloading numpy-1.19.2-cp38-cp38-win_amd64.whl (13.0 MB)
     |████████████████████████████████| 13.0 MB 6.8 MB/s
Collecting pillow>=4.1.1
  Downloading Pillow-8.0.0-cp38-cp38-win_amd64.whl (2.1 MB)
     |████████████████████████████████| 2.1 MB 6.4 MB/s
Using legacy 'setup.py install' for future, since package 'wheel' is not installed.
Installing collected packages: future, numpy, torch, pillow, torchvision
    Running setup.py install for future ... done
Successfully installed future-0.18.2 numpy-1.19.2 pillow-8.0.0 torch-1.6.0 torchvision-0.7.0

PIL seems to be the name used in Python 2.x, and pillow seems to be the name used in Python 3.x.

Others

Click here to see command-line logs

C:\Users\AAA>pip install wheel scikit-image matplotlib argparse
Collecting wheel
  Using cached wheel-0.35.1-py2.py3-none-any.whl (33 kB)
Collecting scikit-image
  Downloading scikit_image-0.17.2-cp38-cp38-win_amd64.whl (11.7 MB)
     |████████████████████████████████| 11.7 MB 3.3 MB/s
Collecting matplotlib
  Downloading matplotlib-3.3.2-cp38-cp38-win_amd64.whl (8.5 MB)
     |████████████████████████████████| 8.5 MB 6.4 MB/s
Collecting argparse
  Using cached argparse-1.4.0-py2.py3-none-any.whl (23 kB)
Collecting scipy>=1.0.1
  Downloading scipy-1.5.3-cp38-cp38-win_amd64.whl (31.4 MB)
     |████████████████████████████████| 31.4 MB 6.4 MB/s
Collecting imageio>=2.3.0
  Downloading imageio-2.9.0-py3-none-any.whl (3.3 MB)
     |████████████████████████████████| 3.3 MB ...
Requirement already satisfied: pillow!=7.1.0,!=7.1.1,>=4.3.0 in c:\home\sdk\python38\lib\site-packages (from scikit-image) (8.0.0)
Collecting tifffile>=2019.7.26
  Downloading tifffile-2020.10.1-py3-none-any.whl (152 kB)
     |████████████████████████████████| 152 kB 6.8 MB/s
Collecting PyWavelets>=1.1.1
  Downloading PyWavelets-1.1.1-cp38-cp38-win_amd64.whl (4.3 MB)
     |████████████████████████████████| 4.3 MB 6.8 MB/s
Requirement already satisfied: numpy>=1.15.1 in c:\home\sdk\python38\lib\site-packages (from scikit-image) (1.19.2)
Collecting networkx>=2.0
  Downloading networkx-2.5-py3-none-any.whl (1.6 MB)
     |████████████████████████████████| 1.6 MB ...
Collecting pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.3
  Using cached pyparsing-2.4.7-py2.py3-none-any.whl (67 kB)
Collecting certifi>=2020.06.20
  Using cached certifi-2020.6.20-py2.py3-none-any.whl (156 kB)
Collecting python-dateutil>=2.1
  Using cached python_dateutil-2.8.1-py2.py3-none-any.whl (227 kB)
Collecting cycler>=0.10
  Using cached cycler-0.10.0-py2.py3-none-any.whl (6.5 kB)
Collecting kiwisolver>=1.0.1
  Downloading kiwisolver-1.2.0-cp38-none-win_amd64.whl (58 kB)
     |████████████████████████████████| 58 kB ...
Collecting decorator>=4.3.0
  Downloading decorator-4.4.2-py2.py3-none-any.whl (9.2 kB)
Collecting six>=1.5
  Using cached six-1.15.0-py2.py3-none-any.whl (10 kB)
Installing collected packages: wheel, scipy, imageio, tifffile, PyWavelets, pyparsing, certifi, six, python-dateutil, cycler, kiwisolver, matplotlib, decorator, networkx, scikit-image, argparse
Successfully installed PyWavelets-1.1.1 argparse-1.4.0 certifi-2020.6.20 cycler-0.10.0 decorator-4.4.2 imageio-2.9.0 kiwisolver-1.2.0 matplotlib-3.3.2 networkx-2.5 pyparsing-2.4.7 python-dateutil-2.8.1 scikit-image-0.17.2 scipy-1.5.3 six-1.15.0 tifffile-2020.10.1 wheel-0.35.1

c:\home\sdk\python38 is a directory name that is unique to my environment.

First run

Click here to see command-line logs

C:\home\src\colorization>python demo_release.py -i imgs/ansel_adams3.jpg
Downloading: "https://colorizers.s3.us-east-2.amazonaws.com/colorization_release_v2-9b330a0b.pth" to C:\Users\AAA/.cache\torch\hub\checkpoints\colorization_release_v2-9b330a0b.pth
100.0%
Downloading: "https://colorizers.s3.us-east-2.amazonaws.com/siggraph17-df00044c.pth" to C:\Users\AAA/.cache\torch\hub\checkpoints\siggraph17-df00044c.pth
100.0%
C:\home\sdk\python38\lib\site-packages\torch\nn\functional.py:3118: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details.
  warnings.warn("Default upsampling behavior when mode={} is changed "
C:\home\sdk\python38\lib\site-packages\skimage\color\colorconv.py:1128: UserWarning: Color data out of range: Z < 0 in 367 pixels
  return xyz2rgb(lab2xyz(lab, illuminant, observer))
C:\home\sdk\python38\lib\site-packages\skimage\color\colorconv.py:1128: UserWarning: Color data out of range: Z < 0 in 33 pixels
  return xyz2rgb(lab2xyz(lab, illuminant, observer))![](https://storage.googleapis.com/zenn-user-upload/owfupao294w6w9z4j0wr2w813agg)

Note

This page is a clone of the Zenn article.