Updated 29/Nov/2021 by Yoshihisa Nitta 
Analysis of Cycle Generative Adversarial Network for VidTIMIT dataset with Tensorflow 2 on Google Colab (WGAN-GP)¶
Assuming that you have already executed CycleGAN_VidTIMIT_Train.ipynb, analyzethe Model.
VidTIMIT データセットに対する Cycle Generative Adversarial Network をGoogle Colab 上の Tensorflow 2 で解析する¶
既に CycleGAN_VidTIMIT_Train.ipynb を実行していることを前提とし, 学習済みモデルを解析する。
In [ ]:
save_path = '/content/drive/MyDrive/ColabRun/CycleGAN_VidTIMIT01'
VERBOSE = False
In [ ]:
#! pip install tensorflow==2.7.0
In [ ]:
! pip install tensorflow_addons
In [ ]:
%tensorflow_version 2.x
import tensorflow as tf
print(tf.__version__)
In [ ]:
import numpy as np
np.random.seed(2022)
In [ ]:
if VERBOSE:
! nvidia-smi
! cat /proc/cpuinfo
! cat /etc/issue
! free -h
In [ ]:
from google.colab import drive
drive.mount('/content/drive')
In [ ]:
if VERBOSE:
! ls /content/drive
Download source file from Google Drive or nw.tsuda.ac.jp¶
Basically, gdown from Google Drive.
Only if you cannot download from Google Drive, wget from nw.tsuda.ac.jp.
Google Drive または nw.tsuda.ac.jp からファイルをダウンロードする¶
基本的に、Google Drive から gdown してください。
Google Drive からダウンロードができない場合にのみ、nw.tsuda.ac.jp から wget してください。
In [ ]:
# Download source file
nw_path = './nw'
! rm -rf {nw_path}
! mkdir -p {nw_path}
if True: # from Google Drive
url_model = 'https://drive.google.com/uc?id=1aNvpPDNeDWYQFu_PA1kOtFlzcO5seHky'
! (cd {nw_path}; gdown {url_model})
else: # from nw.tsuda.ac.jp
URL_NW = 'https://nw.tsuda.ac.jp/lec/GoogleColab/pub'
url_model = f'{URL_NW}/models/CycleGAN.py'
! wget -nd {url_model} -P {nw_path}
In [ ]:
if VERBOSE:
! cat {nw_path}/CycleGAN.py
Preparing VidTIMIT dataset¶
Official WWW of VidTIMIT dataset: http://conradsanderson.id.au/vidtimit/
zip files of 2 persons of VidTIMIT dataset:
https://zenodo.org/record/158963/files/fadg0.zip
https://zenodo.org/record/158963/files/faks0.zip
zip files mirrored on my Google Drive:
https://drive.google.com/uc?id=
VidTIMIT データセットを用意する¶
VidTIMIT データセットの公式ページ: http://conradsanderson.id.au/vidtimit/
VidTIMIT の2名の顔写真の zip ファイル:
https://zenodo.org/record/158963/files/fadg0.zip
https://zenodo.org/record/158963/files/faks0.zip
自分の Google Drive 上にミラーした顔写真:
https://drive.google.com/uc?id=
https://drive.google.com/uc?id=
In [ ]:
# Download zip files
VidTIMIT_site = 'https://zenodo.org/record/158963/files/'
VidTIMIT_fnames = [ 'fadg0', 'faks0']
Mirrored_files = [
'https://drive.google.com/uc?id=1_Fv4p9MDNphMZMnLpEvtCtnwXgN8N5Cj',
'https://drive.google.com/uc?id=1Y8j7ThPVqB0gbx4hb9aMEp9Ptr9wFuoz'
]
data_dir = './datasets'
! rm -rf $data_dir
! mkdir -p $data_dir
for i, fname in enumerate(VidTIMIT_fnames):
fzip = fname + '.zip'
if False:
url = VidTIMIT_site + fzip
!wget {url}
else:
url = Mirrored_files[i]
!gdown {url}
!unzip -q {fzip} -d {data_dir}
In [ ]:
IMAGE_SIZE = 128
In [ ]:
import os
import glob
imgA_paths = glob.glob(os.path.join(data_dir, VidTIMIT_fnames[0], 'video/*/[0-9]*'))
imgB_paths = glob.glob(os.path.join(data_dir, VidTIMIT_fnames[1], 'video/*/[0-9]*'))
In [ ]:
import numpy as np
validation_split = 0.05
nA, nB = len(imgA_paths), len(imgB_paths)
splitA = int(nA * (1 - validation_split))
splitB = int(nB * (1 - validation_split))
np.random.shuffle(imgA_paths)
np.random.shuffle(imgB_paths)
train_imgA_paths = imgA_paths[:splitA]
test_imgA_paths = imgA_paths[splitA:]
train_imgB_paths = imgB_paths[:splitB]
test_imgB_paths = imgB_paths[splitB:]
In [ ]:
# Image: [-1, 1] --> [0, 1]
def M1P1_ZeroP1(imgs):
imgs = (imgs + 1) * 0.5
return np.clip(imgs, 0, 1)
# Image: [0, 1] --> [-1, 1]
def ZeroP1_M1P1(imgs):
return imgs * 2 - 1
In [ ]:
from nw.CycleGAN import PairDataset
pair_flow = PairDataset(train_imgA_paths, train_imgB_paths, target_size=(IMAGE_SIZE, IMAGE_SIZE))
test_pair_flow = PairDataset(test_imgA_paths, test_imgB_paths, target_size=(IMAGE_SIZE, IMAGE_SIZE))
In [ ]:
from nw.CycleGAN import CycleGAN
gan = CycleGAN.load(save_path, 350)
print(gan.epoch)
In [ ]:
# Display the graph of losses in training
%matplotlib inline
gan.showLoss()
In [ ]:
# Display images
# 画像を表示する。
%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np
def showImages(imgs, rows=-1, cols=-1, w=2, h=2):
N = len(imgs)
if rows < 0: rows = 1
if cols < 0: cols = (N + rows -1) // rows
fig, ax = plt.subplots(rows, cols, figsize=(w*cols, h*rows))
idx = 0
for row in range(rows):
for col in range(cols) :
if rows == 1 and cols == 1:
axis = ax
elif rows == 1:
axis = ax[col]
elif cols == 1:
axis = ax[row]
else:
axis = ax[row][col]
if idx < N:
axis.imshow(imgs[idx])
axis.axis('off')
idx += 1
plt.show()
In [ ]:
# Display generated and cycle images.
# 生成画像とサイクル画像を表示する。
test_pairs = test_pair_flow[:5]
test_imgsA = test_pairs[:,0]
test_imgsB = test_pairs[:,1]
imgsAB = gan.generate_image_from_A(test_imgsA)
imgsBA = gan.generate_image_from_B(test_imgsB)
print('A-->B-->A, ID')
showImages(M1P1_ZeroP1(imgsAB), 4)
print('B-->A-->B, ID')
showImages(M1P1_ZeroP1(imgsBA), 4)
In [ ]:
from nw.CycleGAN import CycleGAN
test_pairs = test_pair_flow[:5]
test_imgsA = test_pairs[:,0]
test_imgsB = test_pairs[:,1]
showImages(M1P1_ZeroP1(np.concatenate([test_imgsA, test_imgsB],axis=0)), 2)
for epoch in [200, 250, 300, 350, 400, 450, 500, 550, 600]:
gan = CycleGAN.load(save_path, epoch)
print(gan.epoch)
imgsAB = gan.generate_image_from_A(test_imgsA)[5:10]
imgsBA = gan.generate_image_from_B(test_imgsB)[5:10]
showImages(M1P1_ZeroP1(np.concatenate([imgsAB, imgsBA],axis=0)), 2)
In [ ]: