本コードは、Google Colab、あるいは、環境構築済みのpythonで動かすことができます。また、Google Colabのサンプルプログラムを修正したものです。
2, 3行目のcontent_path, style_pathに対して、自分が指定したい画像ファイル名を書き込む。
tips: 絶対パスとは、ディレクトリ（フォルダ）も含めたファイル名指定のこと。今回は、tmpディレクトリに画像を格納しているので、「/tmp/xxx.jpg」のような指定となる。
# ここにコンテンツ画像とスタイル画像の絶対パスを入力
content_path = "/tmp/Sample_Content_cit.jpg"
style_path = "/tmp/Sample_Style_guruguru.jpg"
# png -> jpg 拡張子変換 / ここは触らないこと！
# コンテンツ画像
from PIL import Image
for i in range(len(content_path)-2):
if content_path[i]+content_path[i+1]+content_path[i+2] == "png":
print("コンテンツ画像は png 拡張子だったため、jpgに変換します")
input_im = Image.open(content_path)
rgb_im = input_im.convert('RGB')
jpgfilename = content_path[:i] + "jpg"
rgb_im.save(jpgfilename, quality=30)
content_path = jpgfilename
# コンテンツ画像
from PIL import Image
for i in range(len(style_path)-2):
if style_path[i]+style_path[i+1]+style_path[i+2] == "png":
print("スタイル画像は png 拡張子だったため、jpgに変換します")
input_im = Image.open(style_path)
rgb_im = input_im.convert('RGB')
jpgfilename = style_path[:i] + "jpg"
rgb_im.save(jpgfilename, quality=30)
style_path = jpgfilename
これはどこも触らないでください。
import matplotlib.pyplot as plt
import matplotlib as mpl
mpl.rcParams['figure.figsize'] = (10,10)
mpl.rcParams['axes.grid'] = False
import numpy as np
from PIL import Image
import time
import functools
import tensorflow as tf
from tensorflow.python.keras.preprocessing import image as kp_image
from tensorflow.python.keras import models
from tensorflow.python.keras import losses
from tensorflow.python.keras import layers
from tensorflow.python.keras import backend as K
def load_img(path_to_img):
max_dim = 512
img = Image.open(path_to_img)
long = max(img.size)
scale = max_dim/long
img = img.resize((round(img.size[0]*scale), round(img.size[1]*scale)), Image.ANTIALIAS)
img = kp_image.img_to_array(img)
# We need to broadcast the image array such that it has a batch dimension
img = np.expand_dims(img, axis=0)
return img
def imshow(img, title=None):
# Remove the batch dimension
out = np.squeeze(img, axis=0)
# Normalize for display
out = out.astype('uint8')
plt.imshow(out)
if title is not None:
plt.title(title)
plt.imshow(out)
plt.figure(figsize=(10,10))
content = load_img(content_path).astype('uint8')
style = load_img(style_path).astype('uint8')
plt.subplot(1, 2, 1)
imshow(content, 'Content Image')
plt.subplot(1, 2, 2)
imshow(style, 'Style Image')
plt.show()
def load_and_process_img(path_to_img):
img = load_img(path_to_img)
img = tf.keras.applications.vgg19.preprocess_input(img)
return img
def deprocess_img(processed_img):
x = processed_img.copy()
if len(x.shape) == 4:
x = np.squeeze(x, 0)
assert len(x.shape) == 3, ("Input to deprocess image must be an image of "
"dimension [1, height, width, channel] or [height, width, channel]")
if len(x.shape) != 3:
raise ValueError("Invalid input to deprocessing image")
# perform the inverse of the preprocessiing step
x[:, :, 0] += 103.939
x[:, :, 1] += 116.779
x[:, :, 2] += 123.68
x = x[:, :, ::-1]
x = np.clip(x, 0, 255).astype('uint8')
return x
# Content layer where will pull our feature maps
content_layers = ['block5_conv2']
# Style layer we are interested in
style_layers = ['block1_conv1',
'block2_conv1',
'block3_conv1',
'block4_conv1',
'block5_conv1'
]
num_content_layers = len(content_layers)
num_style_layers = len(style_layers)
def get_model():
""" Creates our model with access to intermediate layers.
This function will load the VGG19 model and access the intermediate layers.
These layers will then be used to create a new model that will take input image
and return the outputs from these intermediate layers from the VGG model.
Returns:
returns a keras model that takes image inputs and outputs the style and
content intermediate layers.
"""
# Load our model. We load pretrained VGG, trained on imagenet data
vgg = tf.keras.applications.vgg19.VGG19(include_top=False, weights='imagenet')
vgg.trainable = False
# Get output layers corresponding to style and content layers
style_outputs = [vgg.get_layer(name).output for name in style_layers]
content_outputs = [vgg.get_layer(name).output for name in content_layers]
model_outputs = style_outputs + content_outputs
# Build model
return models.Model(vgg.input, model_outputs)
def get_content_loss(base_content, target):
return tf.reduce_mean(tf.square(base_content - target))
def gram_matrix(input_tensor):
# We make the image channels first
channels = int(input_tensor.shape[-1])
a = tf.reshape(input_tensor, [-1, channels])
n = tf.shape(a)[0]
gram = tf.matmul(a, a, transpose_a=True)
return gram / tf.cast(n, tf.float32)
def get_style_loss(base_style, gram_target):
"""Expects two images of dimension h, w, c"""
# height, width, num filters of each layer
# We scale the loss at a given layer by the size of the feature map and the number of filters
height, width, channels = base_style.get_shape().as_list()
gram_style = gram_matrix(base_style)
return tf.reduce_mean(tf.square(gram_style - gram_target))# / (4. * (channels ** 2) * (width * height) ** 2)
def get_feature_representations(model, content_path, style_path):
"""Helper function to compute our content and style feature representations.
This function will simply load and preprocess both the content and style
images from their path. Then it will feed them through the network to obtain
the outputs of the intermediate layers.
Arguments:
model: The model that we are using.
content_path: The path to the content image.
style_path: The path to the style image
Returns:
returns the style features and the content features.
"""
# Load our images in
content_image = load_and_process_img(content_path)
style_image = load_and_process_img(style_path)
# batch compute content and style features
style_outputs = model(style_image)
content_outputs = model(content_image)
# Get the style and content feature representations from our model
style_features = [style_layer[0] for style_layer in style_outputs[:num_style_layers]]
content_features = [content_layer[0] for content_layer in content_outputs[num_style_layers:]]
return style_features, content_features
def compute_loss(model, loss_weights, init_image, gram_style_features, content_features):
"""This function will compute the loss total loss.
Arguments:
model: The model that will give us access to the intermediate layers
loss_weights: The weights of each contribution of each loss function.
(style weight, content weight, and total variation weight)
init_image: Our initial base image. This image is what we are updating with
our optimization process. We apply the gradients wrt the loss we are
calculating to this image.
gram_style_features: Precomputed gram matrices corresponding to the
defined style layers of interest.
content_features: Precomputed outputs from defined content layers of
interest.
Returns:
returns the total loss, style loss, content loss, and total variational loss
"""
style_weight, content_weight = loss_weights
# Feed our init image through our model. This will give us the content and
# style representations at our desired layers. Since we're using eager
# our model is callable just like any other function!
model_outputs = model(init_image)
style_output_features = model_outputs[:num_style_layers]
content_output_features = model_outputs[num_style_layers:]
style_score = 0
content_score = 0
# Accumulate style losses from all layers
# Here, we equally weight each contribution of each loss layer
weight_per_style_layer = 1.0 / float(num_style_layers)
for target_style, comb_style in zip(gram_style_features, style_output_features):
style_score += weight_per_style_layer * get_style_loss(comb_style[0], target_style)
# Accumulate content losses from all layers
weight_per_content_layer = 1.0 / float(num_content_layers)
for target_content, comb_content in zip(content_features, content_output_features):
content_score += weight_per_content_layer* get_content_loss(comb_content[0], target_content)
style_score *= style_weight
content_score *= content_weight
# Get total loss
loss = style_score + content_score
return loss, style_score, content_score
def compute_grads(cfg):
with tf.GradientTape() as tape:
all_loss = compute_loss(**cfg)
# Compute gradients wrt input image
total_loss = all_loss[0]
return tape.gradient(total_loss, cfg['init_image']), all_loss
# 以下以降、具体的な関数を使用して学習の設定
import IPython.display
def run_style_transfer(content_path,
style_path,
LR = 5,
num_iterations=1000,
content_weight=1e3,
style_weight=1e-2):
# We don't need to (or want to) train any layers of our model, so we set their
# trainable to false.
model = get_model()
for layer in model.layers:
layer.trainable = False
# Get the style and content feature representations (from our specified intermediate layers)
style_features, content_features = get_feature_representations(model, content_path, style_path)
gram_style_features = [gram_matrix(style_feature) for style_feature in style_features]
# Set initial image
init_image = load_and_process_img(content_path)
init_image = tf.Variable(init_image, dtype=tf.float32)
# Create our optimizer
#opt = tf.train.AdamOptimizer(learning_rate=LR, beta1=0.99, epsilon=1e-1) # エラーが出るので変更
opt = tf.optimizers.Adam(learning_rate=LR, epsilon=1e-1)
# For displaying intermediate images
iter_count = 1
# Store our best result
best_loss, best_img = float('inf'), None
# Create a nice config
loss_weights = (style_weight, content_weight)
cfg = {
'model': model,
'loss_weights': loss_weights,
'init_image': init_image,
'gram_style_features': gram_style_features,
'content_features': content_features
}
# For displaying
num_rows = 2
num_cols = 5
display_interval = num_iterations/(num_rows*num_cols)
start_time = time.time()
global_start = time.time()
norm_means = np.array([103.939, 116.779, 123.68])
min_vals = -norm_means
max_vals = 255 - norm_means
imgs = []
for i in range(num_iterations):
start_time = time.time() # 開始
grads, all_loss = compute_grads(cfg)
loss, style_score, content_score = all_loss
opt.apply_gradients([(grads, init_image)])
clipped = tf.clip_by_value(init_image, min_vals, max_vals)
init_image.assign(clipped)
end_time = time.time()
if loss < best_loss:
# Update best loss and best image from total loss.
best_loss = loss
best_img = deprocess_img(init_image.numpy())
if i % display_interval== 0:
# Use the .numpy() method to get the concrete numpy array
plot_img = init_image.numpy()
plot_img = deprocess_img(plot_img)
imgs.append(plot_img)
IPython.display.clear_output(wait=True)
IPython.display.display_png(Image.fromarray(plot_img))
# 状況
print('Iteration: {}'.format(i))
print('Total loss: {:.4e}, '
'style loss: {:.4e}, '
'content loss: {:.4e}, '
'time: {:.4f}s'.format(loss, style_score, content_score, time.time() - start_time))
print('Total time: {:.4f}s'.format(time.time() - global_start))
# 反復for文の外
IPython.display.clear_output(wait=True)
plt.figure(figsize=(14,4))
for i,img in enumerate(imgs):
plt.subplot(num_rows,num_cols,i+1)
plt.imshow(img)
plt.xticks([])
plt.yticks([])
return best_img, best_loss
LR（学習係数）と num_iterations（学習回数）を指定して、実行します。学習係数とは、1回の学習に対する人工知能の思考の変化の大きさ、学習回数とは思考を変化させる回数を表します。それぞれ、以下の特徴路有します。
LR（学習係数）: 3〜10程度。大きくすると学習が早まるものの、結果が大雑把な学習になり、うまくいかない場合も多い。小さくすると、学習が精密になるが、学習回数を多くしなければならなくなる。
num_iterations（学習回数）: 仮で実施したい場合は2〜3千回、本格的に実施したい場合は2〜3万回を標準とします。ただしこれは、LR（学習係数）が5の場合です。LRを変化させた場合は、必要な学習回数も変わってきます。
なお、参考値としては、
くらいです。仮で結果を見てみたい場合は学習回数を2千回程度、本格的にやりたい場合は2万回にしましょう。
best, best_loss = run_style_transfer(content_path,
style_path,
LR = 5, # 学習係数（3〜10程度）
num_iterations = 3000 # 学習回数
)
学習結果の画像を出力するコードです。右クリックで、自分のパソコンに保存できると思います。
plt.imshow(best)
plt.tick_params(labelbottom=False,
labelleft=False,
labelright=False,
labeltop=False)
plt.tick_params(bottom=False,
left=False,
right=False,
top=False)
plt.savefig("/tmp/output.jpg")