1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681
|
<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<link rel="canonical" href="http://keras.io/examples/neural_doodle/">
<link rel="shortcut icon" href="../../img/favicon.ico">
<title>Neural doodle - Keras Documentation</title>
<link href='https://fonts.googleapis.com/css?family=Lato:400,700|Source+Sans+Pro:400,700|Inconsolata:400,700' rel='stylesheet' type='text/css'>
<link rel="stylesheet" href="../../css/theme.css" type="text/css" />
<link rel="stylesheet" href="../../css/theme_extra.css" type="text/css" />
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/9.12.0/styles/github.min.css">
<script>
// Current page data
var mkdocs_page_name = "Neural doodle";
var mkdocs_page_input_path = "examples/neural_doodle.md";
var mkdocs_page_url = "/examples/neural_doodle/";
</script>
<script src="../../js/jquery-2.1.1.min.js" defer></script>
<script src="../../js/modernizr-2.8.3.min.js" defer></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/9.12.0/highlight.min.js"></script>
<script>hljs.initHighlightingOnLoad();</script>
<script>
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','https://www.google-analytics.com/analytics.js','ga');
ga('create', 'UA-61785484-1', 'keras.io');
ga('send', 'pageview');
</script>
</head>
<body class="wy-body-for-nav" role="document">
<div class="wy-grid-for-nav">
<nav data-toggle="wy-nav-shift" class="wy-nav-side stickynav">
<div class="wy-side-scroll">
<a href="">
<div class="keras-logo">
<img src="/img/keras-logo-small.jpg" class="keras-logo-img">
Keras Documentation
</div>
</a>
<div class="wy-side-nav-search">
<div role="search">
<form id ="rtd-search-form" class="wy-form" action="../../search.html" method="get">
<input type="text" name="q" placeholder="Search docs" title="Type search term here" />
</form>
</div>
</div>
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
<ul>
<li class="toctree-l1"><a class="reference internal" href="../..">Home</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../why-use-keras/">Why use Keras</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Getting started</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../getting-started/sequential-model-guide/">Guide to the Sequential model</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../getting-started/functional-api-guide/">Guide to the Functional API</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../getting-started/faq/">FAQ</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Models</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../models/about-keras-models/">About Keras models</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../models/sequential/">Sequential</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../models/model/">Model (functional API)</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Layers</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../layers/about-keras-layers/">About Keras layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/core/">Core Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/convolutional/">Convolutional Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/pooling/">Pooling Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/local/">Locally-connected Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/recurrent/">Recurrent Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/embeddings/">Embedding Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/merge/">Merge Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/advanced-activations/">Advanced Activations Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/normalization/">Normalization Layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/noise/">Noise layers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/wrappers/">Layer wrappers</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../layers/writing-your-own-keras-layers/">Writing your own Keras layers</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Preprocessing</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../preprocessing/sequence/">Sequence Preprocessing</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../preprocessing/text/">Text Preprocessing</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../preprocessing/image/">Image Preprocessing</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../losses/">Losses</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../metrics/">Metrics</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../optimizers/">Optimizers</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../activations/">Activations</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../callbacks/">Callbacks</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../datasets/">Datasets</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../applications/">Applications</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../backend/">Backend</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../initializers/">Initializers</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../regularizers/">Regularizers</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../constraints/">Constraints</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../visualization/">Visualization</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../scikit-learn-api/">Scikit-learn API</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../utils/">Utils</a>
</li>
</ul>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../contributing/">Contributing</a>
</li>
</ul>
<p class="caption"><span class="caption-text">Examples</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../addition_rnn/">Addition RNN</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../antirectifier/">Custom layer - antirectifier</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../babi_rnn/">Baby RNN</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../babi_memnn/">Baby MemNN</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../cifar10_cnn/">CIFAR-10 CNN</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../cifar10_resnet/">CIFAR-10 ResNet</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../conv_filter_visualization/">Convolution filter visualization</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../conv_lstm/">Convolutional LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../deep_dream/">Deep Dream</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../image_ocr/">Image OCR</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../imdb_bidirectional_lstm/">Bidirectional LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../imdb_cnn/">1D CNN for text classification</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../imdb_cnn_lstm/">Sentiment classification CNN-LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../imdb_fasttext/">Fasttext for text classification</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../imdb_lstm/">Sentiment classification LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../lstm_seq2seq/">Sequence to sequence - training</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../lstm_seq2seq_restore/">Sequence to sequence - prediction</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../lstm_stateful/">Stateful LSTM</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../lstm_text_generation/">LSTM for text generation</a>
</li>
<li class="toctree-l1"><a class="reference internal" href="../mnist_acgan/">Auxiliary Classifier GAN</a>
</li>
</ul>
</div>
</div>
</nav>
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
<nav class="wy-nav-top" role="navigation" aria-label="top navigation">
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
<a href="../..">Keras Documentation</a>
</nav>
<div class="wy-nav-content">
<div class="rst-content">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="wy-breadcrumbs">
<li><a href="../..">Docs</a> »</li>
<li>Neural doodle</li>
<li class="wy-breadcrumbs-aside">
<a href="https://github.com/keras-team/keras/tree/master/docs"
class="icon icon-github"> Edit on GitHub</a>
</li>
</ul>
<hr/>
</div>
<div role="main">
<div class="section">
<p>Neural doodle with Keras</p>
<h1 id="script-usage">Script Usage</h1>
<h2 id="arguments">Arguments</h2>
<pre><code>--nlabels: # of regions (colors) in mask images
--style-image: image to learn style from
--style-mask: semantic labels for style image
--target-mask: semantic labels for target image (your doodle)
--content-image: optional image to learn content from
--target-image-prefix: path prefix for generated target images
</code></pre>
<h2 id="example-1-doodle-using-a-style-image-style-mask">Example 1: doodle using a style image, style mask</h2>
<p>and target mask.</p>
<pre><code>python neural_doodle.py --nlabels 4 --style-image Monet/style.png --style-mask Monet/style_mask.png --target-mask Monet/target_mask.png --target-image-prefix generated/monet
</code></pre>
<h2 id="example-2-doodle-using-a-style-image-style-mask">Example 2: doodle using a style image, style mask,</h2>
<p>target mask and an optional content image.</p>
<pre><code>python neural_doodle.py --nlabels 4 --style-image Renoir/style.png --style-mask Renoir/style_mask.png --target-mask Renoir/target_mask.png --content-image Renoir/creek.jpg --target-image-prefix generated/renoir
</code></pre>
<h1 id="references">References</h1>
<ul>
<li>[Dmitry Ulyanov's blog on fast-neural-doodle]
(http://dmitryulyanov.github.io/feed-forward-neural-doodle/)</li>
<li>[Torch code for fast-neural-doodle]
(https://github.com/DmitryUlyanov/fast-neural-doodle)</li>
<li>[Torch code for online-neural-doodle]
(https://github.com/DmitryUlyanov/online-neural-doodle)</li>
<li>[Paper Texture Networks: Feed-forward Synthesis of Textures and Stylized Images]
(http://arxiv.org/abs/1603.03417)</li>
<li>[Discussion on parameter tuning]
(https://github.com/keras-team/keras/issues/3705)</li>
</ul>
<h1 id="resources">Resources</h1>
<p>Example images can be downloaded from
https://github.com/DmitryUlyanov/fast-neural-doodle/tree/master/data</p>
<pre><code class="python">from __future__ import print_function
import time
import argparse
import numpy as np
from scipy.optimize import fmin_l_bfgs_b
from keras import backend as K
from keras.layers import Input, AveragePooling2D
from keras.models import Model
from keras.preprocessing.image import load_img, save_img, img_to_array
from keras.applications import vgg19
# Command line arguments
parser = argparse.ArgumentParser(description='Keras neural doodle example')
parser.add_argument('--nlabels', type=int,
help='number of semantic labels'
' (regions in differnet colors)'
' in style_mask/target_mask')
parser.add_argument('--style-image', type=str,
help='path to image to learn style from')
parser.add_argument('--style-mask', type=str,
help='path to semantic mask of style image')
parser.add_argument('--target-mask', type=str,
help='path to semantic mask of target image')
parser.add_argument('--content-image', type=str, default=None,
help='path to optional content image')
parser.add_argument('--target-image-prefix', type=str,
help='path prefix for generated results')
args = parser.parse_args()
style_img_path = args.style_image
style_mask_path = args.style_mask
target_mask_path = args.target_mask
content_img_path = args.content_image
target_img_prefix = args.target_image_prefix
use_content_img = content_img_path is not None
num_labels = args.nlabels
num_colors = 3 # RGB
# determine image sizes based on target_mask
ref_img = img_to_array(load_img(target_mask_path))
img_nrows, img_ncols = ref_img.shape[:2]
num_iterations = 50
total_variation_weight = 50.
style_weight = 1.
content_weight = 0.1 if use_content_img else 0
content_feature_layers = ['block5_conv2']
# To get better generation qualities, use more conv layers for style features
style_feature_layers = ['block1_conv1', 'block2_conv1', 'block3_conv1',
'block4_conv1', 'block5_conv1']
# helper functions for reading/processing images
def preprocess_image(image_path):
img = load_img(image_path, target_size=(img_nrows, img_ncols))
img = img_to_array(img)
img = np.expand_dims(img, axis=0)
img = vgg19.preprocess_input(img)
return img
def deprocess_image(x):
if K.image_data_format() == 'channels_first':
x = x.reshape((3, img_nrows, img_ncols))
x = x.transpose((1, 2, 0))
else:
x = x.reshape((img_nrows, img_ncols, 3))
# Remove zero-center by mean pixel
x[:, :, 0] += 103.939
x[:, :, 1] += 116.779
x[:, :, 2] += 123.68
# 'BGR'->'RGB'
x = x[:, :, ::-1]
x = np.clip(x, 0, 255).astype('uint8')
return x
def kmeans(xs, k):
assert xs.ndim == 2
try:
from sklearn.cluster import k_means
_, labels, _ = k_means(xs.astype('float64'), k)
except ImportError:
from scipy.cluster.vq import kmeans2
_, labels = kmeans2(xs, k, missing='raise')
return labels
def load_mask_labels():
'''Load both target and style masks.
A mask image (nr x nc) with m labels/colors will be loaded
as a 4D boolean tensor:
(1, m, nr, nc) for 'channels_first' or (1, nr, nc, m) for 'channels_last'
'''
target_mask_img = load_img(target_mask_path,
target_size=(img_nrows, img_ncols))
target_mask_img = img_to_array(target_mask_img)
style_mask_img = load_img(style_mask_path,
target_size=(img_nrows, img_ncols))
style_mask_img = img_to_array(style_mask_img)
if K.image_data_format() == 'channels_first':
mask_vecs = np.vstack([style_mask_img.reshape((3, -1)).T,
target_mask_img.reshape((3, -1)).T])
else:
mask_vecs = np.vstack([style_mask_img.reshape((-1, 3)),
target_mask_img.reshape((-1, 3))])
labels = kmeans(mask_vecs, num_labels)
style_mask_label = labels[:img_nrows *
img_ncols].reshape((img_nrows, img_ncols))
target_mask_label = labels[img_nrows *
img_ncols:].reshape((img_nrows, img_ncols))
stack_axis = 0 if K.image_data_format() == 'channels_first' else -1
style_mask = np.stack([style_mask_label == r for r in range(num_labels)],
axis=stack_axis)
target_mask = np.stack([target_mask_label == r for r in range(num_labels)],
axis=stack_axis)
return (np.expand_dims(style_mask, axis=0),
np.expand_dims(target_mask, axis=0))
# Create tensor variables for images
if K.image_data_format() == 'channels_first':
shape = (1, num_colors, img_nrows, img_ncols)
else:
shape = (1, img_nrows, img_ncols, num_colors)
style_image = K.variable(preprocess_image(style_img_path))
target_image = K.placeholder(shape=shape)
if use_content_img:
content_image = K.variable(preprocess_image(content_img_path))
else:
content_image = K.zeros(shape=shape)
images = K.concatenate([style_image, target_image, content_image], axis=0)
# Create tensor variables for masks
raw_style_mask, raw_target_mask = load_mask_labels()
style_mask = K.variable(raw_style_mask.astype('float32'))
target_mask = K.variable(raw_target_mask.astype('float32'))
masks = K.concatenate([style_mask, target_mask], axis=0)
# index constants for images and tasks variables
STYLE, TARGET, CONTENT = 0, 1, 2
# Build image model, mask model and use layer outputs as features
# image model as VGG19
image_model = vgg19.VGG19(include_top=False, input_tensor=images)
# mask model as a series of pooling
mask_input = Input(tensor=masks, shape=(None, None, None), name='mask_input')
x = mask_input
for layer in image_model.layers[1:]:
name = 'mask_%s' % layer.name
if 'conv' in layer.name:
x = AveragePooling2D((3, 3), padding='same', strides=(
1, 1), name=name)(x)
elif 'pool' in layer.name:
x = AveragePooling2D((2, 2), name=name)(x)
mask_model = Model(mask_input, x)
# Collect features from image_model and task_model
image_features = {}
mask_features = {}
for img_layer, mask_layer in zip(image_model.layers, mask_model.layers):
if 'conv' in img_layer.name:
assert 'mask_' + img_layer.name == mask_layer.name
layer_name = img_layer.name
img_feat, mask_feat = img_layer.output, mask_layer.output
image_features[layer_name] = img_feat
mask_features[layer_name] = mask_feat
# Define loss functions
def gram_matrix(x):
assert K.ndim(x) == 3
features = K.batch_flatten(x)
gram = K.dot(features, K.transpose(features))
return gram
def region_style_loss(style_image, target_image, style_mask, target_mask):
'''Calculate style loss between style_image and target_image,
for one common region specified by their (boolean) masks
'''
assert 3 == K.ndim(style_image) == K.ndim(target_image)
assert 2 == K.ndim(style_mask) == K.ndim(target_mask)
if K.image_data_format() == 'channels_first':
masked_style = style_image * style_mask
masked_target = target_image * target_mask
num_channels = K.shape(style_image)[0]
else:
masked_style = K.permute_dimensions(
style_image, (2, 0, 1)) * style_mask
masked_target = K.permute_dimensions(
target_image, (2, 0, 1)) * target_mask
num_channels = K.shape(style_image)[-1]
num_channels = K.cast(num_channels, dtype='float32')
s = gram_matrix(masked_style) / K.mean(style_mask) / num_channels
c = gram_matrix(masked_target) / K.mean(target_mask) / num_channels
return K.mean(K.square(s - c))
def style_loss(style_image, target_image, style_masks, target_masks):
'''Calculate style loss between style_image and target_image,
in all regions.
'''
assert 3 == K.ndim(style_image) == K.ndim(target_image)
assert 3 == K.ndim(style_masks) == K.ndim(target_masks)
loss = K.variable(0)
for i in range(num_labels):
if K.image_data_format() == 'channels_first':
style_mask = style_masks[i, :, :]
target_mask = target_masks[i, :, :]
else:
style_mask = style_masks[:, :, i]
target_mask = target_masks[:, :, i]
loss = loss + region_style_loss(style_image,
target_image,
style_mask,
target_mask)
return loss
def content_loss(content_image, target_image):
return K.sum(K.square(target_image - content_image))
def total_variation_loss(x):
assert 4 == K.ndim(x)
if K.image_data_format() == 'channels_first':
a = K.square(x[:, :, :img_nrows - 1, :img_ncols - 1] -
x[:, :, 1:, :img_ncols - 1])
b = K.square(x[:, :, :img_nrows - 1, :img_ncols - 1] -
x[:, :, :img_nrows - 1, 1:])
else:
a = K.square(x[:, :img_nrows - 1, :img_ncols - 1, :] -
x[:, 1:, :img_ncols - 1, :])
b = K.square(x[:, :img_nrows - 1, :img_ncols - 1, :] -
x[:, :img_nrows - 1, 1:, :])
return K.sum(K.pow(a + b, 1.25))
# Overall loss is the weighted sum of content_loss, style_loss and tv_loss
# Each individual loss uses features from image/mask models.
loss = K.variable(0)
for layer in content_feature_layers:
content_feat = image_features[layer][CONTENT, :, :, :]
target_feat = image_features[layer][TARGET, :, :, :]
loss = loss + content_weight * content_loss(content_feat, target_feat)
for layer in style_feature_layers:
style_feat = image_features[layer][STYLE, :, :, :]
target_feat = image_features[layer][TARGET, :, :, :]
style_masks = mask_features[layer][STYLE, :, :, :]
target_masks = mask_features[layer][TARGET, :, :, :]
sl = style_loss(style_feat, target_feat, style_masks, target_masks)
loss = loss + (style_weight / len(style_feature_layers)) * sl
loss = loss + total_variation_weight * total_variation_loss(target_image)
loss_grads = K.gradients(loss, target_image)
# Evaluator class for computing efficiency
outputs = [loss]
if isinstance(loss_grads, (list, tuple)):
outputs += loss_grads
else:
outputs.append(loss_grads)
f_outputs = K.function([target_image], outputs)
def eval_loss_and_grads(x):
if K.image_data_format() == 'channels_first':
x = x.reshape((1, 3, img_nrows, img_ncols))
else:
x = x.reshape((1, img_nrows, img_ncols, 3))
outs = f_outputs([x])
loss_value = outs[0]
if len(outs[1:]) == 1:
grad_values = outs[1].flatten().astype('float64')
else:
grad_values = np.array(outs[1:]).flatten().astype('float64')
return loss_value, grad_values
class Evaluator(object):
def __init__(self):
self.loss_value = None
self.grads_values = None
def loss(self, x):
assert self.loss_value is None
loss_value, grad_values = eval_loss_and_grads(x)
self.loss_value = loss_value
self.grad_values = grad_values
return self.loss_value
def grads(self, x):
assert self.loss_value is not None
grad_values = np.copy(self.grad_values)
self.loss_value = None
self.grad_values = None
return grad_values
evaluator = Evaluator()
# Generate images by iterative optimization
if K.image_data_format() == 'channels_first':
x = np.random.uniform(0, 255, (1, 3, img_nrows, img_ncols)) - 128.
else:
x = np.random.uniform(0, 255, (1, img_nrows, img_ncols, 3)) - 128.
for i in range(num_iterations):
print('Start of iteration', i, '/', num_iterations)
start_time = time.time()
x, min_val, info = fmin_l_bfgs_b(evaluator.loss, x.flatten(),
fprime=evaluator.grads, maxfun=20)
print('Current loss value:', min_val)
# save current generated image
img = deprocess_image(x.copy())
fname = target_img_prefix + '_at_iteration_%d.png' % i
save_img(fname, img)
end_time = time.time()
print('Image saved as', fname)
print('Iteration %d completed in %ds' % (i, end_time - start_time))
</code></pre>
</div>
</div>
<footer>
<hr/>
<div role="contentinfo">
<!-- Copyright etc -->
</div>
Built with <a href="https://www.mkdocs.org/">MkDocs</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
</footer>
</div>
</div>
</section>
</div>
<div class="rst-versions" role="note" aria-label="versions">
<span class="rst-current-version" data-toggle="rst-current-version">
<a href="http://github.com/keras-team/keras/" class="fa fa-github" style="float: left; color: #fcfcfc"> GitHub</a>
</span>
</div>
<script>var base_url = '../..';</script>
<script src="../../js/theme.js" defer></script>
<script src="../../search/main.js" defer></script>
<script type="text/javascript" defer>
window.onload = function () {
SphinxRtdTheme.Navigation.enable(true);
};
</script>
</body>
</html>
|