Functions — github.com/YoungGod/sturcture-inpainting

hub / github.com/YoungGod/sturcture-inpainting / functions

Functions209 in github.com/YoungGod/sturcture-inpainting

⨍ Functions 209◇ Types & classes 19

↓ 17 callersFunctiongen_conv

Define conv for generator. Args: x: Input. cnum: Channel number. ksize: Kernel size. Stride: Convolution st

src/ops.py:9

↓ 17 callersFunctiongen_conv

Define conv for generator. Args: x: Input. cnum: Channel number. ksize: Kernel size. Stride: Convolution st

painter/ops.py:8

↓ 13 callersFunctioninverse_transform

(images)

src/ops.py:445

↓ 7 callersFunctionresize

(x, scale=2, to_shape=None, align_corners=True, dynamic=False, func=tf.image.resize_bilinear, name

painter/ops.py:110

↓ 6 callersMethodl1_loss

(self, image, predict, mask, type='foreground')

src/loss.py:110

↓ 6 callersFunctionpriority_loss_mask

(mask, ksize=5, sigma=1, iteration=2)

src/loss.py:572

↓ 5 callersFunctiondis_conv

Define conv for discriminator. Activation is set to leaky_relu. Args: x: Input. cnum: Channel number. ksize: Ke

src/ops.py:215

↓ 5 callersFunctiongrad_matching_loss

(im_src, im_dst, config)

src/loss.py:542

↓ 5 callersFunctionimsave

(images, size, path)

src/ops.py:442

↓ 5 callersMethodload

(self)

painter/painter.py:119

↓ 4 callersMethodbuild_inpaint_net

Inpaint network. Args: x: incomplete image[-1, 1] with shape of (batch_size, h, w, c) edge: incomplete edge {0, 1

src/inpaint_model.py:26

↓ 4 callersFunctiongen_deconv

Define deconv for generator. The deconv is defined to be a x2 resize_nearest_neighbor operation with additional gen_conv operation. A

src/ops.py:64

↓ 4 callersFunctiongen_deconv

Define deconv for generator. The deconv is defined to be a x2 resize_nearest_neighbor operation with additional gen_conv operation. A

painter/ops.py:63

↓ 3 callersMethod__calculate_CS

(self, scaled_distances, axis_for_normalization=3)

src/loss.py:295

↓ 3 callersFunctionattention

(x, channels, neighbors=1, use_bias=True, sn=False, down_scale = 2, pool_scale=2, name='attent

src/ops.py:339

↓ 3 callersFunctionattention

(x, channels, neighbors=1, use_bias=True, sn=False, down_scale = 2, pool_scale=2, name='attent

painter/ops.py:432

↓ 3 callersFunctionavg_l1

(x, pred_x)

src/metrics.py:24

↓ 3 callersMethodcalc_relative_distances

(self, axis=3)

src/loss.py:377

↓ 3 callersFunctioncalculate_fid_given_paths

Calculates the FID of two paths.

src/frechet_inception_distance.py:215

↓ 3 callersFunctionconv

(x, channels, kernel=3, stride=1, pad=0, pad_type='REFLECT', use_bias=True, sn=False, scope='conv_0',

src/ops.py:270

↓ 3 callersFunctionconv

(x, channels, kernel=3, stride=1, pad=0, pad_type='REFLECT', use_bias=True, sn=False, scope='conv_0', reuse=Fa

painter/ops.py:270

↓ 3 callersFunctionhw_flatten

(x)

src/ops.py:324

↓ 3 callersFunctionhw_flatten

(x)

painter/ops.py:323

↓ 3 callersFunctionmerge

src/vgg_network.py:38

↓ 3 callersFunctionpsnr

PSNR

src/metrics.py:3

↓ 3 callersFunctionresize

(x, scale=2, to_shape=None, align_corners=True, dynamic=False, func=tf.image.resize_bilinear, name

src/ops.py:111

↓ 3 callersFunctionspectral_norm

(w, iteration=1)

src/ops.py:294

↓ 3 callersFunctionspectral_norm

(w, iteration=1)

painter/ops.py:293

↓ 3 callersFunctionssmi

SSMI

src/metrics.py:10

↓ 3 callersFunctiontv_loss

(pred_x)

src/metrics.py:28

↓ 2 callersMethod_compute_gram

(self, feature)

src/loss.py:87

↓ 2 callersFunction_handle_path

(path, sess)

src/frechet_inception_distance.py:202

↓ 2 callersFunction_tensor_size

(tensor)

src/loss.py:124

↓ 2 callersMethodactivate_button

(self, some_button, eraser_mode=False)

painter/painter.py:237

↓ 2 callersFunctionaverage_gradients

(tower_grads)

src/train_inpaint_model.py:98

↓ 2 callersMethodbuild_discriminator_256

Patch GAN discriminator component, receptive filed: 70*70

src/inpaint_model.py:102

↓ 2 callersMethodbuild_graph_with_losses

(self, x, mask, edge, edge_128, edge_64, args, training=True, reuse=False)

src/inpaint_model.py:116

↓ 2 callersFunctioncalculate_frechet_distance

Numpy implementation of the Frechet Distance. The Frechet distance between two multivariate Gaussians X_1 ~ N(mu_1, C_1) and X_2 ~ N(mu_2, C_2

src/frechet_inception_distance.py:125

↓ 2 callersMethodcheckResp

(self)

painter/painter.py:116

↓ 2 callersFunctioncheck_or_download_inception

Checks if the path to the inception file is valid, or downloads the file if it is not present.

src/frechet_inception_distance.py:184

↓ 2 callersMethodclear

(self)

painter/painter.py:214

↓ 2 callersMethodcontent_loss

(self, content_fake, layers=None)

src/loss.py:35

↓ 2 callersFunctioncreate_inception_graph

Creates a graph from saved GraphDef file.

src/frechet_inception_distance.py:52

↓ 2 callersFunctioncreate_mask

create_mask(imgw, imgh, imgw // 2, imgh, 0 if random.random() < 0.5 else imgw // 2, 0) delta: margin between mask and image boundary

src/utils_fn.py:279

↓ 2 callersFunctioncreate_validation_mask

create_mask(imgw, imgh, imgw // 2, imgh, 0 if random.random() < 0.5 else imgw // 2, 0)

src/utils_fn.py:341

↓ 2 callersFunctioncrop_quarters

(feature_tensor)

src/loss.py:513

↓ 2 callersFunctiondataset_len

(args)

src/utils_fn.py:415

↓ 2 callersFunctionget_activations

Calculates the activations of the pool_3 layer for all images. Params: -- images : Numpy array of dimension (n_images, hi, wi, 3). The v

src/frechet_inception_distance.py:88

↓ 2 callersFunctioninverse_transform

(images)

painter/ops.py:537

↓ 2 callersFunctionl2_norm

(v, eps=1e-12)

src/ops.py:321

↓ 2 callersFunctionl2_norm

(v, eps=1e-12)

painter/ops.py:320

↓ 2 callersMethodl2_normalize_channelwise

(features)

src/loss.py:428

↓ 2 callersFunctionload_img_scale_edge

Load image data

src/utils_fn.py:546

↓ 2 callersFunctionload_mask

(args)

src/utils_fn.py:249

↓ 2 callersFunctionmax_pooling

(x, pool_size=2)

src/ops.py:327

↓ 2 callersFunctionmax_pooling

(x, pool_size=2)

painter/ops.py:326

↓ 2 callersFunctionpriority_loss_mask

(mask, ksize=5, sigma=1, iteration=2)

structure_loss.py:44

↓ 2 callersFunctionpriority_loss_mask

(mask, ksize=5, sigma=1, iteration=2)

painter/structure_loss.py:44

↓ 2 callersFunctionrandom_sampling

(tensor_in, n, indices=None)

src/loss.py:480

↓ 2 callersMethodsave

(self)

painter/painter.py:152

↓ 2 callersFunctionshow_all_variables

Show all the variables of an tf model.

src/utils_fn.py:420

↓ 2 callersMethodstyle_loss

(self, style_fake, layers=None)

src/loss.py:49

↓ 1 callersMethod_content_loss_helper

(self, vgg_A, vgg_B)

src/loss.py:73

↓ 1 callersFunction_conv_layer_from

(input, weights, bias)

src/vgg_network.py:68

↓ 1 callersFunction_get_inception_layer

Prepares inception net for batched usage and returns pool_3 layer.

src/frechet_inception_distance.py:64

↓ 1 callersFunction_is_convolutional_layer

(name)

src/vgg_network.py:59

↓ 1 callersFunction_is_pooling_layer

(name)

src/vgg_network.py:65

↓ 1 callersFunction_is_relu_layer

(name)

src/vgg_network.py:62

↓ 1 callersFunction_pooling_layer_from

(input)

src/vgg_network.py:73

↓ 1 callersMethod_style_loss_helper

(self, vgg_A, vgg_B)

src/loss.py:79

↓ 1 callersFunctionattention_with_neighbors

(x, channels, ksize=3, use_bias=True, sn=False, stride=2, down_scale = 2, pool_sca

src/ops.py:389

↓ 1 callersFunctionattention_with_neighbors

(x, channels, ksize=3, use_bias=True, sn=False, stride=2, down_scale = 2, pool_sca

painter/ops.py:481

↓ 1 callersFunctionattention_with_pooling

(x, channels, ksize=4, use_bias=True, sn=False, down_scale = 2, pool_scale=2, name

src/ops.py:348

↓ 1 callersFunctionattention_with_pooling

(x, channels, ksize=4, use_bias=True, sn=False, down_scale = 2, pool_scale=2, name='attention_pooling', traini

painter/ops.py:441

↓ 1 callersMethodbuild_inpaint_net

Inpaint network. Args: x: incomplete image[-1, 1] with shape of (batch_size, h, w, c) edge: incomplete edge {0, 1

painter/inpaint_model.py:14

↓ 1 callersMethodbuild_test_model

(self, x, mask, args, training=False, reuse=False)

src/inpaint_model.py:645

↓ 1 callersMethodbuild_validation_model

(self, x, mask, args, training=False, reuse=True)

src/inpaint_model.py:477

↓ 1 callersFunctioncalculate_activation_statistics

Calculation of the statistics used by the FID. Params: -- images : Numpy array of dimension (n_images, hi, wi, 3). The values

src/frechet_inception_distance.py:156

↓ 1 callersMethodcenter_by_T

(self, T_features, I_features)

src/loss.py:407

↓ 1 callersFunctioncompare_mae

(img_true, img_test)

src/evaluation/metrics.py:16

↓ 1 callersMethodcreate

(I_features, T_features, distance : Distance, nnsigma=float(1.0), b=float(1.0), args=None)

src/loss.py:393

↓ 1 callersFunctioncreate_test_mask

create_mask(imgw, imgh, imgw // 2, imgh, 0 if random.random() < 0.5 else imgw // 2, 0)

src/utils_fn.py:403

↓ 1 callersMethodcreate_using_L2

(I_features, T_features, sigma=float(0.1), b=float(1.0))

src/loss.py:310

↓ 1 callersMethodcreate_using_dotP

(I_features, T_features, sigma=float(1.0), b=float(1.0), args=None)

src/loss.py:346

↓ 1 callersFunctionevaluate

(path_true, path_pred, inception_dir = '../inception')

src/evaluation/metrics.py:21

↓ 1 callersMethodevaluate

(self, x, edge, mask, args, training=False, reuse=False)

painter/inpaint_model.py:90

↓ 1 callersFunctionfree_form_mask

(parts, maxVertex=16, maxLength=60, maxBrushWidth=14, maxAngle=360, im_size=(256, 256))

src/utils_fn.py:67

↓ 1 callersFunctiongaussian_kernel_2d_opencv

ref: https://blog.csdn.net/qq_16013649/article/details/78784791 ref: tensorflow (1) https://stackoverflow.com/questions/52012657/how-

structure_loss.py:12

↓ 1 callersFunctiongaussian_kernel_2d_opencv

ref: https://blog.csdn.net/qq_16013649/article/details/78784791 ref: tensorflow (1) https://stackoverflow.com/questions/52012657/how-

src/loss.py:561

↓ 1 callersFunctiongaussian_kernel_2d_opencv

ref: https://blog.csdn.net/qq_16013649/article/details/78784791 ref: tensorflow (1) https://stackoverflow.com/questions/52012657/how-

painter/structure_loss.py:12

↓ 1 callersFunctiongradients_penalty

Improved Training of Wasserstein GANs - https://arxiv.org/abs/1704.00028

src/loss.py:182

↓ 1 callersFunctionid_mrf_reg_feat

(feat_A, feat_B, config, args)

src/loss.py:525

↓ 1 callersFunctionimsave

(images, size, path)

src/utils_fn.py:453

↓ 1 callersFunctioninverse_transform

(images)

src/utils_fn.py:456

↓ 1 callersFunctionload_test_img_edge

Load image data

src/utils_fn.py:518

↓ 1 callersFunctionload_test_mask

(args)

src/utils_fn.py:379

↓ 1 callersFunctionload_val_img_scale_edge

Load image data

src/utils_fn.py:578

↓ 1 callersFunctionload_validation_mask

(args)

src/utils_fn.py:317

↓ 1 callersFunctionmain

Training

src/train_inpaint_model.py:156

next →1–100 of 209, ranked by callers