MCPcopy Create free account

hub / github.com/bruno-31/GAN-manifold-regularization / functions

Functions58 in github.com/bruno-31/GAN-manifold-regularization

⨍ Functions 58◇ Types & classes 0
↓ 12 callersFunctiondiscriminator
(inp, is_training, init=False, reuse=False, getter =None)
svhn_gan.py:23
↓ 12 callersFunctiondiscriminator
(inp, is_training, init=False, reuse=False, getter =None)
cifar_gan.py:23
↓ 11 callersFunctionint_shape
(x)
nn.py:12
↓ 6 callersFunctiongenerator
(z_seed, is_training, init=False,reuse=False)
svhn_gan.py:56
↓ 6 callersFunctiongenerator
(z_seed, is_training, init=False,reuse=False)
cifar_gan.py:56
↓ 4 callersFunctionget_name
utlity for keeping track of layer names
nn.py:138
↓ 3 callersFunctionget_vars_maybe_avg
utility for retrieving polyak averaged params
nn.py:105
↓ 2 callersFunction_leakyReLu_impl
(x, alpha)
svhn_gan.py:15
↓ 2 callersFunction_leakyReLu_impl
(x, alpha)
cifar_gan.py:15
↓ 2 callersFunctionconv2d
convolutional layer
nn.py:182
↓ 2 callersFunctiondeconv2d
transposed convolutional layer
nn.py:219
↓ 2 callersFunctionget_var_maybe_avg
utility for retrieving polyak averaged params
nn.py:98
↓ 2 callersFunctionlog_sum_exp
numerically stable log_sum_exp implementation that prevents overflow
nn.py:20
↓ 2 callersFunctionrescale
(mat)
train_svhn.py:73
↓ 1 callersFunction_unpickle_file
(filename)
data/cifar10_input.py:14
↓ 1 callersFunctiondense
fully connected layer
nn.py:147
↓ 1 callersFunctiondisplay_progression_epoch
(j, id_max)
train_cifar_validation.py:55
↓ 1 callersFunctiondisplay_progression_epoch
(j, id_max)
train_svhn_validation.py:53
↓ 1 callersFunctiondisplay_progression_epoch
(j, id_max)
train_svhn.py:53
↓ 1 callersFunctiondisplay_progression_epoch
(j, id_max)
train_cifar.py:53
↓ 1 callersFunctionget_getter
(ema)
train_cifar_validation.py:47
↓ 1 callersFunctionget_getter
(ema)
train_svhn_validation.py:45
↓ 1 callersFunctionget_getter
(ema)
train_svhn.py:45
↓ 1 callersFunctionget_getter
(ema)
train_cifar.py:45
↓ 1 callersFunctionlinear_decay
(decay_start, decay_end, epoch)
train_cifar_validation.py:61
↓ 1 callersFunctionlinear_decay
(decay_start, decay_end, epoch)
train_svhn_validation.py:59
↓ 1 callersFunctionlinear_decay
(decay_start, decay_end, epoch)
train_svhn.py:59
↓ 1 callersFunctionlinear_decay
(decay_start, decay_end, epoch)
train_cifar.py:59
↓ 1 callersFunctionlog_prob_from_logits
numerically stable log_softmax implementation that prevents overflow
nn.py:27
↓ 1 callersFunctionmaybe_download
(data_dir)
data/svhn_data.py:6
↓ 1 callersFunctionnin
a network in network layer (1x1 CONV)
nn.py:262
↓ 1 callersFunctionrescale
(mat)
train_svhn_validation.py:72
Function_get_dataset
(path,split)
data/cifar10_input.py:30
Function_progress
(count, block_size, total_size)
data/svhn_data.py:10
Functionadam_updates
Adam optimizer
nn.py:112
Functionconcat_elu
like concatenated ReLU (http://arxiv.org/abs/1603.05201), but then with ELU
nn.py:15
Functiondiscretized_mix_logistic_loss
log-likelihood for mixture of discretized logistics, assumes the data has been rescaled to [-1,1] interval
nn.py:33
Functiondown_right_shifted_conv2d
(x, num_filters, filter_size=[2,2], stride=[1,1], **kwargs)
nn.py:319
Functiondown_right_shifted_deconv2d
(x, num_filters, filter_size=[2,2], stride=[1,1], **kwargs)
nn.py:324
Functiondown_shift
(x)
nn.py:299
Functiondown_shifted_conv2d
(x, num_filters, filter_size=[2,3], stride=[1,1], **kwargs)
nn.py:308
Functiondown_shifted_deconv2d
(x, num_filters, filter_size=[2,3], stride=[1,1], **kwargs)
nn.py:313
Functionema_getter
(getter, name, *args, **kwargs)
train_cifar_validation.py:48
Functionema_getter
(getter, name, *args, **kwargs)
train_svhn_validation.py:46
Functionema_getter
(getter, name, *args, **kwargs)
train_svhn.py:46
Functionema_getter
(getter, name, *args, **kwargs)
train_cifar.py:46
Functiongated_resnet
(x, a=None, h=None, nonlinearity=concat_elu, conv=conv2d, init=False, counters={}, ema=None, dropout_p=0., **k
nn.py:272
Functiongaussian_noise_layer
(input_layer, std)
svhn_gan.py:18
Functiongaussian_noise_layer
(input_layer, std)
cifar_gan.py:18
FunctionleakyReLu
(x, alpha=0.2, name=None)
svhn_gan.py:7
FunctionleakyReLu
(x, alpha=0.2, name=None)
cifar_gan.py:7
Functionload
(data_dir, subset='train')
data/svhn_data.py:16
Functionmain
(_)
train_cifar_validation.py:65
Functionmain
(_)
train_svhn_validation.py:63
Functionmain
(_)
train_svhn.py:63
Functionmain
(_)
train_cifar.py:63
Functionright_shift
(x)
nn.py:303
Functionsample_from_discretized_mix_logistic
(l,nr_mix)
nn.py:76