Functions — github.com/ZhugeKongan/Attention-mechanism-implementation

hub / github.com/ZhugeKongan/Attention-mechanism-implementation / functions

Functions81 in github.com/ZhugeKongan/Attention-mechanism-implementation

⨍ Functions 81 ◇ Types & classes 43

↓ 4 callersMethod_make_layer

(self, block, planes, num_blocks, stride)

train.py:92

↓ 4 callersMethodupdate

(self, val, n=1)

train.py:161

↓ 3 callersMethod__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_dot_product.py:94

↓ 3 callersMethod__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_embedded_gaussian.py:100

↓ 3 callersMethod__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_gaussian.py:96

↓ 3 callersMethod__init__

(self,channels,M=2)

models/SK_block.py:121

↓ 3 callersMethod__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_concatenation.py:110

↓ 2 callersMethod__init__

(self, block, num_blocks, num_classes=10)

train.py:79

↓ 2 callersMethod__init__

(self, channel)

models/ANN.py:28

↓ 2 callersMethod__init__

(self, planes)

models/CBAM_blocks.py:49

↓ 2 callersMethod__init__

(self, channel,ratio = 16)

models/SEvariants.py:7

↓ 2 callersMethod__init__

(self, in_channels, nclass, aux=True, norm_layer=nn.BatchNorm2d, norm_kwargs=None, **kwargs)

models/DaNet.py:51

↓ 2 callersFunctionaccuracy

Computes the precision@k for the specified values of k

train.py:166

↓ 1 callersFunctionConv1

(in_planes, places, stride=2)

train.py:38

↓ 1 callersFunctionResNet18

(**kwargs)

train.py:113

↓ 1 callersMethod__init__

(self)

models/TripletAttention.py:18

↓ 1 callersMethodreset

(self)

train.py:155

↓ 1 callersMethodspatial_pool

(self, x)

models/GlobalContextBlock.py:45

↓ 1 callersFunctiontrain

(trainloader, model, criterion, optimizer, epoch)

train.py:198

↓ 1 callersFunctionvalidate

(val_loader, model, criterion)

train.py:223

Method__init__

(self, in_planes, planes, stride=1)

train.py:49

Method__init__

(self, optimizer, total_iters, last_epoch=-1)

models/TripletAttention.py:34

Method__init__

(self, features, WH, M, G, r, stride=1, L=32)

models/SKNet.py:6

Method__init__

(self, in_channels, inter_channels=None, dimension=3, sub_sample=True, bn_layer=True)

models/non_local_dot_product.py:7

Method__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_dot_product.py:102

Method__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_dot_product.py:110

Method__init__

(self, output_sizes = [1, 3, 6, 8])

(self, in_chnls, ratio)

models/SE_block.py:13

Method__init__

(self, inplanes, ratio, pooling_type='att',

models/GlobalContextBlock.py:7

Method__init__

(self,block, in_channels, out_channels, size1, size2, size3)

models/residual_attention_blocks.py:7

Method__init__

:param in_channels: :param inter_channels: :param dimension: :param sub_sample: :param bn_layer:

models/non_local_embedded_gaussian.py:7

Method__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_embedded_gaussian.py:108

Method__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_embedded_gaussian.py:116

Method__init__

(self, in_channels, inter_channels=None, dimension=3, sub_sample=True, bn_layer=True)

models/non_local_gaussian.py:7

Method__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_gaussian.py:104

Method__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_gaussian.py:112

Method__init__

(self, in_planes, ratio=16)

models/CBAM_blocks.py:8

Method__init__

(self, kernel_size=7)

models/CBAM_blocks.py:31

Method__init__

(self, channel)

models/SEvariants.py:24

Method__init__

(self, channel,ratio = 16)

models/SEvariants.py:36

Method__init__

(self, in_channels, **kwargs)

(self,in_channels,out_channels,stride=1,M=2,r=16,L=32)

models/SK_block.py:4

Method__init__

(self, channel, k_size=3)

models/SK_block.py:103

Method__init__

(self, in_channels, inter_channels=None, dimension=3, sub_sample=True, bn_layer=True)

models/non_local_concatenation.py:7

Method__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True)

models/non_local_concatenation.py:118

Method__init__

(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True,)

models/non_local_concatenation.py:126

Method__init__

(self, channel)

models/NonLocalBlock.py:7

models/TripletAttention.py:13

Methodforward

(self, x)

models/TripletAttention.py:28

Methodforward

(self, x)

models/TripletAttention.py:42

:param x: (b, c, t, h, w) :param return_nl_map: if True return z, nl_map, else only return z. :return:

models/non_local_dot_product.py:63

models/SE_block.py:19

Methodforward

(self, x)

models/GlobalContextBlock.py:71

Methodforward

(self, x)

models/residual_attention_blocks.py:59

Methodforward

:param x: (b, c, t, h, w) :param return_nl_map: if True return z, nl_map, else only return z. :return:

models/non_local_embedded_gaussian.py:70

Methodforward

:param x: (b, c, t, h, w) :param return_nl_map: if True return z, nl_map, else only return z. :return:

models/non_local_gaussian.py:57

Methodforward

(self, x)

models/CBAM_blocks.py:23

Methodforward

(self, x)

models/CBAM_blocks.py:40

Methodforward

(self, x)

models/CBAM_blocks.py:54

Methodforward

(self, x)

models/SEvariants.py:16

Methodforward

(self, x)

models/SEvariants.py:30

Methodforward

(self, x)

models/SEvariants.py:41

models/SK_block.py:20

Methodforward

(self, x)

models/SK_block.py:109

Methodforward

(self, input)

models/SK_block.py:135

Methodforward

:param x: (b, c, t, h, w) :param return_nl_map: if True return z, nl_map, else only return z. :return:

models/non_local_concatenation.py:68

Methodforward

(self, x)

models/NonLocalBlock.py:16

Methodget_lr

we will use the first m batches, and set the learning rate to base_lr * m / total_iters

train.py:144