Functions — github.com/XLearning-SCU/2022-CVPR-DART

CodeHub MCPcopy Create free account

hub / github.com/XLearning-SCU/2022-CVPR-DART / functions

Functions92 in github.com/XLearning-SCU/2022-CVPR-DART

⨍ Functions 92 ◇ Types & classes 24

↓ 5 callersMethod__init__

(self, power=2)

↓ 5 callersFunctionremove_fc

Remove the fc layer parameters from state_dict.

↓ 5 callersMethodupdate

(self, val, n=1)

↓ 4 callersMethod_make_layer

(self, block, planes, blocks, stride=1, dilation=1)

↓ 4 callersFunctionextract_gall_feat

(gall_loader, net)

↓ 4 callersFunctionextract_query_feat

(query_loader, net)

↓ 4 callersFunctionprocess_test_regdb

(img_dir, trial = 1, modal = 'visible')

data_manager.py:69

↓ 4 callersFunctionsoftmax_weights

↓ 3 callersMethod__init__

↓ 3 callersFunctioneval_regdb

(distmat, q_pids, g_pids, max_rank = 20)

eval_metrics.py:86

↓ 3 callersFunctionprocess_gallery_sysu

(data_path, mode = 'all', trial = 0, relabel=False)

data_manager.py:37

↓ 3 callersFunctionresnet50

Constructs a ResNet-50 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet

↓ 2 callersFunctionGenIdx

( train_color_label, train_thermal_label)

↓ 2 callersMethod__init__

(self, block, layers, last_conv_stride=2, last_conv_dilation=1)

↓ 2 callersFunctionadjust_learning_rate

Sets the learning rate to the initial LR decayed by 10 every 30 epochs

↓ 2 callersMethodclose

↓ 2 callersFunctionconv3x3

3x3 convolution with padding

↓ 2 callersFunctioneval_sysu

Evaluation with sysu metric Key: for each query identity, its gallery images from the same camera view are discarded. "Following the original sett

eval_metrics.py:6

↓ 2 callersFunctioneval_train

(net, dataloader, type)

↓ 2 callersFunctionload_data

(input_data_path)

data_loader.py:383

↓ 2 callersFunctionnormalize

Normalizing to unit length along the specified dimension. Args: x: pytorch Variable Returns: x: pytorch Variable, same shape as in

↓ 2 callersFunctionpdist_torch

compute the eucilidean distance matrix between embeddings1 and embeddings2 using gpu

↓ 2 callersFunctionprocess_query_sysu

(data_path, mode = 'all', relabel=False)

data_manager.py:6

↓ 2 callersFunctiontrain

(epoch, net, optimizer, trainloader)

↓ 2 callersFunctionwarmup

(epoch, net, optimizer, dataloader)

↓ 1 callersFunctionmkdir_if_missing

↓ 1 callersMethodreset

↓ 1 callersFunctionset_seed

(seed, cuda=True)

↓ 1 callersFunctiontest

FunctionExtractCam

FunctionGenCamIdx

(gall_img, gall_label, mode)

ChannelAug.py:26

ChannelAug.py:65

ChannelAug.py:115

data_loader.py:23

Method__enter__

Method__getitem__

data_loader.py:169

Method__getitem__

data_loader.py:339

Method__getitem__

data_loader.py:374

(self, alpha=1, gamma=1, square=0)

(self, batch_size, margin)

(self, train_color_label, train_thermal_label, color_pos, thermal_pos, num_pos, batchSize, epoch)

(self, fpath=None)

(self, dataset, train_color_label, train_thermal_label, shuffle=True)

(self, probability = 0.5)

ChannelAug.py:22

(self, probability = 0.5)

ChannelAug.py:61

(self, probability = 0.5, sl = 0.02, sh = 0.4, r1 = 0.3, mean=[0.4914, 0.4822, 0.4465])

ChannelAug.py:107

data_loader.py:20

(self, data_dir, transform=None, colorIndex=None, thermalIndex=None, noise_rate=0., noise_file='', mode='', pr

data_loader.py:48

(self, data_dir, trial, transform=None, colorIndex=None, thermalIndex=None, noise_rate=0., no

data_loader.py:195

(self, test_img_file, test_label, transform=None, img_size=(144, 288))

data_loader.py:362

(self, in_channels, reduc_ratio=2)

(self, arch='resnet50')

(self, arch='resnet50')

(self, arch='resnet50')

(self, class_num, no_local= 'on', gm_pool = 'on', arch='resnet50')

(self, inplanes, planes, stride=1, downsample=None, dilation=1)

(self, inplanes, planes, stride=1, downsample=None, dilation=1)

data_loader.py:379

(self, inputs, label_assign, targets, true_targets, prob, threshold=0.6, alpha=100, normalize_feature=False)

(self, inputs, label_assign, targets, true_targets, prob, threshold=0.6, alpha=100, normalize_feature=False)

(self, pred, label)

(self, inputs, prediction, targets, true_targets, prob, threshold)

:param x: (b, c, t, h, w) :return:

(self, x1, x2, modal=0)

Functionload_data

(input_data_path )

Functionpdist_np

compute the eucilidean distance matrix between embeddings1 and embeddings2 using cpu

Functionplot_pair_distribution

(type, X, clean_index, noisy_index, save_path='')

Functionresnet101

Constructs a ResNet-101 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet

Functionresnet152

Constructs a ResNet-152 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet

Functionresnet18

Constructs a ResNet-18 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet

Functionresnet34

Constructs a ResNet-34 model. Args: pretrained (bool): If True, returns a model pre-trained on ImageNet

Functionset_requires_grad

Set requies_grad=Fasle for all the networks to avoid unnecessary computations Parameters: nets (network list) -- a list

Functionweights_init_classifier

Functionweights_init_kaiming