linting and formatting code

.style.yapf

@@ -1,8 +1,6 @@
 [style]
 BASED_ON_STYLE = pep8
 BLANK_LINE_BEFORE_NESTED_CLASS_OR_DEF = true
-BLANK_LINE_BEFORE_MODULE_DOCSTRING = true
-BLANK_LINE_BEFORE_CLASS_DOCSTRING = true
 SPLIT_BEFORE_EXPRESSION_AFTER_OPENING_PAREN = true
 DEDENT_CLOSING_BRACKETS = true
 SPACES_BEFORE_COMMENT = 1

projects/OSNet_AIN/osnet_child.py

@@ -7,7 +7,6 @@ from torch.nn import functional as F
 # Basic layers
 ##########
 class ConvLayer(nn.Module):
-
     """Convolution layer (conv + bn + relu)."""
 
     def __init__(
@@ -43,7 +42,6 @@ class ConvLayer(nn.Module):
 
 
 class Conv1x1(nn.Module):
-
     """1x1 convolution + bn + relu."""
 
     def __init__(self, in_channels, out_channels, stride=1, groups=1):
@@ -67,7 +65,6 @@ class Conv1x1(nn.Module):
 
 
 class Conv1x1Linear(nn.Module):
-
     """1x1 convolution + bn (w/o non-linearity)."""
 
     def __init__(self, in_channels, out_channels, stride=1, bn=True):
@@ -87,7 +84,6 @@ class Conv1x1Linear(nn.Module):
 
 
 class Conv3x3(nn.Module):
-
     """3x3 convolution + bn + relu."""
 
     def __init__(self, in_channels, out_channels, stride=1, groups=1):
@@ -111,7 +107,6 @@ class Conv3x3(nn.Module):
 
 
 class LightConv3x3(nn.Module):
-
     """Lightweight 3x3 convolution.
 
     1x1 (linear) + dw 3x3 (nonlinear).
@@ -142,7 +137,6 @@ class LightConv3x3(nn.Module):
 
 
 class LightConvStream(nn.Module):
-
     """Lightweight convolution stream."""
 
     def __init__(self, in_channels, out_channels, depth):
@@ -164,7 +158,6 @@ class LightConvStream(nn.Module):
 # Building blocks for omni-scale feature learning
 ##########
 class ChannelGate(nn.Module):
-
     """A mini-network that generates channel-wise gates conditioned on input tensor."""
 
     def __init__(
@@ -226,7 +219,6 @@ class ChannelGate(nn.Module):
 
 
 class OSBlock(nn.Module):
-
     """Omni-scale feature learning block."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -260,7 +252,6 @@ class OSBlock(nn.Module):
 
 
 class OSBlockINv1(nn.Module):
-
     """Omni-scale feature learning block with instance normalization."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -296,7 +287,6 @@ class OSBlockINv1(nn.Module):
 
 
 class OSBlockINv2(nn.Module):
-
     """Omni-scale feature learning block with instance normalization."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -332,7 +322,6 @@ class OSBlockINv2(nn.Module):
 
 
 class OSBlockINv3(nn.Module):
-
     """Omni-scale feature learning block with instance normalization."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -373,7 +362,6 @@ class OSBlockINv3(nn.Module):
 # Network architecture
 ##########
 class OSNet(nn.Module):
-
     """Omni-Scale Network.
     
     Reference:

projects/OSNet_AIN/osnet_search.py

@@ -11,7 +11,6 @@ NORM_AFFINE = False # enable affine transformations for normalization layer
 # Basic layers
 ##########
 class ConvLayer(nn.Module):
-
     """Convolution layer (conv + bn + relu)."""
 
     def __init__(
@@ -47,7 +46,6 @@ class ConvLayer(nn.Module):
 
 
 class Conv1x1(nn.Module):
-
     """1x1 convolution + bn + relu."""
 
     def __init__(
@@ -76,7 +74,6 @@ class Conv1x1(nn.Module):
 
 
 class Conv1x1Linear(nn.Module):
-
     """1x1 convolution + bn (w/o non-linearity)."""
 
     def __init__(self, in_channels, out_channels, stride=1, bn=True):
@@ -96,7 +93,6 @@ class Conv1x1Linear(nn.Module):
 
 
 class Conv3x3(nn.Module):
-
     """3x3 convolution + bn + relu."""
 
     def __init__(self, in_channels, out_channels, stride=1, groups=1):
@@ -120,7 +116,6 @@ class Conv3x3(nn.Module):
 
 
 class LightConv3x3(nn.Module):
-
     """Lightweight 3x3 convolution.
 
     1x1 (linear) + dw 3x3 (nonlinear).
@@ -151,7 +146,6 @@ class LightConv3x3(nn.Module):
 
 
 class LightConvStream(nn.Module):
-
     """Lightweight convolution stream."""
 
     def __init__(self, in_channels, out_channels, depth):
@@ -173,7 +167,6 @@ class LightConvStream(nn.Module):
 # Building blocks for omni-scale feature learning
 ##########
 class ChannelGate(nn.Module):
-
     """A mini-network that generates channel-wise gates conditioned on input tensor."""
 
     def __init__(
@@ -235,7 +228,6 @@ class ChannelGate(nn.Module):
 
 
 class OSBlock(nn.Module):
-
     """Omni-scale feature learning block."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -269,7 +261,6 @@ class OSBlock(nn.Module):
 
 
 class OSBlockINv1(nn.Module):
-
     """Omni-scale feature learning block with instance normalization."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -305,7 +296,6 @@ class OSBlockINv1(nn.Module):
 
 
 class OSBlockINv2(nn.Module):
-
     """Omni-scale feature learning block with instance normalization."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -341,7 +331,6 @@ class OSBlockINv2(nn.Module):
 
 
 class OSBlockINv3(nn.Module):
-
     """Omni-scale feature learning block with instance normalization."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -379,7 +368,6 @@ class OSBlockINv3(nn.Module):
 
 
 class NASBlock(nn.Module):
-
     """Neural architecture search layer."""
 
     def __init__(self, in_channels, out_channels, search_space=None):
@@ -425,7 +413,6 @@ class NASBlock(nn.Module):
 # Network architecture
 ##########
 class OSNet(nn.Module):
-
     """Omni-Scale Network.
     
     Reference:

torchreid/data/datamanager.py

@@ -7,7 +7,6 @@ from torchreid.data.transforms import build_transforms
 
 
 class DataManager(object):
-
     r"""Base data manager.
 
     Args:
@@ -87,7 +86,6 @@ class DataManager(object):
 
 
 class ImageDataManager(DataManager):
-
     r"""Image data manager.
 
     Args:
@@ -330,7 +328,6 @@ class ImageDataManager(DataManager):
 
 
 class VideoDataManager(DataManager):
-
     r"""Video data manager.
 
     Args:

torchreid/data/datasets/dataset.py

@@ -10,7 +10,6 @@ from torchreid.utils import read_image, download_url, mkdir_if_missing
 
 
 class Dataset(object):
-
     """An abstract class representing a Dataset.
 
     This is the base class for ``ImageDataset`` and ``VideoDataset``.
@@ -248,7 +247,6 @@ class Dataset(object):
 
 
 class ImageDataset(Dataset):
-
     """A base class representing ImageDataset.
 
     All other image datasets should subclass it.
@@ -297,7 +295,6 @@ class ImageDataset(Dataset):
 
 
 class VideoDataset(Dataset):
-
     """A base class representing VideoDataset.
 
     All other video datasets should subclass it.

torchreid/data/datasets/image/cuhk01.py

@@ -10,7 +10,6 @@ from ..dataset import ImageDataset
 
 
 class CUHK01(ImageDataset):
-
     """CUHK01.
 
     Reference:

torchreid/data/datasets/image/cuhk02.py

@@ -6,7 +6,6 @@ from ..dataset import ImageDataset
 
 
 class CUHK02(ImageDataset):
-
     """CUHK02.
 
     Reference:

torchreid/data/datasets/image/cuhk03.py

@@ -7,7 +7,6 @@ from ..dataset import ImageDataset
 
 
 class CUHK03(ImageDataset):
-
     """CUHK03.
 
     Reference:

torchreid/data/datasets/image/dukemtmcreid.py

@@ -7,7 +7,6 @@ from ..dataset import ImageDataset
 
 
 class DukeMTMCreID(ImageDataset):
-
     """DukeMTMC-reID.
 
     Reference:

torchreid/data/datasets/image/grid.py

@@ -9,7 +9,6 @@ from ..dataset import ImageDataset
 
 
 class GRID(ImageDataset):
-
     """GRID.
 
     Reference:

torchreid/data/datasets/image/ilids.py

@@ -11,7 +11,6 @@ from ..dataset import ImageDataset
 
 
 class iLIDS(ImageDataset):
-
     """QMUL-iLIDS.
 
     Reference:

torchreid/data/datasets/image/market1501.py

@@ -8,7 +8,6 @@ from ..dataset import ImageDataset
 
 
 class Market1501(ImageDataset):
-
     """Market1501.
 
     Reference:

torchreid/data/datasets/image/msmt17.py

@@ -23,7 +23,6 @@ VERSION_DICT = {
 
 
 class MSMT17(ImageDataset):
-
     """MSMT17.
 
     Reference:

torchreid/data/datasets/image/prid.py

@@ -8,7 +8,6 @@ from ..dataset import ImageDataset
 
 
 class PRID(ImageDataset):
-
     """PRID (single-shot version of prid-2011)
 
     Reference:

torchreid/data/datasets/image/sensereid.py

@@ -7,7 +7,6 @@ from ..dataset import ImageDataset
 
 
 class SenseReID(ImageDataset):
-
     """SenseReID.
 
     This dataset is used for test purpose only.

torchreid/data/datasets/image/viper.py

@@ -9,7 +9,6 @@ from ..dataset import ImageDataset
 
 
 class VIPeR(ImageDataset):
-
     """VIPeR.
 
     Reference:

torchreid/data/datasets/video/dukemtmcvidreid.py

@@ -9,7 +9,6 @@ from ..dataset import VideoDataset
 
 
 class DukeMTMCVidReID(VideoDataset):
-
     """DukeMTMCVidReID.
 
     Reference:

torchreid/data/datasets/video/ilidsvid.py

@@ -9,7 +9,6 @@ from ..dataset import VideoDataset
 
 
 class iLIDSVID(VideoDataset):
-
     """iLIDS-VID.
 
     Reference:

torchreid/data/datasets/video/mars.py

@@ -7,7 +7,6 @@ from ..dataset import VideoDataset
 
 
 class Mars(VideoDataset):
-
     """MARS.
 
     Reference:

torchreid/data/datasets/video/prid2011.py

@@ -8,7 +8,6 @@ from ..dataset import VideoDataset
 
 
 class PRID2011(VideoDataset):
-
     """PRID2011.
 
     Reference:

torchreid/data/sampler.py

@@ -9,7 +9,6 @@ AVAI_SAMPLERS = ['RandomIdentitySampler', 'SequentialSampler', 'RandomSampler']
 
 
 class RandomIdentitySampler(Sampler):
-
     """Randomly samples N identities each with K instances.
 
     Args:

torchreid/data/transforms.py

@@ -8,7 +8,6 @@ from torchvision.transforms import *
 
 
 class Random2DTranslation(object):
-
     """Randomly translates the input image with a probability.
 
     Specifically, given a predefined shape (height, width), the input is first
@@ -48,7 +47,6 @@ class Random2DTranslation(object):
 
 
 class RandomErasing(object):
-
     """Randomly erases an image patch.
 
     Origin: `<https://github.com/zhunzhong07/Random-Erasing>`_
@@ -107,7 +105,6 @@ class RandomErasing(object):
 
 
 class ColorAugmentation(object):
-
     """Randomly alters the intensities of RGB channels.
 
     Reference:
@@ -143,7 +140,6 @@ class ColorAugmentation(object):
 
 
 class RandomPatch(object):
-
     """Random patch data augmentation.
 
     There is a patch pool that stores randomly extracted pathces from person images.

torchreid/engine/engine.py

@@ -15,7 +15,6 @@ from torchreid.losses import DeepSupervision
 
 
 class Engine(object):
-
     r"""A generic base Engine class for both image- and video-reid.
 
     Args:

torchreid/engine/image/softmax.py

@@ -12,7 +12,6 @@ from ..engine import Engine
 
 
 class ImageSoftmaxEngine(Engine):
-
     r"""Softmax-loss engine for image-reid.
 
     Args:

torchreid/engine/image/triplet.py

@@ -12,7 +12,6 @@ from ..engine import Engine
 
 
 class ImageTripletEngine(Engine):
-
     r"""Triplet-loss engine for image-reid.
 
     Args:

torchreid/engine/video/softmax.py

@@ -5,7 +5,6 @@ from torchreid.engine.image import ImageSoftmaxEngine
 
 
 class VideoSoftmaxEngine(ImageSoftmaxEngine):
-
     """Softmax-loss engine for video-reid.
 
     Args:

torchreid/engine/video/triplet.py

@@ -5,7 +5,6 @@ from torchreid.engine.video import VideoSoftmaxEngine
 
 
 class VideoTripletEngine(ImageTripletEngine, VideoSoftmaxEngine):
-
     """Triplet-loss engine for video-reid.
 
     Args:

torchreid/losses/cross_entropy_loss.py

@@ -4,7 +4,6 @@ import torch.nn as nn
 
 
 class CrossEntropyLoss(nn.Module):
-
     r"""Cross entropy loss with label smoothing regularizer.
     
     Reference:

torchreid/losses/hard_mine_triplet_loss.py

@@ -4,7 +4,6 @@ import torch.nn as nn
 
 
 class TripletLoss(nn.Module):
-
     """Triplet loss with hard positive/negative mining.
     
     Reference:

torchreid/models/densenet.py

@@ -100,7 +100,6 @@ class _Transition(nn.Sequential):
 
 
 class DenseNet(nn.Module):
-
     """Densely connected network.
     
     Reference:

torchreid/models/hacnn.py

@@ -7,7 +7,6 @@ __all__ = ['HACNN']
 
 
 class ConvBlock(nn.Module):
-
     """Basic convolutional block.
     
     convolution + batch normalization + relu.
@@ -90,7 +89,6 @@ class InceptionB(nn.Module):
 
 
 class SpatialAttn(nn.Module):
-
     """Spatial Attention (Sec. 3.1.I.1)"""
 
     def __init__(self):
@@ -115,7 +113,6 @@ class SpatialAttn(nn.Module):
 
 
 class ChannelAttn(nn.Module):
-
     """Channel Attention (Sec. 3.1.I.2)"""
 
     def __init__(self, in_channels, reduction_rate=16):
@@ -134,7 +131,6 @@ class ChannelAttn(nn.Module):
 
 
 class SoftAttn(nn.Module):
-
     """Soft Attention (Sec. 3.1.I)
     
     Aim: Spatial Attention + Channel Attention
@@ -157,7 +153,6 @@ class SoftAttn(nn.Module):
 
 
 class HardAttn(nn.Module):
-
     """Hard Attention (Sec. 3.1.II)"""
 
     def __init__(self, in_channels):
@@ -183,7 +178,6 @@ class HardAttn(nn.Module):
 
 
 class HarmAttn(nn.Module):
-
     """Harmonious Attention (Sec. 3.1)"""
 
     def __init__(self, in_channels):
@@ -198,7 +192,6 @@ class HarmAttn(nn.Module):
 
 
 class HACNN(nn.Module):
-
     """Harmonious Attention Convolutional Neural Network.
 
     Reference:

torchreid/models/inceptionresnetv2.py

@@ -253,7 +253,6 @@ class Block8(nn.Module):
 
 
 class InceptionResNetV2(nn.Module):
-
     """Inception-ResNet-V2.
 
     Reference:

torchreid/models/inceptionv4.py

@@ -298,7 +298,6 @@ class Inception_C(nn.Module):
 
 
 class InceptionV4(nn.Module):
-
     """Inception-v4.
 
     Reference:

torchreid/models/mlfn.py

@@ -96,7 +96,6 @@ class MLFNBlock(nn.Module):
 
 
 class MLFN(nn.Module):
-
     """Multi-Level Factorisation Net.
 
     Reference:

torchreid/models/mobilenetv2.py

@@ -16,7 +16,6 @@ model_urls = {
 
 
 class ConvBlock(nn.Module):
-
     """Basic convolutional block.
     
     convolution (bias discarded) + batch normalization + relu6.
@@ -68,7 +67,6 @@ class Bottleneck(nn.Module):
 
 
 class MobileNetV2(nn.Module):
-
     """MobileNetV2.
 
     Reference:

torchreid/models/mudeep.py

@@ -7,7 +7,6 @@ __all__ = ['MuDeep']
 
 
 class ConvBlock(nn.Module):
-
     """Basic convolutional block.
     
     convolution + batch normalization + relu.
@@ -30,7 +29,6 @@ class ConvBlock(nn.Module):
 
 
 class ConvLayers(nn.Module):
-
     """Preprocessing layers."""
 
     def __init__(self):
@@ -47,7 +45,6 @@ class ConvLayers(nn.Module):
 
 
 class MultiScaleA(nn.Module):
-
     """Multi-scale stream layer A (Sec.3.1)"""
 
     def __init__(self):
@@ -77,7 +74,6 @@ class MultiScaleA(nn.Module):
 
 
 class Reduction(nn.Module):
-
     """Reduction layer (Sec.3.1)"""
 
     def __init__(self):
@@ -99,7 +95,6 @@ class Reduction(nn.Module):
 
 
 class MultiScaleB(nn.Module):
-
     """Multi-scale stream layer B (Sec.3.1)"""
 
     def __init__(self):
@@ -131,7 +126,6 @@ class MultiScaleB(nn.Module):
 
 
 class Fusion(nn.Module):
-
     """Saliency-based learning fusion layer (Sec.3.2)"""
 
     def __init__(self):
@@ -155,7 +149,6 @@ class Fusion(nn.Module):
 
 
 class MuDeep(nn.Module):
-
     """Multiscale deep neural network.
 
     Reference:

torchreid/models/nasnet.py

@@ -889,7 +889,6 @@ class ReductionCell1(nn.Module):
 
 
 class NASNetAMobile(nn.Module):
-
     """Neural Architecture Search (NAS).
 
     Reference:

torchreid/models/osnet.py

@@ -25,7 +25,6 @@ pretrained_urls = {
 # Basic layers
 ##########
 class ConvLayer(nn.Module):
-
     """Convolution layer (conv + bn + relu)."""
 
     def __init__(
@@ -62,7 +61,6 @@ class ConvLayer(nn.Module):
 
 
 class Conv1x1(nn.Module):
-
     """1x1 convolution + bn + relu."""
 
     def __init__(self, in_channels, out_channels, stride=1, groups=1):
@@ -87,7 +85,6 @@ class Conv1x1(nn.Module):
 
 
 class Conv1x1Linear(nn.Module):
-
     """1x1 convolution + bn (w/o non-linearity)."""
 
     def __init__(self, in_channels, out_channels, stride=1):
@@ -104,7 +101,6 @@ class Conv1x1Linear(nn.Module):
 
 
 class Conv3x3(nn.Module):
-
     """3x3 convolution + bn + relu."""
 
     def __init__(self, in_channels, out_channels, stride=1, groups=1):
@@ -129,7 +125,6 @@ class Conv3x3(nn.Module):
 
 
 class LightConv3x3(nn.Module):
-
     """Lightweight 3x3 convolution.
 
     1x1 (linear) + dw 3x3 (nonlinear).
@@ -164,7 +159,6 @@ class LightConv3x3(nn.Module):
 # Building blocks for omni-scale feature learning
 ##########
 class ChannelGate(nn.Module):
-
     """A mini-network that generates channel-wise gates conditioned on input tensor."""
 
     def __init__(
@@ -226,7 +220,6 @@ class ChannelGate(nn.Module):
 
 
 class OSBlock(nn.Module):
-
     """Omni-scale feature learning block."""
 
     def __init__(
@@ -286,7 +279,6 @@ class OSBlock(nn.Module):
 # Network architecture
 ##########
 class OSNet(nn.Module):
-
     """Omni-Scale Network.
     
     Reference:

torchreid/models/osnet_ain.py

@@ -15,7 +15,6 @@ pretrained_urls = {
 # Basic layers
 ##########
 class ConvLayer(nn.Module):
-
     """Convolution layer (conv + bn + relu)."""
 
     def __init__(
@@ -51,7 +50,6 @@ class ConvLayer(nn.Module):
 
 
 class Conv1x1(nn.Module):
-
     """1x1 convolution + bn + relu."""
 
     def __init__(self, in_channels, out_channels, stride=1, groups=1):
@@ -75,7 +73,6 @@ class Conv1x1(nn.Module):
 
 
 class Conv1x1Linear(nn.Module):
-
     """1x1 convolution + bn (w/o non-linearity)."""
 
     def __init__(self, in_channels, out_channels, stride=1, bn=True):
@@ -95,7 +92,6 @@ class Conv1x1Linear(nn.Module):
 
 
 class Conv3x3(nn.Module):
-
     """3x3 convolution + bn + relu."""
 
     def __init__(self, in_channels, out_channels, stride=1, groups=1):
@@ -119,7 +115,6 @@ class Conv3x3(nn.Module):
 
 
 class LightConv3x3(nn.Module):
-
     """Lightweight 3x3 convolution.
 
     1x1 (linear) + dw 3x3 (nonlinear).
@@ -150,7 +145,6 @@ class LightConv3x3(nn.Module):
 
 
 class LightConvStream(nn.Module):
-
     """Lightweight convolution stream."""
 
     def __init__(self, in_channels, out_channels, depth):
@@ -172,7 +166,6 @@ class LightConvStream(nn.Module):
 # Building blocks for omni-scale feature learning
 ##########
 class ChannelGate(nn.Module):
-
     """A mini-network that generates channel-wise gates conditioned on input tensor."""
 
     def __init__(
@@ -234,7 +227,6 @@ class ChannelGate(nn.Module):
 
 
 class OSBlock(nn.Module):
-
     """Omni-scale feature learning block."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -268,7 +260,6 @@ class OSBlock(nn.Module):
 
 
 class OSBlockINin(nn.Module):
-
     """Omni-scale feature learning block with instance normalization."""
 
     def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs):
@@ -307,7 +298,6 @@ class OSBlockINin(nn.Module):
 # Network architecture
 ##########
 class OSNet(nn.Module):
-
     """Omni-Scale Network.
     
     Reference:

torchreid/models/pcb.py

@@ -129,7 +129,6 @@ class DimReduceLayer(nn.Module):
 
 
 class PCB(nn.Module):
-
     """Part-based Convolutional Baseline.
 
     Reference:

torchreid/models/resnet.py

@@ -155,7 +155,6 @@ class Bottleneck(nn.Module):
 
 
 class ResNet(nn.Module):
-
     """Residual network.
     
     Reference:

torchreid/models/resnet_ibn_a.py

@@ -129,7 +129,6 @@ class Bottleneck(nn.Module):
 
 
 class ResNet(nn.Module):
-
     """Residual network + IBN layer.
     
     Reference:

torchreid/models/resnet_ibn_b.py

@@ -112,7 +112,6 @@ class Bottleneck(nn.Module):
 
 
 class ResNet(nn.Module):
-
     """Residual network + IBN layer.
     
     Reference:

torchreid/models/resnetmid.py

@@ -106,7 +106,6 @@ class Bottleneck(nn.Module):
 
 
 class ResNetMid(nn.Module):
-
     """Residual network + mid-level features.
     
     Reference:

torchreid/models/senet.py

@@ -113,7 +113,6 @@ class SEModule(nn.Module):
 
 
 class Bottleneck(nn.Module):
-
     """
     Base class for bottlenecks that implements `forward()` method.
     """
@@ -142,7 +141,6 @@ class Bottleneck(nn.Module):
 
 
 class SEBottleneck(Bottleneck):
-
     """
     Bottleneck for SENet154.
     """
@@ -175,7 +173,6 @@ class SEBottleneck(Bottleneck):
 
 
 class SEResNetBottleneck(Bottleneck):
-
     """
     ResNet bottleneck with a Squeeze-and-Excitation module. It follows Caffe
     implementation and uses `stride=stride` in `conv1` and not in `conv2`
@@ -209,7 +206,6 @@ class SEResNetBottleneck(Bottleneck):
 
 
 class SEResNeXtBottleneck(Bottleneck):
-
     """ResNeXt bottleneck type C with a Squeeze-and-Excitation module"""
     expansion = 4
 
@@ -248,7 +244,6 @@ class SEResNeXtBottleneck(Bottleneck):
 
 
 class SENet(nn.Module):
-
     """Squeeze-and-excitation network.
     
     Reference:

torchreid/models/shufflenet.py

@@ -97,7 +97,6 @@ cfg = {
 
 
 class ShuffleNet(nn.Module):
-
     """ShuffleNet.
 
     Reference:

torchreid/models/shufflenetv2.py

@@ -116,7 +116,6 @@ class InvertedResidual(nn.Module):
 
 
 class ShuffleNetV2(nn.Module):
-
     """ShuffleNetV2.
     
     Reference:

torchreid/models/squeezenet.py

@@ -46,7 +46,6 @@ class Fire(nn.Module):
 
 
 class SqueezeNet(nn.Module):
-
     """SqueezeNet.
 
     Reference:

torchreid/models/xception.py

@@ -143,7 +143,6 @@ class Block(nn.Module):
 
 
 class Xception(nn.Module):
-
     """Xception.
     
     Reference:

torchreid/utils/avgmeter.py

@@ -4,7 +4,6 @@ __all__ = ['AverageMeter']
 
 
 class AverageMeter(object):
-
     """Computes and stores the average and current value.
 
     Examples::

torchreid/utils/loggers.py

@@ -9,7 +9,6 @@ __all__ = ['Logger', 'RankLogger']
 
 
 class Logger(object):
-
     """Writes console output to external text file.
 
     Imported from `<https://github.com/Cysu/open-reid/blob/master/reid/utils/logging.py>`_
@@ -61,7 +60,6 @@ class Logger(object):
 
 
 class RankLogger(object):
-
     """Records the rank1 matching accuracy obtained for each
     test dataset at specified evaluation steps and provides a function
     to show the summarized results, which are convenient for analysis.

torchreid/utils/rerank.py

@@ -1,6 +1,5 @@
 #!/usr/bin/env python2/python3
 # -*- coding: utf-8 -*-
-
 """
 Source: https://github.com/zhunzhong07/person-re-ranking