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

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

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

Sets the learning rate to the initial LR decayed by 0.2 every steep step

Computes the accuracy over the k top predictions for the specified values of k

Computes the accuracy over the k top predictions for the specified values of k

3x3 convolution with padding

Draw N samples from multinomial :param N: number of samples :return: samples

model_ema = m * model_ema + (1 - m) model

generate shuffle ids for ShuffleBN

get the train loader

one epoch training

one epoch training

one epoch training

one epoch training for instance discrimination

one epoch training for instance discrimination

evaluation

Args: index (int): Index Returns: tuple: (image, target, index) where target is class_index of the target cla

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

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

set the learning rate