How to use the memcnn.models.resnet.BottleneckSub function in memcnn

def __init__(self, inplanes, planes, stride=1, downsample=None, noactivation=False):
        super(RevBottleneck, self).__init__()
        if downsample is None and stride == 1:
            gm = BottleneckSub(inplanes // 2, planes // 2, stride, noactivation)
            fm = BottleneckSub(inplanes // 2, planes // 2, stride, noactivation)
            self.revblock = ReversibleBlock(gm, fm)
        else:
            self.bottleneck_sub = BottleneckSub(inplanes, planes, stride, noactivation)
        self.downsample = downsample
        self.stride = stride

def __init__(self, inplanes, planes, stride=1, downsample=None, noactivation=False):
        super(RevBottleneck, self).__init__()
        if downsample is None and stride == 1:
            gm = BottleneckSub(inplanes // 2, planes // 2, stride, noactivation)
            fm = BottleneckSub(inplanes // 2, planes // 2, stride, noactivation)
            self.revblock = ReversibleBlock(gm, fm)
        else:
            self.bottleneck_sub = BottleneckSub(inplanes, planes, stride, noactivation)
        self.downsample = downsample
        self.stride = stride

def __init__(self, inplanes, planes, stride=1, downsample=None, noactivation=False):
        super(RevBottleneck, self).__init__()
        if downsample is None and stride == 1:
            gm = BottleneckSub(inplanes // 2, planes // 2, stride, noactivation)
            fm = BottleneckSub(inplanes // 2, planes // 2, stride, noactivation)
            self.revblock = ReversibleBlock(gm, fm)
        else:
            self.bottleneck_sub = BottleneckSub(inplanes, planes, stride, noactivation)
        self.downsample = downsample
        self.stride = stride

def __init__(self, inplanes, planes, stride=1, noactivation=False):
        super(BottleneckSub, self).__init__()
        self.noactivation = noactivation
        if not self.noactivation:
            self.bn1 = batch_norm(inplanes)
        self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
        self.bn2 = batch_norm(planes)
        self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride,
                               padding=1, bias=False)
        self.bn3 = batch_norm(planes)
        self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
        self.relu = nn.ReLU(inplace=True)