How to use the torchmeta.modules.MetaSequential function in torchmeta

def __init__(self, in_channels, out_features, hidden_size=64):
        super(ConvolutionalNeuralNetwork, self).__init__()
        self.in_channels = in_channels
        self.out_features = out_features
        self.hidden_size = hidden_size

        self.features = MetaSequential(
            conv3x3(in_channels, hidden_size),
            conv3x3(hidden_size, hidden_size),
            conv3x3(hidden_size, hidden_size),
            conv3x3(hidden_size, hidden_size)
        )

        self.classifier = MetaLinear(hidden_size, out_features)

def conv3x3(in_channels, out_channels, **kwargs):
    return MetaSequential(
        MetaConv2d(in_channels, out_channels, kernel_size=3, padding=1, **kwargs),
        MetaBatchNorm2d(out_channels, momentum=1., track_running_stats=False),
        nn.ReLU(),
        nn.MaxPool2d(2)
    )