class ContrastiveLoss(torch.nn.Module):
    def __init__(self, margin=1.0):
        super(ContrastiveLoss, self).__init__()
        self.margin = margin

    def forward(self, x, y):
        #<your code>
        L_sum = 0
        for i in range(x.size()[0] - 1):
            for j in range(i+1, x.size()[0]):
                y_ij = 1
                if y[i] == y[j]:
                    y_ij = 0
                norm = torch.norm((x[i] - x[j]), p=2) ** 2

                if self.margin - norm > 0:
                    L_sum += (1 - y_ij) * norm + y_ij * (self.margin - norm)
                else:
                    L_sum += (1 - y_ij) * norm

        return 2 * L_sum / (x.shape[0] * (x.shape[0] - 1))