import math
import torch
import torch.nn as nn


class SparseLinear(nn.Module):
    def __init__(self, in_features, out_features, bias=True):
        super(SparseLinear, self).__init__()
        self.in_features = in_features
        self.out_features = out_features
        self.weight = nn.Parameter(torch.Tensor(in_features, out_features))
        if bias:
            self.bias = nn.Parameter(torch.Tensor(out_features))
        else:
            self.register_parameter('bias', None)
        self.reset_parameters()

    def reset_parameters(self):
        nn.init.kaiming_uniform_(self.weight, a=math.sqrt(5))
        if self.bias is not None:
            fan_in, _ = nn.init._calculate_fan_in_and_fan_out(self.weight)
            bound = 1 / math.sqrt(fan_in)
            nn.init.uniform_(self.bias, -bound, bound)

    def forward(self, input_sparse_tensor):
        return torch.sparse.mm(input_sparse_tensor, self.weight) + self.bias

    def extra_repr(self):
        return 'in_features={}, out_features={}, bias={}'.format(
            self.in_features, self.out_features, self.bias is not None
        )


if __name__ == '__main__':
    i = torch.LongTensor([[0, 1, 1],
                          [2, 0, 2]])
    v = torch.FloatTensor([3, 4, 5])
    input_tensor = torch.sparse.FloatTensor(i, v, torch.Size([2, 3])).requires_grad_(True)

    layer = SparseLinear(3, 5)

    output_tensor = layer(input_tensor)
    print('Size: {}'.format(output_tensor.size()))

    target = output_tensor.sum()
    target.backward()

    print('Gradients: {}'.format(input_tensor.grad))