Random connections in MADE perform better than deterministic masks

[jaanli]

I found that constructing masks using the original MADE paper https://arxiv.org/abs/1502.03509 (nodes are randomly assigned ids) works better than assigning an id deterministically as in

pytorch-flows/flows.py

Line 19 in 5520ebe

in_degrees = torch.arange(in_features) % in_flow_features

(The reason for this is likely simple, that if the output size is not an exact multiple of the number of hidden units, some input features will be assigned more hidden nodes than others)