[RickarySanchez]

Any new rust based ML library is great in my book. I've taken a brief look at the candle tensor implementation and just took an even briefer look at the can-t tensor implementation. What was the reasoning behind a global equation to store all tensors grads etc ? My initial assumption would be that it might make more sense to store that data scoped within a `Graph` type or similar.

[TuckerBMorgan]

Basically the idea, and I need to write up something on this, is that as much as possible client usage of the library should feel like you are writing PyTorch. For better of worse, people are used to that lib, and its patterns. So if anyone wanted to bring code over, it is easier if can basically copy and paste it over.