Backward Propagation Pointer Bug

[karanchahal]

When we try to overload operations such that complex operations can be done in a single expression. Something like

Tensor<float> = a;
Tensor<float> = b;
Tensor<float> = c;
Tensor<float> = d;
Tensor<float> e = a*b + c*d;
e.backward();

We get several garbage value Tensor objects when we debug the backOp of e. This is a very puzzling bug.

Also we cannot perform operations where there is a temporary on the right hand side.

Tensor<float> e = a*b + c*d;

is an example of that where a*b is a temporary.

There are bugs in the operation overloading and that is a medium priority item. It would be very good to get fixed but we can proceed using the assemly coding style one op per line approach

[karanchahal]

Update on this is as follows:

1. Added another way to perform operations on tensors by just taking the pointer of two tensors and adding them. This approach removes a lot of bugs and headaches, The only problem is via a user perspective with the API being tensorOps::add(t1,t2) instead of t1+t2.

Maybe this will change in the future to the latter hopefully.

Changes will be pushed through the CUDA dot product feature