Broadcasting is a feature of Numpy arrays that allows arithmetic operations between arrays that are not necessarily of compatible dimension for the computation.

During such an operation, the arrays are "broadcast" to a certain size that makes the operartion possible.

Knowledge of broadcasting is useful in a number of places, for example deep learning methids.

Create a Numpy array A as a $(3,3)$ array of random integers.

# Your code here

Create a second numpy array, b of size $(3,)$ of random integers.

# Your code here

Perform the operation A$*$b

# Your code below

Explain why it was possible to compute A$*$b in the above case.

As part of your explanation, show the broadcast versions of A and b

Explanation here.

Now create a third array c that is a $(2,)$ array of random integers.

# Your code here

Perform the operation A$*$c

# Your code here

Explain, in terms of broadcasting, why it was not possible to compute A$*$c

Your explanation here.