copapy.grad

copapy.grad(x, y)

Returns the partial derivative dx/dy where x needs to be a scalar and y might be a scalar, a list of scalars, a vector, matrix or array. It uses automatic differentiation in reverse-mode.

Overloads:

• x (Any), y (value[Any]) → value[float] | float

• x (Any), y (vector[Any]) → vector[float]

• x (Any), y (tensor[Any]) → tensor[float]

• x (Any), y (array[Any]) → array[float]

• x (Any), y (quaternion) → quaternion

• x (Any), y (Sequence[value[Any]]) → list[value[float] | float]

Element-wise array operations, sum, min, max, dot and matrix products are differentiated by array operations. Convolutions, pooling, sorting and strided copies (transposing, slicing, broadcasting) are not supported.

Parameters:

• x (Any) – Value to return derivative of

• y (value[Any] | Sequence[value[Any]] | vector[Any] | tensor[Any] | array[Any] | quaternion) – Value(s) to derive in respect to

Returns:

Derivative of x with the type and dimensions of y