InputInfo

class nvtripy.InputInfo(shape: Sequence[int | DimensionSize | Tuple[int | DimensionSize, int | DimensionSize, int | DimensionSize]], dtype: dtype)

Bases: object

Captures information about an input to a compiled function.

Parameters:

• shape (Sequence[int | DimensionSize | Tuple[int | DimensionSize, int | DimensionSize, int | DimensionSize]]) – The shape of the input. To indicate dynamic dimensions, provide the minimum, optimum, and maximum values for the dimension.

• dtype (dtype) – The data type of the input.

Example

inp = tp.InputInfo((2, 4), dtype=tp.float32)

Local Variables

>>> inp
InputInfo(min=[2, 4], opt=[2, 4], max=[2, 4], dtype=float32)

Example: Dynamic Dimensions

# The first dimension will support values in the range [1, 3],
# optimizing for a size of 2.
inp = tp.InputInfo(((1, 2, 3), 4), dtype=tp.float32)

Local Variables

>>> inp
InputInfo(min=[1, 4], opt=[2, 4], max=[3, 4], dtype=float32)