tessellate until threshold p is happy... @TODO warped to compensate for non-linear stretching

Summarizes one or more benchmark runs and prepares a pandas table showing the run per sample run-time and speed-up numbers. :param basel

Summarizes one or more benchmark runs and prepares a pandas table showing the per operator run-time and speed-up numbers. :param baselin

* \brief support Get support size. * * \return support size. */

Convert Tensor to Image using the foreign interface.

Convert an image in host data to cvcuda.Image Args: host_data (numpy array): Tensor in host data Returns: cvcuda.Image: The

Convert a tensor in host or CUDA data with layout to cvcuda.Tensor Args: data (numpy array or CUDA array): Tensor in host data la

Un-flattens (i.e expands) the data present in benchmark_dict and also calculates additions numbers.

Attempts to visualize the output produced by the operator as an image by writing it down to the disk. May raise exceptions if visuali