Class TensorView

gpu.hpp:116–120

* @brief Represents a non-owning view into a tensor specifying an offset and a * subspan. This is useful for specifying a slice of a tensor on the GPU * without copying the data. * * @code * TensorView view = {tensor, 0, 256}; * @endcode */

 * @endcode
 */
struct TensorView {
  Tensor data; // non-owning view
  size_t offset = 0;
  size_t span = 0;
};

/**
 * @brief Represents an ordered collection of WGPUBuffers (wrapped as tensors,