into_contiguous_packed

Function into_contiguous_packed 

Source
pub fn into_contiguous_packed<R: Runtime, I: Int>(
    client: &ComputeClient<R::Server>,
    input: &TensorHandleRef<'_, R>,
    shape: &[usize],
    packing: u32,
) -> TensorHandle<R, I>
Expand description

Make a jit tensor contiguous, using the pitched allocator if available. See create_tensor. Handles unpacking and repacking packed tensors (i.e. quantized values). shape refers to the actual (unpacked) shape of the tensor, while packing specifies the number of elements in each storage element.

ยงWarning

This assumes u32 or u8 packing.