Add parameter type to `to_tensor` #2015

Job	Run time
linux (linux.g5.4xlarge.nvidia.gpu, cuda, 11.8) / linux-job	48s
linux (linux.12xlarge, cpu) / linux-job	42s
macos (macos-12) / macos-job	1m 16s
macos (macos-m1-stable) / macos-job	58s
windows (windows.4xlarge, cpu) / windows-job	2m 48s
windows (windows.g5.4xlarge.nvidia.gpu, cuda, 11.8) / windows-job	37s
	7m 9s

Provide feedback