tests/testthat/_snaps/jit-ops.md

In torch: Tensors and Neural Networks with 'GPU' Acceleration

can print ops objects at different levels

Code
  jit_ops
Output
  Object of class <torch_ops>

Code
  jit_ops$sparse
Output
  <torch_ops>: Handle to namespace  sparse

Code
  jit_ops$prim$ChunkSizes
Output
  function_schema (name = prim::ChunkSizes)
  arguments: <none> 
  returns: <none>

Code
  jit_ops$aten$fft_fft
Output
  function_schema (name = aten::fft_fft)
  arguments: self - TensorType, n - OptionalType, dim - IntType, norm - OptionalType 
  returns: TensorType

torch documentation built on June 7, 2023, 6:19 p.m.