torch-mlir/frontends/pytorch/test/ivalue_import/methods.py

44 lines
1.5 KiB
Python

# -*- Python -*-
# This file is licensed under a pytorch-style license
# See frontends/pytorch/LICENSE for license information.
import typing
import torch
import torch_mlir
# RUN: %PYTHON %s | npcomp-opt | FileCheck %s
mb = torch_mlir.ModuleBuilder()
# Function names in the Torch compilation unit are systematic -- they
# are effectively Python dotted paths. E.g. a Python module "foo" with a class
# "bar" with a method "baz" will result in a function in the compilation unit
# called "foo.bar.baz" when it gets `torch.jit.script`'ed.
# (with the exception that `__main__` is replaced with `__torch__`).
#
# Given how systematic this is, we don't treat the symbol names as opaque (i.e.
# we don't need to capture their names when FileCheck testing).
# CHECK-LABEL: func private @__torch__.TestModule.forward
# CHECK-SAME: (%[[SELF:.*]]: !torch.nn.Module<"__torch__.TestModule">, %[[X:.*]]: !torch.tensor) -> !torch.tensor {
# CHECK: return %[[X]] : !torch.tensor
# CHECK: }
#
# CHECK-LABEL: torch.class_type @__torch__.TestModule {
# CHECK: torch.method "forward", @__torch__.TestModule.forward
# CHECK: }
class TestModule(torch.nn.Module):
def __init__(self):
super().__init__()
def forward(self, x):
return x
test_module = TestModule()
recursivescriptmodule = torch.jit.script(test_module)
# TODO: Automatically handle unpacking Python class RecursiveScriptModule into the underlying ScriptModule.
mb.import_module(recursivescriptmodule._c)
mb.module.operation.print()