1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
|
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT License.
from __future__ import annotations
import functools
import os
import pathlib
import unittest
import numpy as np
import onnx
import onnxruntime
import torch
from onnxscript import optimizer
from onnxscript._legacy_ir import visitor
from onnxscript.rewriter import onnxruntime as ort_rewriter
from onnxscript.utils import evaluation_utils
class TestBase(unittest.TestCase):
"""The base class for testing ONNX Script functions for internal use."""
def validate(self, fn):
"""Validate script function translation."""
return fn.to_function_proto()
def skip_if_no_cuda(reason: str):
def skip_dec(func):
@functools.wraps(func)
def wrapper(self, *args, **kwargs):
if not torch.cuda.is_available() or not onnxruntime.get_device() == "GPU":
raise unittest.SkipTest(f"GPU is not available. {reason}")
return func(self, *args, **kwargs)
return wrapper
return skip_dec
class OpTypeAnalysisVisitor(visitor.ProtoVisitorCore):
def __init__(self):
super().__init__()
self.op_types = set()
def visit_model(self, model: onnx.ModelProto):
self.op_types = set()
super().visit_model(model)
def process_node(self, node: onnx.NodeProto):
self.op_types.add((node.domain, node.op_type, getattr(node, "overload", "")))
return super().process_node(node)
def test_onnxruntime_rewrite(
model_basename: str,
model_count: int,
expected_optypes: set[tuple[str, str, str]],
rtol: float = 1e-2,
atol: float = 1e-2,
):
dir_path = pathlib.Path(os.path.dirname(os.path.realpath(__file__)))
unittest_root_dir = dir_path.parent.parent / "testdata" / "unittest_models"
for model_index in range(model_count):
model_name = f"{model_basename}_{model_index}"
model_dir = unittest_root_dir / f"{model_name}"
model_path = model_dir / f"{model_name}.onnx"
model = onnx.load(model_path)
# TODO: Parity issue with randomly generated data. Need investigation.
# inputs = generate_random_input(model)
inputs, expected_outputs = evaluation_utils.load_test_data(
model_dir, [i.name for i in model.graph.input]
)
optimized = optimizer.optimize(
model,
onnx_shape_inference=False,
num_iterations=2,
)
rewritten = ort_rewriter.rewrite(optimized)
# NOTE: uncomment this to save the optimized model.
# onnx.save(rewritten, model_dir / f"{model_name}_opt.onnx")
# Check expected operator is found.
optype_analysis = OpTypeAnalysisVisitor()
optype_analysis.visit_model(rewritten)
for domain, op_type, overload in expected_optypes:
if (domain, op_type, overload) not in optype_analysis.op_types:
raise AssertionError(
f"Expected op type {domain}:{op_type}:{overload} not found in rewritten model."
)
# Run baseline model
providers = ["CUDAExecutionProvider"]
# Run optimized model
optimized_session = onnxruntime.InferenceSession(
rewritten.SerializeToString(), providers=providers
)
optimized_outputs = optimized_session.run(None, inputs)
for i, (baseline_output, optimized_output) in enumerate(
zip(expected_outputs, optimized_outputs)
):
try:
np.testing.assert_equal(baseline_output.shape, optimized_output.shape)
np.testing.assert_allclose(
baseline_output, optimized_output, rtol=rtol, atol=atol
)
except AssertionError as e:
print(
f"Failed for model {model_name} and output {i} with rtol={rtol} and atol={atol}\n{e}"
)
raise
|
