mirror of
https://github.com/saymrwulf/onnxruntime.git
synced 2026-07-04 04:07:22 +00:00
Added MatMul tests for QNN EP (#15956)
### Description <!-- Describe your changes. --> Added test coverage for QNN EP MatMul op ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. --> Created test coverage for HTP based MatMul with broadcasting. --------- Co-authored-by: Hector Li <hecli@microsoft.com>
This commit is contained in:
Aung T Naing
GitHub
parent
7b51a1b17d
commit
e83993bbaf
1 changed files with 180 additions and 0 deletions
180
onnxruntime/test/providers/qnn/matmul_test.cpp
Normal file
180
onnxruntime/test/providers/qnn/matmul_test.cpp
Normal file
|
|
@ -0,0 +1,180 @@
|
|||
// Copyright (c) Microsoft Corporation. All rights reserved.
|
||||
// Licensed under the MIT License.
|
||||
|
||||
#if !defined(ORT_MINIMAL_BUILD)
|
||||
|
||||
#include <string>
|
||||
#include <unordered_map>
|
||||
|
||||
#include "test/optimizer/qdq_test_utils.h"
|
||||
#include "test/providers/qnn/qnn_test_utils.h"
|
||||
|
||||
#include "onnx/onnx_pb.h"
|
||||
|
||||
#include "gtest/gtest.h"
|
||||
|
||||
namespace onnxruntime {
|
||||
namespace test {
|
||||
|
||||
// Returns a function that creates a graph with MatMul operator.
|
||||
static GetTestModelFn BuildMatMulOpTestCase(const std::vector<int64_t>& input1_shape,
|
||||
const std::vector<int64_t>& input2_shape) {
|
||||
return [input1_shape, input2_shape](ModelTestBuilder& builder) {
|
||||
// Random input data
|
||||
auto input1 = builder.MakeInput<float>(input1_shape, 0.0f, 10.0f);
|
||||
auto input2 = builder.MakeInput<float>(input2_shape, 0.0f, 10.0f);
|
||||
|
||||
auto* output = builder.MakeOutput();
|
||||
builder.AddNode("MatMul", {input1, input2}, {output});
|
||||
};
|
||||
}
|
||||
|
||||
// Returns a function that creates a graph with a QDQ AveragePool operator.
|
||||
template <typename QuantType>
|
||||
GetQDQTestCaseFn BuildMatMulOpQDQTestCase(const std::vector<int64_t>& input1_shape,
|
||||
const std::vector<int64_t>& input2_shape) {
|
||||
return [input1_shape, input2_shape](ModelTestBuilder& builder) {
|
||||
float pool_output_scale = 0.0038f;
|
||||
float q_scale = 0.0038f;
|
||||
QuantType pool_output_zp = std::numeric_limits<QuantType>::max() / 2;
|
||||
QuantType q_zp = std::numeric_limits<QuantType>::max() / 2;
|
||||
|
||||
auto* input_arg = builder.MakeInput<float>(input1_shape, -1.f, 1.f);
|
||||
auto* output_arg = builder.MakeOutput();
|
||||
|
||||
using InputLimits = std::numeric_limits<QuantType>;
|
||||
|
||||
// add QDQ input
|
||||
auto* q1_output = builder.MakeIntermediate();
|
||||
auto* dq1_output = builder.MakeIntermediate();
|
||||
builder.AddQuantizeLinearNode<QuantType>(input_arg,
|
||||
pool_output_scale,
|
||||
pool_output_zp,
|
||||
q1_output);
|
||||
builder.AddDequantizeLinearNode<QuantType>(q1_output,
|
||||
q_scale,
|
||||
q_zp,
|
||||
dq1_output);
|
||||
|
||||
// add input b initializer (NNAPI only supports case of MatMul A*B - B is an initializer)
|
||||
auto* dq_2_output = builder.MakeIntermediate();
|
||||
auto* input_b = builder.MakeInitializer<QuantType>(input2_shape, InputLimits::min(), InputLimits::max());
|
||||
builder.AddDequantizeLinearNode<QuantType>(input_b,
|
||||
q_scale,
|
||||
q_zp,
|
||||
dq_2_output);
|
||||
|
||||
// add MatMul operator
|
||||
auto* matmul_op_output = builder.MakeIntermediate();
|
||||
builder.AddNode("MatMul", {dq1_output, dq_2_output}, {matmul_op_output});
|
||||
|
||||
// add QDQ output
|
||||
auto* q3_output = builder.MakeIntermediate();
|
||||
builder.AddQuantizeLinearNode<QuantType>(matmul_op_output,
|
||||
pool_output_scale,
|
||||
pool_output_zp,
|
||||
q3_output);
|
||||
builder.AddDequantizeLinearNode<QuantType>(q3_output,
|
||||
q_scale,
|
||||
q_zp,
|
||||
output_arg);
|
||||
};
|
||||
}
|
||||
|
||||
// Runs an AveragePool model on the QNN CPU backend. Checks the graph node assignment, and that inference
|
||||
// outputs for QNN and CPU match.
|
||||
static void RunMatMulOpOpTest(const std::vector<int64_t>& input1_shape,
|
||||
const std::vector<int64_t>& input2_shape,
|
||||
ExpectedEPNodeAssignment expected_ep_assignment, const char* test_description,
|
||||
int opset = 13) {
|
||||
ProviderOptions provider_options;
|
||||
#if defined(_WIN32)
|
||||
provider_options["backend_path"] = "QnnCpu.dll";
|
||||
#else
|
||||
provider_options["backend_path"] = "libQnnCpu.so";
|
||||
#endif
|
||||
|
||||
constexpr int expected_nodes_in_partition = 1;
|
||||
RunQnnModelTest(BuildMatMulOpTestCase(input1_shape, input2_shape),
|
||||
provider_options,
|
||||
opset,
|
||||
expected_ep_assignment,
|
||||
expected_nodes_in_partition,
|
||||
test_description);
|
||||
}
|
||||
|
||||
// Runs a QDQ AveragePool model on the QNN HTP backend. Checks the graph node assignment, and that inference
|
||||
// outputs for QNN and CPU match.
|
||||
template <typename QuantType>
|
||||
static void RunQDQMatMulOpOpTest(const std::vector<int64_t>& input1_shape,
|
||||
const std::vector<int64_t>& input2_shape,
|
||||
ExpectedEPNodeAssignment expected_ep_assignment, const char* test_description,
|
||||
int opset = 18, float fp32_abs_err = 1e-5f) {
|
||||
ProviderOptions provider_options;
|
||||
#if defined(_WIN32)
|
||||
provider_options["backend_path"] = "QnnHtp.dll";
|
||||
#else
|
||||
provider_options["backend_path"] = "libQnnHtp.so";
|
||||
#endif
|
||||
|
||||
constexpr int expected_nodes_in_partition = 1;
|
||||
RunQnnModelTest(BuildMatMulOpQDQTestCase<QuantType>(input1_shape, input2_shape),
|
||||
provider_options,
|
||||
opset,
|
||||
expected_ep_assignment,
|
||||
expected_nodes_in_partition,
|
||||
test_description,
|
||||
fp32_abs_err);
|
||||
}
|
||||
|
||||
//
|
||||
// CPU tests:
|
||||
//
|
||||
|
||||
TEST_F(QnnCPUBackendTests, TestMatMulOp) {
|
||||
RunMatMulOpOpTest({2, 2} /* input_shape1 */,
|
||||
{2, 2} /* input_shape2 */,
|
||||
ExpectedEPNodeAssignment::All, "TestMatMulOp", 18);
|
||||
}
|
||||
|
||||
// QNN broadcast issue
|
||||
TEST_F(QnnCPUBackendTests, DISABLED_TestMatMulOp2) {
|
||||
RunMatMulOpOpTest({28, 1, 64} /* input_shape1 */,
|
||||
{64, 32} /* input_shape2 */,
|
||||
ExpectedEPNodeAssignment::All, "TestMatMulOp2", 18);
|
||||
}
|
||||
|
||||
#if defined(__aarch64__) || defined(_M_ARM64) || defined(__linux__)
|
||||
//
|
||||
// HTP tests:
|
||||
//
|
||||
|
||||
TEST_F(QnnHTPBackendTests, TestMatMulOp_HTP_u8) {
|
||||
RunQDQMatMulOpOpTest<uint8_t>({2, 2} /* input_shape1 */,
|
||||
{2, 2} /* input_shape2 */,
|
||||
ExpectedEPNodeAssignment::All, "TestMatMulOp_HTP_u8",
|
||||
18, 0.00381f);
|
||||
}
|
||||
|
||||
// QNN broadcast issue
|
||||
TEST_F(QnnHTPBackendTests, DISABLED_TestMatMulOp2_HTP_u8) {
|
||||
RunQDQMatMulOpOpTest<uint8_t>({28, 1, 64} /* input_shape1 */,
|
||||
{64, 32} /* input_shape2 */,
|
||||
ExpectedEPNodeAssignment::All, "TestMatMulOp2_HTP_u8",
|
||||
18, 0.00381f);
|
||||
}
|
||||
|
||||
// QNN broadcast issue
|
||||
TEST_F(QnnHTPBackendTests, DISABLED_TestMatMulOp3_HTP_u8) {
|
||||
RunQDQMatMulOpOpTest<uint8_t>({28, 1, 32} /* input_shape1 */,
|
||||
{32, 2} /* input_shape2 */,
|
||||
ExpectedEPNodeAssignment::All, "TestMatMulOp3_HTP_u8",
|
||||
18, 0.00381f);
|
||||
}
|
||||
|
||||
#endif // defined(__aarch64__) || defined(_M_ARM64) || defined(__linux__)
|
||||
|
||||
} // namespace test
|
||||
} // namespace onnxruntime
|
||||
|
||||
#endif // !defined(ORT_MINIMAL_BUILD)
|
||||
Loading…
Reference in a new issue