opencv/modules/dnn/test/test_int.cpp
alexlyulkov 1d1faaabef
Merge pull request #24411 from alexlyulkov:al/dnn-type-inference
Added int32, int64 support and type inference to dnn #24411

**Added a type inference to dnn similar to the shape inference, added int32 and int64 support.**

- Added getTypes method for layers that calculates layer outputs types and internals types from inputs types (Similar to getMemoryShapes). By default outputs and internals types = input[0] type
- Added type inference pipeline similar to shape inference pipeline. LayersShapes struct (that is used in shape inference pipeline) now contains both shapes and types
- All layers output blobs are now allocated using the calculated types from the type inference.
- Inputs and constants with int32 and int64 types are not automatically converted into float32 now.
- Added int32 and int64 support for all the layers with indexing and for all the layers required in tests.

Added  int32 and int64 support for CUDA:
- Added host<->device data moving for int32 and int64
- Added int32 and int64 support for several layers (just slightly modified CUDA C++ templates)

Passed all the accuracy tests on CPU, OCL, OCL_FP16, CUDA, CUDA_FP16. (except RAFT model)

**CURRENT PROBLEMS**:
-  ONNX parser always converts int64 constants and layers attributes to int32, so some models with int64 constants doesn't work (e.g. RAFT). The solution is to disable int64->int32 conversion and fix attributes reading in a lot of ONNX layers parsers (https://github.com/opencv/opencv/issues/25102)
- I didn't add type inference and int support to VULCAN, so it doesn't work at all now.
- Some layers don't support int yet, so some unknown models may not work.

**CURRENT WORKAROUNDS**:
- CPU arg_layer indides are implemented in int32 followed by a int32->int64 conversion (the master branch has the same workaround with int32->float conversion)
- CPU and OCL pooling_layer indices are implemented in float followed by a float->int64 conversion
- CPU gather_layer indices are implemented in int32, so int64 indices are converted to int32 (the master branch has the same workaround with float->int32 conversion)

**DISABLED TESTS**:
- RAFT model

**REMOVED TESTS**:
- Greater_input_dtype_int64 (because it doesn't fit ONNX rules, the whole test is just comparing float tensor with int constant)

**TODO IN NEXT PULL REQUESTS**:
- Add int64 support for ONNX parser
- Add int support for more layers
- Add int support for OCL (currently int layers just run on CPU)
- Add int tests
- Add int support for other backends
2024-03-01 17:07:38 +03:00

57 lines
1.6 KiB
C++

// This file is part of OpenCV project.
// It is subject to the license terms in the LICENSE file found in the top-level directory
// of this distribution and at http://opencv.org/license.html.
//
// Copyright (C) 2017, Intel Corporation, all rights reserved.
// Third party copyrights are property of their respective owners.
#include "test_precomp.hpp"
namespace opencv_test { namespace {
typedef testing::TestWithParam<tuple<Backend, Target> > Test_int64_sum;
TEST_P(Test_int64_sum, basic)
{
Backend backend = get<0>(GetParam());
Target target = get<1>(GetParam());
int64_t a_value = 1000000000000000ll;
int64_t b_value = 1;
int64_t result_value = 1000000000000001ll;
EXPECT_NE(int64_t(float(a_value) + float(b_value)), result_value);
Mat a(3, 5, CV_64SC1, cv::Scalar_<int64_t>(a_value));
Mat b = Mat::ones(3, 5, CV_64S);
Net net;
LayerParams lp;
lp.type = "NaryEltwise";
lp.name = "testLayer";
lp.set("operation", "sum");
int id = net.addLayerToPrev(lp.name, lp.type, lp);
net.connect(0, 1, id, 1);
vector<String> inpNames(2);
inpNames[0] = "a";
inpNames[1] = "b";
net.setInputsNames(inpNames);
net.setInput(a, inpNames[0]);
net.setInput(b, inpNames[1]);
net.setPreferableBackend(backend);
net.setPreferableTarget(target);
Mat re;
re = net.forward();
EXPECT_EQ(re.depth(), CV_64S);
auto ptr_re = (int64_t *) re.data;
for (int i = 0; i < re.total(); i++)
ASSERT_EQ(result_value, ptr_re[i]);
}
INSTANTIATE_TEST_CASE_P(/*nothing*/, Test_int64_sum,
dnnBackendsAndTargets()
);
}} // namespace