Open armandok opened 3 years ago
vik748
commented
3 years ago This is not working code yet, I haven't had the time to get all the pieces working. Having said that it should still be able to compile. Follow the steps in .travis.yml to ensure you have everything setup correctly.
rener2020
commented
2 years ago many bugs here
if u ONLY want to solve this bug just rewrite some codes in src/lib/conversion.cpp
good luck
class NumpyAllocator : public MatAllocator { public: NumpyAllocator() { stdAllocator = Mat::getStdAllocator(); } ~NumpyAllocator() {}
UMatData* allocate(PyObject* o, int dims, const int* sizes, int type, size_t* step) const
{
UMatData* u = new UMatData(this);
u->data = u->origdata = (uchar*)PyArray_DATA((PyArrayObject*) o);
npy_intp* _strides = PyArray_STRIDES((PyArrayObject*) o);
for( int i = 0; i < dims - 1; i++ )
step[i] = (size_t)_strides[i];
step[dims-1] = CV_ELEM_SIZE(type);
u->size = sizes[0]*step[0];
u->userdata = o;
return u;
}
UMatData *allocate(int dims, const int *sizes, int type, void *data, size_t *step, AccessFlag flags, UMatUsageFlags usageFlags) const
{
USAGE_DEFAULT;
if( data != 0 )
{
CV_Error(Error::StsAssert, "The data should normally be NULL!");
// probably this is safe to do in such extreme case
return stdAllocator->allocate(dims, sizes, type, data, step, flags, usageFlags);
}
PyEnsureGIL gil;
int depth = CV_MAT_DEPTH(type);
int cn = CV_MAT_CN(type);
const int f = (int)(sizeof(size_t)/8);
int typenum = depth == CV_8U ? NPY_UBYTE : depth == CV_8S ? NPY_BYTE :
depth == CV_16U ? NPY_USHORT : depth == CV_16S ? NPY_SHORT :
depth == CV_32S ? NPY_INT : depth == CV_32F ? NPY_FLOAT :
depth == CV_64F ? NPY_DOUBLE : f*NPY_ULONGLONG + (f^1)*NPY_UINT;
int i = dims;
cv::AutoBuffer<npy_intp> _sizes(dims + 1);
for( i = 0; i < dims; i++ )
_sizes[i] = sizes[i];
if( cn > 1 )
_sizes[dims++] = cn;
PyObject* o = PyArray_SimpleNew(dims, _sizes, typenum);
if(!o)
CV_Error_(Error::StsError, ("The numpy array of typenum=%d, ndims=%d can not be created", typenum, dims));
return allocate(o, dims, sizes, type, step);
}
bool allocate(UMatData* data, AccessFlag accessflags, UMatUsageFlags usageFlags) const
{
return stdAllocator->allocate(data, accessflags, usageFlags);
}
void deallocate(UMatData* data) const
{
if(data)
{
PyEnsureGIL gil;
PyObject* o = (PyObject*)data->userdata;
Py_XDECREF(o);
delete data;
}
}
const MatAllocator* stdAllocator;};
iljastas
commented
2 years ago many bugs here
if u ONLY want to solve this bug just rewrite some codes in src/lib/conversion.cpp
good luck
if OPENCV_3 class NumpyAllocator : public MatAllocator { public: NumpyAllocator() { stdAllocator = Mat::getStdAllocator(); } ~NumpyAllocator() {}
UMatData* allocate(PyObject* o, int dims, const int* sizes, int type, size_t* step) const { UMatData* u = new UMatData(this); u->data = u->origdata = (uchar*)PyArray_DATA((PyArrayObject*) o); npy_intp* _strides = PyArray_STRIDES((PyArrayObject*) o); for( int i = 0; i < dims - 1; i++ ) step[i] = (size_t)_strides[i]; step[dims-1] = CV_ELEM_SIZE(type); u->size = sizes[0]*step[0]; u->userdata = o; return u; } UMatData *allocate(int dims, const int *sizes, int type, void *data, size_t *step, AccessFlag flags, UMatUsageFlags usageFlags) const { USAGE_DEFAULT; if( data != 0 ) { CV_Error(Error::StsAssert, "The data should normally be NULL!"); // probably this is safe to do in such extreme case return stdAllocator->allocate(dims, sizes, type, data, step, flags, usageFlags); } PyEnsureGIL gil; int depth = CV_MAT_DEPTH(type); int cn = CV_MAT_CN(type); const int f = (int)(sizeof(size_t)/8); int typenum = depth == CV_8U ? NPY_UBYTE : depth == CV_8S ? NPY_BYTE : depth == CV_16U ? NPY_USHORT : depth == CV_16S ? NPY_SHORT : depth == CV_32S ? NPY_INT : depth == CV_32F ? NPY_FLOAT : depth == CV_64F ? NPY_DOUBLE : f*NPY_ULONGLONG + (f^1)*NPY_UINT; int i = dims; cv::AutoBuffer<npy_intp> _sizes(dims + 1); for( i = 0; i < dims; i++ ) _sizes[i] = sizes[i]; if( cn > 1 ) _sizes[dims++] = cn; PyObject* o = PyArray_SimpleNew(dims, _sizes, typenum); if(!o) CV_Error_(Error::StsError, ("The numpy array of typenum=%d, ndims=%d can not be created", typenum, dims)); return allocate(o, dims, sizes, type, step); } bool allocate(UMatData* data, AccessFlag accessflags, UMatUsageFlags usageFlags) const { return stdAllocator->allocate(data, accessflags, usageFlags); } void deallocate(UMatData* data) const { if(data) { PyEnsureGIL gil; PyObject* o = (PyObject*)data->userdata; Py_XDECREF(o); delete data; } } const MatAllocator* stdAllocator;};
Hi rener2020,
where is the file src/lib/conversion.cpp? Is this a file in openCV?
rener2020
commented
2 years ago no in this project
rener2020
commented
2 years ago pyfbow/src/lib/conversion.cpp
When compiling the wrapper, I get this error:
Scanning dependencies of target np_opencv_converter [ 20%] Building CXX object CMakeFiles/np_opencv_converter.dir/lib/conversion.cpp.o In file included from /projectnb/semslam/armandok/.conda/envs/pyslamu/lib/python3.7/site-packages/numpy/core/include/numpy/ndarraytypes.h:1822:0, from /projectnb/semslam/armandok/.conda/envs/pyslamu/lib/python3.7/site-packages/numpy/core/include/numpy/ndarrayobject.h:12, from /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.h:10, from /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:4: /projectnb/semslam/armandok/.conda/envs/pyslamu/lib/python3.7/site-packages/numpy/core/include/numpy/npy_1_7_deprecated_api.h:17:2: warning: #warning "Using deprecated NumPy API, disable it with " "#define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION" [-Wcpp]
warning "Using deprecated NumPy API, disable it with " \
^
~~ /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp: In function 'int init()': /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:12:5: warning: converting to non-pointer type 'int' from NULL [-Wconversion-null] import_array(); ^~~~ /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp: In member function 'cv::UMatData NumpyAllocator::allocate(int, const int, int, void, size_t, int, cv::UMatUsageFlags) const': /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:99:92: error: invalid conversion from 'int' to 'cv::AccessFlag' [-fpermissive] return stdAllocator->allocate(dims0, sizes, type, data, step, flags, usageFlags); ^ In file included from /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/core.hpp:59:0, from /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/opencv.hpp:52, from /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.h:8, from /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:4: /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/core/mat.hpp:475:23: note: initializing argument 6 of 'virtual cv::UMatData cv::MatAllocator::allocate(int, const int, int, void, size_t, cv::AccessFlag, cv::UMatUsageFlags) const' virtual UMatData allocate(int dims, const int sizes, int type, ^~~~ /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp: In member function 'bool NumpyAllocator::allocate(cv::UMatData, int, cv::UMatUsageFlags) const': /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:124:59: error: invalid conversion from 'int' to 'cv::AccessFlag' [-fpermissive] return stdAllocator->allocate(u, accessFlags, usageFlags); ^ In file included from /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/core.hpp:59:0, from /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/opencv.hpp:52, from /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.h:8, from /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:4: /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/core/mat.hpp:477:18: note: initializing argument 2 of 'virtual bool cv::MatAllocator::allocate(cv::UMatData, cv::AccessFlag, cv::UMatUsageFlags) const' virtual bool allocate(UMatData data, AccessFlag accessflags, UMatUsageFlags usageFlags) const = 0; ^~~~ /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp: At global scope: /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:194:16: error: cannot declare variable 'g_numpyAllocator' to be of abstract type 'NumpyAllocator' NumpyAllocator g_numpyAllocator; ^~~~/usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:73:7: note: because the following virtual functions are pure within 'NumpyAllocator': class NumpyAllocator : public MatAllocator ^~~~~~ In file included from /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/core.hpp:59:0, from /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/opencv.hpp:52, from /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.h:8, from /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:4: /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/core/mat.hpp:475:23: note: virtual cv::UMatData cv::MatAllocator::allocate(int, const int, int, void, size_t, cv::AccessFlag, cv::UMatUsageFlags) const virtual UMatData allocate(int dims, const int sizes, int type, ^~~~ /share/pkg.7/opencv/4.5.0/install/include/opencv4/opencv2/core/mat.hpp:477:18: note: virtual bool cv::MatAllocator::allocate(cv::UMatData, cv::AccessFlag, cv::UMatUsageFlags) const virtual bool allocate(UMatData* data, AccessFlag accessflags, UMatUsageFlags usageFlags) const = 0; ^~~~ /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp: In member function 'int NDArrayConverter::init()': /usr3/graduate/armandok/pyslamu/pyfbow/src/lib/conversion.cpp:200:5: warning: converting to non-pointer type 'int' from NULL [-Wconversion-null] import_array(); ^~~~ make[2]: [CMakeFiles/np_opencv_converter.dir/lib/conversion.cpp.o] Error 1 make[1]: [CMakeFiles/np_opencv_converter.dir/all] Error 2 make: *** [all] Error 2