Fail to pack with openai clip on macos

[YeDaxia]

Context information (for bug reports)

• Output of pyinstaller --version: 5.2
• Version of Python: 3.9.10
• Platform: macOS Monterey 12.3.1 (21E258)
• How you installed Python: virtualenv
• Did you also try this on another platform? Does it work there? (Just try on mac os)

A minimal example program which shows the error

**install clip: **

pip install torch==1.10.0 torchvision==0.11.1 numpy==1.19.5 pandas==1.1.5 ftfy==6.0.3 tqdm==4.62.3
pip install git+https://github.com/openai/CLIP.git

main.py:

import clip  # remove this line, it works.

if __name__ == '__main__':
    print('hello world')

pack main.py:

pyinstaller  main.py

Stacktrace / full error message

run main application:

./dist/main/main

error:

libc++abi: terminating with uncaught exception of type c10::Error: Type c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > could not be converted to any of the known types.

Exception raised from operator() at ../aten/src/ATen/core/jit_type.h:1674 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >) + 98 (0x1135e6522 in libc10.dylib)

frame #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) + 106 (0x1135e4c3a in libc10.dylib)
frame #2: c10::detail::getTypePtr_<c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > >::call()::'lambda'()::operator()() const + 260 (0x16e62a234 in libtorch_cpu.dylib)
frame #3: c10::detail::getTypePtr_<c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > >::call() + 27 (0x16e629e4b in libtorch_cpu.dylib)
frame #4: c10::detail::infer_schema::(anonymous namespace)::createArgumentVector(c10::ArrayRef<c10::detail::infer_schema::ArgumentDef>) + 206 (0x16e13dede in libtorch_cpu.dylib)
frame #5: c10::detail::infer_schema::make_function_schema(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >&&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >&&, c10::ArrayRef<c10::detail::infer_schema::ArgumentDef>, c10::ArrayRef<c10::detail::infer_schema::ArgumentDef>) + 123 (0x16e13dc8b in libtorch_cpu.dylib)
frame #6: c10::detail::infer_schema::make_function_schema(c10::ArrayRef<c10::detail::infer_schema::ArgumentDef>, c10::ArrayRef<c10::detail::infer_schema::ArgumentDef>) + 76 (0x16e13e20c in libtorch_cpu.dylib)
frame #7: std::__1::unique_ptr<c10::FunctionSchema, std::__1::default_delete<c10::FunctionSchema> > c10::detail::inferFunctionSchemaFromFunctor<at::Tensor (*)(at::Tensor, c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&, double, long long)>() + 110 (0x16e6920de in libtorch_cpu.dylib)
frame #8: torch::CppFunction::CppFunction<at::Tensor (at::Tensor, c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&, double, long long)>(at::Tensor (*)(at::Tensor, c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&, double, long long), std::__1::enable_if<c10::guts::is_function_type<at::Tensor (at::Tensor, c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&, double, long long)>::value, std::nullptr_t>::type) + 156 (0x16e691ffc in libtorch_cpu.dylib)
frame #9: torch::Library& torch::Library::impl<at::Tensor (&)(at::Tensor, c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&, double, long long)>(torch::detail::SelectiveStr<true>, at::Tensor (&)(at::Tensor, c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&, double, long long)) & + 36 (0x16e690144 in libtorch_cpu.dylib)
frame #10: at::native::(anonymous namespace)::TORCH_LIBRARY_IMPL_init_quantized_QuantizedCPU_4(torch::Library&) + 36 (0x16e690014 in libtorch_cpu.dylib)
frame #11: torch::detail::TorchLibraryInit::TorchLibraryInit(torch::Library::Kind, void (*)(torch::Library&), char const*, c10::optional<c10::DispatchKey>, char const*, unsigned int) + 204 (0x16dfee07c in libtorch_cpu.dylib)
frame #12: _GLOBAL__sub_I_qconv.cpp + 72 (0x16e698bf8 in libtorch_cpu.dylib)
frame #13: invocation function for block in dyld4::Loader::findAndRunAllInitializers(dyld4::RuntimeState&) const + 182 (0x115ed1ddb in dyld)
frame #14: invocation function for block in dyld3::MachOAnalyzer::forEachInitializer(Diagnostics&, dyld3::MachOAnalyzer::VMAddrConverter const&, void (unsigned int) block_pointer, void const*) const + 129 (0x115ef80bb in dyld)
frame #15: invocation function for block in dyld3::MachOFile::forEachSection(void (dyld3::MachOFile::SectionInfo const&, bool, bool&) block_pointer) const + 557 (0x115eef85a in dyld)
frame #16: dyld3::MachOFile::forEachLoadCommand(Diagnostics&, void (load_command const*, bool&) block_pointer) const + 129 (0x115ebedb3 in dyld)
frame #17: dyld3::MachOFile::forEachSection(void (dyld3::MachOFile::SectionInfo const&, bool, bool&) block_pointer) const + 179 (0x115eef5eb in dyld)
frame #18: dyld3::MachOAnalyzer::forEachInitializerPointerSection(Diagnostics&, void (unsigned int, unsigned int, unsigned char const*, bool&) block_pointer) const + 118 (0x115ef7aec in dyld)
frame #19: dyld3::MachOAnalyzer::forEachInitializer(Diagnostics&, dyld3::MachOAnalyzer::VMAddrConverter const&, void (unsigned int) block_pointer, void const*) const + 386 (0x115ef7d5e in dyld)
frame #20: dyld4::Loader::findAndRunAllInitializers(dyld4::RuntimeState&) const + 144 (0x115ed1d0e in dyld)
frame #21: dyld4::Loader::runInitializersBottomUp(dyld4::RuntimeState&, dyld3::Array<dyld4::Loader const*>&) const + 178 (0x115ed1e9a in dyld)
frame #22: dyld4::Loader::runInitializersBottomUp(dyld4::RuntimeState&, dyld3::Array<dyld4::Loader const*>&) const + 149 (0x115ed1e7d in dyld)
frame #23: dyld4::Loader::runInitializersBottomUp(dyld4::RuntimeState&, dyld3::Array<dyld4::Loader const*>&) const + 149 (0x115ed1e7d in dyld)
frame #24: dyld4::Loader::runInitializersBottomUpPlusUpwardLinks(dyld4::RuntimeState&) const + 108 (0x115ed1f3e in dyld)
frame #25: dyld4::APIs::dlopen_from(char const*, int, void*) + 638 (0x115ee0b6e in dyld)
frame #26: py_dl_open + 121 (0x10e591edf in _ctypes.cpython-39-darwin.so)
frame #27: cfunction_call + 90 (0x10e72857b in Python)
frame #28: _PyObject_MakeTpCall + 127 (0x10e6dcdd4 in Python)
frame #29: call_function + 279 (0x10e7d3fdf in Python)
frame #30: _PyEval_EvalFrameDefault + 20490 (0x10e7cb9da in Python)
frame #31: _PyEval_EvalCode + 411 (0x10e7c58d6 in Python)
frame #32: _PyFunction_Vectorcall + 373 (0x10e6dd87e in Python)
frame #33: method_vectorcall + 365 (0x10e6e0480 in Python)
frame #34: _PyEval_EvalFrameDefault + 2609 (0x10e7c7401 in Python)
frame #35: _PyEval_EvalCode + 411 (0x10e7c58d6 in Python)
frame #36: _PyFunction_Vectorcall + 373 (0x10e6dd87e in Python)
frame #37: _PyObject_FastCallDictTstate + 87 (0x10e6dcfd2 in Python)
frame #38: slot_tp_init + 192 (0x10e74be9c in Python)
frame #39: type_call + 272 (0x10e7512e6 in Python)
frame #40: _PyObject_MakeTpCall + 127 (0x10e6dcdd4 in Python)
frame #41: call_function + 279 (0x10e7d3fdf in Python)
frame #42: _PyEval_EvalFrameDefault + 27838 (0x10e7cd68e in Python)
frame #43: _PyFunction_Vectorcall + 192 (0x10e6dd7c9 in Python)
frame #44: method_vectorcall + 160 (0x10e6e03b3 in Python)
frame #45: call_function + 168 (0x10e7d3f70 in Python)
frame #46: _PyEval_EvalFrameDefault + 27838 (0x10e7cd68e in Python)
frame #47: _PyEval_EvalCode + 411 (0x10e7c58d6 in Python)
frame #48: builtin_exec + 389 (0x10e7c061d in Python)
frame #49: cfunction_vectorcall_FASTCALL + 91 (0x10e728e34 in Python)
frame #50: call_function + 168 (0x10e7d3f70 in Python)
frame #51: _PyEval_EvalFrameDefault + 20490 (0x10e7cb9da in Python)
frame #52: _PyFunction_Vectorcall + 192 (0x10e6dd7c9 in Python)
frame #53: call_function + 168 (0x10e7d3f70 in Python)
frame #54: _PyEval_EvalFrameDefault + 23803 (0x10e7cc6cb in Python)
frame #55: _PyFunction_Vectorcall + 192 (0x10e6dd7c9 in Python)
frame #56: call_function + 168 (0x10e7d3f70 in Python)
frame #57: _PyEval_EvalFrameDefault + 20490 (0x10e7cb9da in Python)
frame #58: _PyFunction_Vectorcall + 192 (0x10e6dd7c9 in Python)
frame #59: call_function + 168 (0x10e7d3f70 in Python)
frame #60: _PyEval_EvalFrameDefault + 20490 (0x10e7cb9da in Python)
frame #61: _PyFunction_Vectorcall + 192 (0x10e6dd7c9 in Python)
frame #62: object_vacall + 264 (0x10e6deca0 in Python)
frame #63: _PyObject_CallMethodIdObjArgs + 252 (0x10e6deedd in Python)

[rokm]

This is very likely caused by duplication of torch shared libraries on macOS - see pyinstaller/pyinstaller-hooks-contrib#375 (from this comment on) for details and for manual work-around.

[YeDaxia]

This is very likely caused by duplication of torch shared libraries on macOS - see pyinstaller/pyinstaller-hooks-contrib#375 (from this comment on) for details and for manual work-around.

Thank you for your prompt reply. It works!