Omegastick
commented
4 years ago Sorry for the slow reply. Not really sure what's going on here. It builds fine on my machine and in CI. Perhaps a difference between gcc 8 and 9. I could imagine that 9 stopped inferring the template parameters or something. I'll have a try at building it with gcc-9 when I get a chance.
compiling with g++ 9.2.1 on Ubuntu 16.04 I get the following compiler error; any ideas what might
output of make command:
Scanning dependencies of target cpprl [ 1%] Building CXX object CMakeFiles/cpprl.dir/src/storage.cpp.o [ 1%] Building CXX object CMakeFiles/cpprl.dir/src/observation_normalizer.cpp.o /home/dimid/Documents/misc/pytorch-cpp-rl/src/observation_normalizer.cpp: In member function ‘std::vector<float, std::allocator > cpprl::ObservationNormalizerImpl::get_mean() const’:
/home/dimid/Documents/misc/pytorch-cpp-rl/src/observation_normalizer.cpp:79:30: error: expected primary-expression before ‘(’ token
79 | return std::vector(mean.data_ptr(), mean.data_ptr() + mean.numel());
| ^
/home/dimid/Documents/misc/pytorch-cpp-rl/src/observation_normalizer.cpp:79:45: error: expected primary-expression before ‘float’
79 | return std::vector(mean.data_ptr(), mean.data_ptr() + mean.numel());
| ^~~~~
/home/dimid/Documents/misc/pytorch-cpp-rl/src/observation_normalizer.cpp:79:69: error: expected primary-expression before ‘float’
79 | return std::vector(mean.data_ptr(), mean.data_ptr() + mean.numel());
| ^~~~~
/home/dimid/Documents/misc/pytorch-cpp-rl/src/observation_normalizer.cpp: In member function ‘std::vector<float, std::allocator > cpprl::ObservationNormalizerImpl::get_variance() const’:
/home/dimid/Documents/misc/pytorch-cpp-rl/src/observation_normalizer.cpp:85:30: error: expected primary-expression before ‘(’ token
85 | return std::vector(variance.data_ptr(), variance.data_ptr() + variance.numel());
| ^
/home/dimid/Documents/misc/pytorch-cpp-rl/src/observation_normalizer.cpp:85:49: error: expected primary-expression before ‘float’
85 | return std::vector(variance.data_ptr(), variance.data_ptr() + variance.numel());
| ^~~~~
/home/dimid/Documents/misc/pytorch-cpp-rl/src/observation_normalizer.cpp:85:77: error: expected primary-expression before ‘float’
85 | return std::vector(variance.data_ptr(), variance.data_ptr() + variance.numel());
| ^~~~~
CMakeFiles/cpprl.dir/build.make:75: recipe for target 'CMakeFiles/cpprl.dir/src/observation_normalizer.cpp.o' failed
make[2]: [CMakeFiles/cpprl.dir/src/observation_normalizer.cpp.o] Error 1
CMakeFiles/Makefile2:285: recipe for target 'CMakeFiles/cpprl.dir/all' failed
make[1]: [CMakeFiles/cpprl.dir/all] Error 2
Makefile:129: recipe for target 'all' failed
make: *** [all] Error 2