Christophe Guillon authored 9 months ago and Maxence Naud committed 8 months ago

Define a Tensor with undefined dims by explicitly setting mSize to 0.
Provide undefined() method which must be used instead of empty()
to test whether a Tensor as associated dimensions and can be used
for forwarding dimensions downward.
This disambiguates from defined scalar Tensor which actually have
empty dimensions, but size of 1.
Note that as soon as a Tensor with undefined dimensions is resized,
it's dimensions are permanently defined (though may still change).

This change solves issues in operators forwardDims() when
input values are scalar, i.e. !undefined() and empty().

- [#include] Remove duplicated includes, add includes
- [attributes] add default values for mDataType, mImpl and mGrad
- [constructors] Order constructors, change array copy assignment to remove code duplication
- calling grad() does not instanciate the gradient anymore, a dedicated function should do it
- move getAvailableBackends() and operator=(const Tensor&) to cpp file
- Tensor::resize() now handles scalar Tensors

- [Add] ``zeros()`` member function to set implpementation elements to 0
- [unit_tests][NF] Add many more cases in Test_TensorImpl.cpp
- [include] update includes in Tensor.hpp

- [Add] ``zeros()`` member function to set implpementation elements to 0
- [unit_tests][NF] Add many more cases in Test_TensorImpl.cpp
- [include] update includes in Tensor.hpp

- [Add] ``zeros()`` member function to set implpementation elements to 0
- [unit_tests][NF] Add many more cases in Test_TensorImpl.cpp
- [include] update includes in Tensor.hpp

Update TensorImpl constructor to take the tensor  dimensions instead of the number of elements. Necessary for backend_opencv.