* @brief Assert Datatype, Backend, data format and dimensions along the GraphView are coherent.
* If not, apply the required transformations.
...
...
@@ -282,13 +282,16 @@ public:
* compatible with the selected kernel.
* If not, add a Transpose Operator.
* 4 - Propagate Tensor dimensions through the consecutive Operators.
@params string: backend @todo: explain params
@params Aidge Datatype: datatype
@params vector of vector of DimSize_t: dims
*/
voidcompile(conststd::string&backend="cpu",
constAidge::DataTypedatatype=DataType::Float32,
DeviceIdx_tdevice=0,
conststd::vector<std::vector<DimSize_t>>dims={});
/**
/** @todo naming the input dims, while there exist currentTensorPtr->dims is a nice way to mess with reader's head maybe some proper variable name would be great
* @brief Compute dimensions of input/output Tensors for each Operator of the
* GraphView object's Nodes, by calling Node::forwardDims().
* This function verifies the following conditions: