45 vector<vector<vector<TensorView>>>
views;
Container of layers forming a feed-forward neural network, with parameter storage and I/O.
Definition neural_network.h:20
Definition adaptive_moment_estimation.h:14
Owning raw byte buffer that lives on CPU or CUDA memory, with aligned (re)allocation.
Definition tensor_utilities.h:166
Buffer data
Definition forward_propagation.h:44
vector< vector< vector< TensorView > > > views
Definition forward_propagation.h:45
ForwardPropagation(const Index=0, NeuralNetwork *=nullptr)
Constructs a workspace for the given batch size and network.
void set(const Index=0, NeuralNetwork *=nullptr)
Reconfigures the workspace for a new batch size or network; reuses allocations when possible.
Index batch_size
Definition forward_propagation.h:40
NeuralNetwork * neural_network
Definition forward_propagation.h:42
TensorView get_outputs() const
Returns the final output tensor of the network.
void print() const
Prints a human-readable summary of the workspace contents.
TensorView get_last_trainable_layer_outputs() const
Returns the output tensor of the last trainable layer.
Non-owning view over a tensor: pointer, shape, and data type with rich reshape helpers.
Definition tensor_utilities.h:293