Inheritance diagram for dlprim::Hardtanh:
Public Member Functions | |
| Hardtanh (Context &ctx, HardtanhConfig const &cfg=HardtanhConfig()) | |
| virtual char const * | operator_type () const |
| name of the operator type | |
| virtual void | forward_cpu_float (size_t n, float const *x, float *y) |
| virtual void | backward_cpu_float (size_t n, float const *, float *, float const *, float const *, float beta) |
| virtual void | forward_gpu (Tensor &x, Tensor &y, ExecutionContext const &q) |
| virtual void | backward_gpu (Tensor &x, Tensor &dx, Tensor &y, Tensor &dy, float beta, ExecutionContext const &q) |
 Public Member Functions inherited from dlprim::PointwiseBase | |
| PointwiseBase (Context &ctx) | |
| virtual void | setup (std::vector< TensorSpecs > const &in, std::vector< TensorSpecs > &out, std::vector< TensorSpecs > ¶meters, size_t &workspace) |
| Convigure operator. More... | |
| virtual void | reshape (std::vector< Shape > const &in, std::vector< Shape > &out, size_t &ws) |
| Reshape layer according to new input size. More... | |
| virtual void | forward (std::vector< Tensor > &input, std::vector< Tensor > &output, std::vector< Tensor > ¶meters, Tensor &ws, ExecutionContext const &q) |
| Enqueue forward propogation computations. More... | |
| virtual void | backward (std::vector< TensorAndGradient > &input, std::vector< TensorAndGradient > &output, std::vector< TensorAndGradient > ¶meters, Tensor &workspace, ExecutionContext const &ctx) |
| Enqueue backward propogation computations. More... | |
| Public Member Functions inherited from dlprim::Operator | |
| Operator (Context const &ctx) | |
| Create operator for specific context (device/platform) | |
| SharedResource & | shared_resource () |
| Getter for object that is shared between operators accross the net, for example random numbers generator. | |
| void | shared_resource (std::shared_ptr< SharedResource > r) |
| Setter of the shared resource. | |
| virtual void | mode (CalculationsMode mode) |
| Can be called with both train and predict before setup() is called. afterwards if original mode was train - it can be switched to predict and back but if original mode was predict it can't be switched to train. More... | |
| virtual CalculationsMode | mode () |
| get current mode | |
| Operator (Operator const &)=delete | |
| operator non-copyable/moveable More... | |
| void | operator= (Operator const &)=delete |
| Operator (Operator &&)=delete | |
| void | operator= (Operator &&)=delete |
| virtual bool | alias_generator () |
| returns true of the operator is alias - generation - it only changes the shape of tensor but not its content it actually does not perform any operation but only changes semantics, in this casse input and output are aliases of each other | |
| virtual void | initialize_params (std::vector< Tensor > &, ExecutionContext const &) |
| Set default parameters iniitalization. | |
Additional Inherited Members | |
| Protected Member Functions inherited from dlprim::PointwiseBase | |
| virtual void | forward_cpu (Tensor &x, Tensor &y) |
| virtual void | backward_cpu (Tensor &x, Tensor &dx, Tensor &y, Tensor &dy, float beta) |
| Protected Attributes inherited from dlprim::Operator | |
| Context | ctx_ |
| OpenCL/CPU Context to work with. | |
| CalculationsMode | mode_ |
| computaions mode | |
| std::shared_ptr< SharedResource > | shared_resource_ |
The documentation for this class was generated from the following file:
- include/dlprim/ops/pointwise.hpp
