dl::optim::Adam Class Reference

Implements the Adam optimization algorithm [4] . More...

#include <adam.hpp>

Inheritance diagram for dl::optim::Adam:

Collaboration diagram for dl::optim::Adam:

Public Member Functions
	Adam (float lr=0.001f, float beta1=0.9f, float beta2=0.999f, float eps=1e-8f)
virtual void	step (dl::TensorPtr &loss) override

Detailed Description

Implements the Adam optimization algorithm [4] .

Definition at line 10 of file adam.hpp.

Constructor & Destructor Documentation

Adam()

dl::optim::Adam::Adam ( float  lr = 0.001f, float  beta1 = 0.9f, float  beta2 = 0.999f, float  eps = 1e-8f  )

inline

Definition at line 18 of file adam.hpp.

                : dl::Optimizer(), lr(lr), beta1(beta1), beta2(beta2), eps(eps) {}

Member Function Documentation

step()

virtual void dl::optim::Adam::step ( dl::TensorPtr &  loss )

inlineoverridevirtual

Implements dl::Optimizer.

Definition at line 21 of file adam.hpp.

{ throw std::runtime_error("Not yet implemented"); }

---

The documentation for this class was generated from the following file:

• dl/learning/optimizers/adam.hpp