tensorflow::ops::ApplyFtrl

#include <training_ops.h>

Perbarui '*var' sesuai dengan skema proksimal Ftrl.

Ringkasan

accum_new = accum + grad * grad linear += grad + (accum_new^(-lr_power) - accum^(-lr_power)) / lr * var quadratic = 1.0 / (accum_new^(lr_power) * lr) + 2 * l2 var = (tanda(linear) * l1 - linear) / kuadrat if |linear| > l1 lain 0,0 accum = accum_new

Argumen:

  • Cakupan: Objek Cakupan
  • var: Seharusnya berasal dari Variable().
  • accum: Seharusnya berasal dari Variable().
  • linear: Harus berasal dari Variable().
  • grad: Gradien.
  • lr: Faktor skala. Harus berupa besaran skalar.
  • l1: Regularisasi L1. Harus berupa skalar.
  • l2: Regularisasi L2. Harus berupa skalar.
  • lr_power: Faktor penskalaan. Harus berupa skalar.

Atribut opsional (lihat Attrs ):

  • use_locking: Jika True , pembaruan tensor var dan accum akan dilindungi oleh kunci; jika tidak, perilakunya tidak terdefinisi, tetapi mungkin menunjukkan lebih sedikit persaingan.

Mengembalikan:

Konstruktor dan Destruktor

ApplyFtrl (const :: tensorflow::Scope & scope, :: tensorflow::Input var, :: tensorflow::Input accum, :: tensorflow::Input linear, :: tensorflow::Input grad, :: tensorflow::Input lr, :: tensorflow::Input l1, :: tensorflow::Input l2, :: tensorflow::Input lr_power)
ApplyFtrl (const :: tensorflow::Scope & scope, :: tensorflow::Input var, :: tensorflow::Input accum, :: tensorflow::Input linear, :: tensorflow::Input grad, :: tensorflow::Input lr, :: tensorflow::Input l1, :: tensorflow::Input l2, :: tensorflow::Input lr_power, const ApplyFtrl::Attrs & attrs)

Atribut publik

operation
out

Fungsi publik

node () const
::tensorflow::Node *
operator::tensorflow::Input () const
operator::tensorflow::Output () const

Fungsi statis publik

UseLocking (bool x)

Struktur

tensorflow::ops::ApplyFtrl::Attrs

Pengaturan atribut opsional untuk ApplyFtrl .

Atribut publik

operasi 

Operation operation

keluar 

::tensorflow::Output out

Fungsi publik

ApplyFtrl 

 ApplyFtrl(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input var,
  ::tensorflow::Input accum,
  ::tensorflow::Input linear,
  ::tensorflow::Input grad,
  ::tensorflow::Input lr,
  ::tensorflow::Input l1,
  ::tensorflow::Input l2,
  ::tensorflow::Input lr_power
)

ApplyFtrl 

 ApplyFtrl(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input var,
  ::tensorflow::Input accum,
  ::tensorflow::Input linear,
  ::tensorflow::Input grad,
  ::tensorflow::Input lr,
  ::tensorflow::Input l1,
  ::tensorflow::Input l2,
  ::tensorflow::Input lr_power,
  const ApplyFtrl::Attrs & attrs
)

simpul 

::tensorflow::Node * node() const

operator::tensorflow::Input 

 operator::tensorflow::Input() const

operator::tensorflow::Output 

 operator::tensorflow::Output() const

Fungsi statis publik

GunakanPenguncian 

Attrs UseLocking(
  bool x
)