tf.raw_ops.ResourceApplyAdaMax

Stay organized with collections Save and categorize content based on your preferences.

Update '*var' according to the AdaMax algorithm.

View aliases

Compat aliases for migration

tf.compat.v1.raw_ops.ResourceApplyAdaMax

tf.raw_ops.ResourceApplyAdaMax(
    var,
    m,
    v,
    beta1_power,
    lr,
    beta1,
    beta2,
    epsilon,
    grad,
    use_locking=False,
    name=None
)

m_t <- beta1 * m_{t-1} + (1 - beta1) * g v_t <- max(beta2 * v_{t-1}, abs(g)) variable <- variable - learning_rate / (1 - beta1^t) * m_t / (v_t + epsilon)

Args
var	A Tensor of type resource. Should be from a Variable().
m	A Tensor of type resource. Should be from a Variable().
v	A Tensor of type resource. Should be from a Variable().
beta1_power	A Tensor. Must be one of the following types: float32, float64, int32, uint8, int16, int8, complex64, int64, qint8, quint8, qint32, bfloat16, qint16, quint16, uint16, complex128, half, uint32, uint64. Must be a scalar.
lr	A Tensor. Must have the same type as beta1_power. Scaling factor. Must be a scalar.
beta1	A Tensor. Must have the same type as beta1_power. Momentum factor. Must be a scalar.
beta2	A Tensor. Must have the same type as beta1_power. Momentum factor. Must be a scalar.
epsilon	A Tensor. Must have the same type as beta1_power. Ridge term. Must be a scalar.
grad	A Tensor. Must have the same type as beta1_power. The gradient.
use_locking	An optional bool. Defaults to False. If True, updating of the var, m, and v tensors will be protected by a lock; otherwise the behavior is undefined, but may exhibit less contention.
name	A name for the operation (optional).

Returns
The created Operation.

tf.raw_ops.ResourceApplyAdaMax | TensorFlow v2.16.1 (original) (raw)

View aliases