tf.raw_ops.ResourceApplyAdamWithAmsgrad

Update '*var' according to the Adam algorithm.

$$\text{lr}_t := \mathrm{learning_rate} * \sqrt{1 - \beta_2^t} / (1 - \beta_1^t)$$
$$m_t := \beta_1 * m_{t-1} + (1 - \beta_1) * g$$
$$v_t := \beta_2 * v_{t-1} + (1 - \beta_2) * g * g$$
$$\hat{v}_t := max{\hat{v}_{t-1}, v_t}$$
$$\text{variable} := \text{variable} - \text{lr}_t * m_t / (\sqrt{\hat{v}_t} + \epsilon)$$
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().
vhat 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, uint16, complex128, half, uint32, uint64. Must be a scalar.
beta2_power A Tensor. Must have the same type as beta1_power. 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.

© 2020 The TensorFlow Authors. All rights reserved.
Licensed under the Creative Commons Attribution License 3.0.
Code samples licensed under the Apache 2.0 License.
https://www.tensorflow.org/versions/r2.4/api_docs/python/tf/raw_ops/ResourceApplyAdamWithAmsgrad