tf.assign()用法

转自TensorFlow官网

tf.assign(
    ref,
    value,
    validate_shape=None,
    use_locking=None,
    name=None
)

Update 'ref' by assigning 'value' to it 

通过给‘ref’分配‘value’完成‘ref’的更新。

This operation outputs a Tensor that holds the new value of 'ref' after the value has been assigned. This makes it easier to chain operations that need to use the reset value.

本操作输出一个‘ref’被更新的张量。这样可以更容易地链接需要使用重置值的操作。

Args:

• ref: A mutable Tensor. Should be from a Variable node. May be uninitialized.
• value: A Tensor. Must have the same type as ref. The value to be assigned to the variable.
• validate_shape: An optional bool. Defaults to True. If true, the operation will validate that the shape of 'value' matches the shape of the Tensor being assigned to. If false, 'ref' will take on the shape of 'value'.
• use_locking: An optional bool. Defaults to True. If True, the assignment will be protected by a lock; otherwise the behavior is undefined, but may exhibit less contention.
• name: A name for the operation (optional).