**Notes**: **1. This API is ONLY available in Dygraph mode** **2. Use it only Variable has gradient, normally we use

Get a Variable by name from this block recursively. Args: name(str): the Variable's name. Returns:

Create parameters for this layer. Parameters: shape(list): Shape of the parameter. The data type in the list must be int.

Multiplies a Tensor by the scale factor and returns scaled outputs. If this instance of :class:`GradScaler` is not enabled, output ar

:api_attr: Static Graph Change the global main program and startup program with ``with`` statement. Layer functions in the Python ``with

Indicating shape of current Variable **Notes: This is a read-only property** Examples: .. code-block:: pycon

\brief Access to the storage instance.

Get the attribute dict. Returns: dict: The Operator's attribute dict, name->attr.

Execute the optimizer and update parameters once. Returns: None Examples: .. code-block:: pycon

Mean of beta distribution.

Pushes a variable onto the stack. Args: val: The variable to be pushed.

Update the auc curve with the given predictions and labels. Args: preds (numpy.array): An numpy array in the shape of

Currently we use grad_node_ to identify if a edge is initialized.

Note: when calling is_float(16), 'st' can't be specific_type_t::None to distinguish FP16/BF16, or use is_float16()/is_bfloat16() for short

Change the operator type into new type. Args: new_type(str): new operator type to be set.