FilesLast commitLast update
1 month ago
1 month ago
1 month ago
1 month ago
1 month ago
22 days ago
9 days ago
README

IsVariableInitialized

English | 简体中文

Product Support Status

Product Supported
Ascend 950PR&950DT products √
Atlas A3 products √
Atlas A2 products √
Atlas 200I/500 A2 inference products √
Atlas inference products √
Atlas training products √

Function Description

Checks whether a tensor is initialized. The IsVariableInitialized operator receives the input tensor x and outputs the scalar Boolean value y, indicating whether the input tensor is initialized. InferShape sets the output shape to a scalar (0-dimensional), and InferDataType sets the output data type to DT_BOOL.

Parameter Description

Parameter Name Input/Output/Attribute Description Data Type Data Format
x Input Input tensor. TensorType::ALL() ND
y Output Scalar Boolean tensor, indicating whether the input tensor is initialized. Its shape is a scalar (0-dimensional), and its data type is DT_BOOL. TensorType({DT_BOOL}) ND

Constraint Description

  • The output y is a scalar (0-dimensional tensor), and its data type is fixed to DT_BOOL.

Invocation Description

The IsVariableInitialized operator is a basic operator of the graph engine. It is usually used to check whether a variable is initialized and is not invoked separately.

Invocation Method Invocation Sample Description
Graph-mode invocation test_geir_is_variable_initialized Invokes the IsVariableInitialized operator by building a graph through the operator IR