Cast

Outputs the given input tensor to the output tensor. The input is transformed from one precision to another.

All conversions between supported data types are allowed.

Attributes

toType The DataType of the output tensor.

Inputs

input: tensor of type T1.

Outputs

output: tensor of type T2.

Data Types

T1: bool, uint8, int8, int32, float16, bfloat16, float32

T2: bool, uint8, int8, int32, float16, bfloat16, float32

Shape Information

input and output are tensors with a shape of \([a_0,...,a_n]\)

Examples

Cast

in1 = network.add_input("input1", dtype=trt.float32, shape=(1, 1, 3, 3))
layer = network.add_cast(in1, trt.int32)
network.mark_output(layer.get_output(0))

inputs[in1.name] = np.array(
    [
        [
            [
                [1.2, 2.3, 3.4],
                [4.5, 5.6, 6.7],
                [7.8, 8.9, 9.1],
            ]
        ]
    ]
)

outputs[layer.get_output(0).name] = layer.get_output(0).shape

expected[layer.get_output(0).name] = np.array(
    [
        [
            [
                [1.0, 2.0, 3.0],
                [4.0, 5.0, 6.0],
                [7.0, 8.0, 9.0],
            ]
        ]
    ]
)

C++ API

For more information about the C++ ICastLayer operator, refer to the C++ ICastLayer documentation.

Python API

For more information about the Python ICastLayer operator, refer to the Python ICastLayer documentation.