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.