PyTorch torch.dequantize Function
PyTorch torch Reference Manual
torch.dequantizeIt is a function in PyTorch used to dequantize a quantized tensor into a floating-point tensor.
Function Definition
torch.dequantize(input)
Usage Example
Example
import torch
# Create the input tensor
input = torch.tensor([[-1.0, 0.0], [1.0, 2.0]])
# Quantized tensor
quantized = torch.quantize_per_tensor(input, scale=0.1, zero_point=10, dtype=torch.quint8)
print("Quantized tensor:")
print(quantized)
# Dequantize
x = torch.dequantize(quantized)
print("Dequantized floating-point tensor:")
print(x)
# Create the input tensor
input = torch.tensor([[-1.0, 0.0], [1.0, 2.0]])
# Quantized tensor
quantized = torch.quantize_per_tensor(input, scale=0.1, zero_point=10, dtype=torch.quint8)
print("Quantized tensor:")
print(quantized)
# Dequantize
x = torch.dequantize(quantized)
print("Dequantized floating-point tensor:")
print(x)
Other Extensions