PyTorch torch.quantize_per_tensor Function
Pytorch torch Reference Manual
torch.quantize_per_tensorIt is a function in PyTorch used to create a quantized tensor with per-tensor quantization.
Function Definition
torch.quantize_per_tensor(input, scale, zero_point, dtype)
Usage Example
Example
import torch
# Create input tensor
input = torch.tensor([[-1.0, 0.0], [1.0, 2.0]])
# Quantize per tensor
# scale: quantization scale factor
# zero_point: zero point
# dtype: quantization data type
x = torch.quantize_per_tensor(input, scale=0.1, zero_point=10, dtype=torch.quint8)
print("Quantized tensor:")
print(x)
print("Dequantized original values:")
print(x.dequantize())
# Create input tensor
input = torch.tensor([[-1.0, 0.0], [1.0, 2.0]])
# Quantize per tensor
# scale: quantization scale factor
# zero_point: zero point
# dtype: quantization data type
x = torch.quantize_per_tensor(input, scale=0.1, zero_point=10, dtype=torch.quint8)
print("Quantized tensor:")
print(x)
print("Dequantized original values:")
print(x.dequantize())
Other Extensions