bitorch.quantizations.base.Quantization¶
- class bitorch.quantizations.base.Quantization[source]¶
superclass for quantization modules
Methods
Quantizes the tensor using this classes quantize-method.
Apply the quantization function to the input tensor.
Attributes
bit_width
bitwidth
name
- forward(x: Tensor) Tensor [source]¶
Quantizes the tensor using this classes quantize-method. Subclasses shall add some semantic there.
- Parameters:
x (torch.Tensor) – tensor to be forwarded.
- Returns:
quantized tensor x
- Return type:
torch.Tensor
- quantize(x: Tensor) Tensor [source]¶
Apply the quantization function to the input tensor.
It is recommended to use a torch.Function to also manipulate backwards behavior. See the implementations of sign or dorefa quantization functions for more examples.
- Parameters:
x (torch.Tensor) – the input to be quantized
- Raises:
NotImplementedError – raised if quantize function of superclass is called.
- Returns:
the quantized tensor
- Return type:
torch.Tensor