tensorplay.functional.quantize_per_channel

tensorplay.functional.quantize_per_channel(input, scales, zero_points, axis, dtype) → Tensor[source]

Converts a float tensor to a per-channel quantized tensor with given scales and zero points.

Parameters:

• input (Tensor) – float tensor to quantize

• scales (Tensor) – float 1D tensor of scales to use, size should match input.size(axis)

• zero_points (int) – integer 1D tensor of offset to use, size should match input.size(axis)

• axis (int) – dimension on which apply per-channel quantization

• dtype (tensorplay.dtype) – the desired data type of returned tensor. Has to be one of the quantized dtypes: tensorplay.quint8, tensorplay.qint8, tensorplay.qint32

Returns:

A newly quantized tensor

Return type:

Tensor

Example:

>>> x = tensorplay.tensor([[-1.0, 0.0], [1.0, 2.0]])
>>> tensorplay.quantize_per_channel(x, tensorplay.tensor([0.1, 0.01]), tensorplay.tensor([10, 0]), 0, tensorplay.quint8)
tensor([[-1.,  0.],
        [ 1.,  2.]], size=(2, 2), dtype=tensorplay.quint8,
       quantization_scheme=tensorplay.per_channel_affine,
       scale=tensor([0.1000, 0.0100], dtype=tensorplay.float64),
       zero_point=tensor([10,  0]), axis=0)
>>> tensorplay.quantize_per_channel(x, tensorplay.tensor([0.1, 0.01]), tensorplay.tensor([10, 0]), 0, tensorplay.quint8).int_repr()
tensor([[  0,  10],
        [100, 200]], dtype=tensorplay.uint8)