mirror of
https://github.com/pytorch/pytorch.git
synced 2025-10-21 05:34:18 +08:00
29f0e1e2ce
Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/48407 T79817692: Fused8BitRowwiseQuantizedToFloat operator support for c2_pt_converter. Also refactored some repeated code from the existing test functions. (Initial commit only has refactoring.) Test Plan: buck test //caffe2/torch/fb/model_transform/c2_convert:c2_pt_converter_test Reviewed By: bugra Differential Revision: D25069936 fbshipit-source-id: 72f6a845a1b4639b9542c6b230c8cd74b06bc5a0
10 lines
261 B
Python
10 lines
261 B
Python
# @package quantization
|
|
# Module caffe2.python.helpers.quantization
|
|
|
|
|
|
def fused_8bit_rowwise_quantized_to_float(
|
|
model, blob_in, blob_out
|
|
):
|
|
"""Fused8BitRowwiseQuantizedToFloat"""
|
|
return model.net.Fused8BitRowwiseQuantizedToFloat(blob_in, blob_out)
|