neural_compressor.data.transforms
Neural Compressor Built-in transforms for multiple framework backends.
Submodules
Package Contents
Classes
|
Transforms collection class. |
|
The base class for transform. |
|
Composes several transforms together. |
|
Resize the input image to the given size. |
|
Resize image with aspect ratio and pad it to max shape(optional). |
|
Scale the values of image to [0,1]. |
|
Normalize a image with mean and standard deviation. |
|
Postprocess the predictions of bert on SQuAD. |
|
Postprocess the predictions of bert on SQuADV1.1. |
|
Coco decoding will be performed automatically from Neural Compressor v1.4. |
|
Just collect the infos to construct a Postprocess. |
|
Convert label to label - label_shift. |
|
Combination of a series of transforms which is applicable to images in Imagenet. |
|
Combination of a series of transforms which is applicable to images in Imagenet. |
|
Label shift by 1 and rescale. |
Functions
|
Class decorator used to register all transform subclasses. |