openfl.pipelines.stc_pipeline.SparsityTransformer

openfl.pipelines.stc_pipeline.SparsityTransformer#

class openfl.pipelines.stc_pipeline.SparsityTransformer(p=0.01)[source]#

Bases: Transformer

A transformer class to sparsify input data.

p#

The sparsity ratio.

Type:

float

lossy#

A flag indicating if the transformation is lossy.

Type:

bool

__init__(p=0.01)[source]#

Initialize.

Parameters:

p (float, optional) – The sparsity ratio. Defaults to 0.01.

Methods

__init__([p])

Initialize.

backward(data, metadata, **kwargs)

Recover data array with the right shape and numerical type.

forward(data, **kwargs)

Sparsify data and pass over only non-sparsified elements by reducing the array size.
backward(data, metadata, **kwargs)[source]#

Recover data array with the right shape and numerical type.

Parameters:

  • data – an numpy array with non-zero values.

  • metadata – dictionary to contain information for recovering back to original data array.

Returns:

an numpy array with original shape.

Return type:

recovered_data

forward(data, **kwargs)[source]#

Sparsify data and pass over only non-sparsified elements by reducing the array size.

Parameters:

data – an numpy array from the model tensor_dict.

Returns:

a flattened, sparse representation of the input

tensor.

metadata: dictionary to store a list of meta information.

Return type:

sparse_data

Contents