HOTTBOX: Higher Order Tensors ToolBOX.
Makefile
utilises pipenv
Makefile
pandas==1.0.0
. Deprecation of as_matrix()
in favour of values
numpy
. Deprecation of non-tuple sequence for multidimensional indexing.CMTF
- An algorithm for Coupled Matrix and Tensor factorization for two Tensors
of order n and 2 with respect to a specified rank
RandomisedCPD
- An algorithm for Randomised Canonical Polyadic Decomposition.Parafac2
- An algorithm for PARAFAC2 model computed via ALSTelVI
and TelVAC
- Algorithms under Tensor Ensemble Learning (TEL) frameworkguzzle_sphinx_theme
hottbox
develop
branchTensor
and vise versa.Tensor
, TensorCPD
, TensorTKD
and TensorTT
classes.Mode
for meta information about data modes of for tensor representations.
It is stored in _modes
as list.State
that tracks data manipulation operation applied to Tensor
.
It is stored in _state
.Tensor
in the unfolded form.Tensor
TensorCPD
, TensorTKD
and TensorTT
classes
by analogy with the Tensor
classkeep_meta
to decompose
methods for the cpd and tucker type decompositions.
Based on its value, meta information of the modes of tensor
to be decomposed can be extracted
and assigned to the TensorCPD
and TensorTKD
respectively.Tensor
objects (redefined __add__
, __eq__
)TensorCPD
and TensorTKD
object (redefined __add__
, __eq__
)__str__
and __repr__
for Tensor
, TensorCPD
, TensorTKD
and TensorTT
__repr__
for tensor decomposition algorithms.Tensor
classfold
, unfold
and mode_n_product
of Tensor
.
Whether they can be called is determined by the current state of the Tensor
object.Tensor
there is characterised by a corresponding Mode
object with meta informationTensor
constructor should be passed as list instead of OrderedDict.
These names are used to create Mode
objects which are stored in a list Tensor._modes
reconstruct
of TensorCPD
, TensorTKD
and TensorTT
classes is now a method
(should have been in the first place). Also it take optional parameter keep_mata
for extraction
of meta information about modesdescribe
functionality is now implemented by __str__
. Instead, describe
provides some
statistics of the Tensor
by analogy with pandas
ft_shape
from the TensorTT
constructorft_shape
and the corresponding attribute from the Tensor
constructor.mode_description
from constructors for all tensor decomposition algorithms_mode_names
from the Tensor
TensorCPD
and TensorTKD
due to new attributescopy
method for the core tensor structuresdescribe
method that describes an instance of Tensor
classTensor
through the use of OrderedDict.
Modes can also be renamedTensor
, TensorCPD
, TensorTKD
, TensorTT
decompose
method for all tensor decomposition algorithmsTensor
, TensorCPD
, TensorTKD
, TensorTT
classes can only be created from numpy arraysTensor
classsuper_diag_tensor
requires to pass a shape of desired tensor instead of its orderreconstruct
was changing the original core so it was not possible to call it several times in a rowdecompose
for HOSVD
and HOOI
classesTensor
, TensorCPD
, TensorTKD
, TensorTT
)super_diag_tensor
, residual_tensor
)CPD
, HOSVD
, HOOI
,TTSVD
)Tensor