Impute

DFMF Imputer

class imml.impute.DFMFImputer(**kwargs)

Bases: DFMF

Impute missing data in a dataset using the DFMF method. [1] [2]

This class extends the DFMF class to provide functionality for filling in incomplete samples by addressing both block-wise and feature-wise missing data. As a subclass of DFMF, DFMFImputer inherits all input parameters and attributes from DFMF. Consequently, it uses the same fit method as DFMF for training the model.

References

[1]

M. Žitnik and B. Zupan, "Data Fusion by Matrix Factorization," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 37, no. 1, pp. 41-53, 1 Jan. 2015, doi: 10.1109/TPAMI.2014.2343973.

[2]

https://github.com/mims-harvard/scikit-fusion/tree/master

See also

DFMF

Example

>>> import numpy as np
>>> import pandas as pd
>>> from imml.impute import DFMFImputer
>>> Xs = [pd.DataFrame(np.random.default_rng(42).random((20, 10))) for i in range(3)]
>>> transformer = DFMFImputer(n_components = 5)
>>> labels = transformer.fit_transform(Xs)

fit_transform(Xs, y=None, **fit_params)

Fit to data, then impute them.

Parameters:

• Xs (list of array-likes objects) --

  • Xs length: n_mods

  • Xs[i] shape: (n_samples_i, n_features_i)

  A list of different mods.

• y (Ignored) -- Not used, present here for API consistency by convention.

• fit_params (Ignored) -- Not used, present here for API consistency by convention.

Returns:

transformed_X -- The transformed data with filled missing samples.

Return type:

array-likes objects of shape (n_samples, n_components)

set_fit_request(*, Xs: bool | None | str = '$UNCHANGED$') → DFMFImputer

Configure whether metadata should be requested to be passed to the fit method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:

Xs (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) -- Metadata routing for Xs parameter in fit.

Returns:

self -- The updated object.

Return type:

object

set_transform_request(*, Xs: bool | None | str = '$UNCHANGED$') → DFMFImputer

Configure whether metadata should be requested to be passed to the transform method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to transform.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:

Xs (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) -- Metadata routing for Xs parameter in transform.

Returns:

self -- The updated object.

Return type:

object

MOFA Imputer

class imml.impute.MOFAImputer(n_components: int = 10, impute: bool = True, data_options: dict = None, data_matrix: dict = None, model_options: dict = None, train_options: dict = None, stochastic_options: dict = None, covariates: dict = None, smooth_options: dict = None, random_state: int = None, verbose=False)

Bases: MOFA

Impute missing data in a dataset using the MOFA method. [3] [4] [5]

This class extends the MOFA class to provide functionality for filling in incomplete samples by addressing both block-wise and feature-wise missing data. As a subclass of MOFA, MOFAImputer inherits all input parameters and attributes from MOFA. Consequently, it uses the same fit method as MOFA for training the model.

References

[3]

Argelaguet R, Velten B, Arnol D, Dietrich S, Zenz T, Marioni JC, Buettner F, Huber W, Stegle O (2018). “Multi‐Omics Factor Analysis—a framework for unsupervised integration of multi‐omics data sets.” Molecular Systems Biology, 14. doi:10.15252/msb.20178124.

[4]

Argelaguet R, Arnol D, Bredikhin D, Deloro Y, Velten B, Marioni JC, Stegle O (2020). “MOFA+: a statistical framework for comprehensive integration of multi-modal single-cell data.” Genome Biology, 21. doi:10.1186/s13059-020-02015-1.

[5]

https://biofam.github.io/MOFA2/index.html

See also

MOFA

Example

>>> import numpy as np
>>> import pandas as pd
>>> from imml.impute import MOFAImputer
>>> Xs = [pd.DataFrame(np.random.default_rng(42).random((20, 10))) for i in range(3)]
>>> transformer = MOFAImputer(n_components = 5)
>>> labels = transformer.fit_transform(Xs)

fit_transform(Xs, y=None, **fit_params)

Fit to data, then impute them.

Parameters:

• Xs (list of array-likes objects) --

  • Xs length: n_mods

  • Xs[i] shape: (n_samples_i, n_features_i)

  A list of different mods.

• y (Ignored) -- Not used, present here for API consistency by convention.

• fit_params (Ignored) -- Not used, present here for API consistency by convention.

Returns:

transformed_X -- The transformed data with filled missing samples.

Return type:

array-likes objects of shape (n_samples, n_components)

set_fit_request(*, Xs: bool | None | str = '$UNCHANGED$') → MOFAImputer

Configure whether metadata should be requested to be passed to the fit method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:

Xs (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) -- Metadata routing for Xs parameter in fit.

Returns:

self -- The updated object.

Return type:

object

set_transform_request(*, Xs: bool | None | str = '$UNCHANGED$') → MOFAImputer

Configure whether metadata should be requested to be passed to the transform method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to transform.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:

Xs (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) -- Metadata routing for Xs parameter in transform.

Returns:

self -- The updated object.

Return type:

object

JNMF Imputer

class imml.impute.JNMFImputer(filling: bool = False, **kwargs)

Bases: JNMF

Impute missing data in a dataset using the JNMF method. [6] [7] [8] [9] [10] [11] [12] [13]

This class extends the JNMF class to provide functionality for filling in incomplete samples by addressing both block-wise and feature-wise missing data. As a subclass of JNMF, JNMFImputer inherits all input parameters and attributes from JNMF. Consequently, it uses the same fit method as JNMF training the model.

References

[6]

Tsuyuzaki et al., (2023). nnTensor: An R package for non-negative matrix/tensor decomposition. Journal of Open Source Software, 8(84), 5015, https://doi.org/10.21105/joss.05015

[7]

Liviu Badea, (2008) Extracting Gene Expression Profiles Common to Colon and Pancreatic Adenocarcinoma using Simultaneous nonnegative matrix factorization. Pacific Symposium on Biocomputing 13:279-290.

[8]

Shihua Zhang, et al. (2012) Discovery of multi-dimensional modules by integrative analysis of cancer genomic data. Nucleic Acids Research 40(19), 9379-9391.

[9]

Zi Yang, et al. (2016) A non-negative matrix factorization method for detecting modules in heterogeneous omics multi-modal data, Bioinformatics 32(1), 1-8.

[10]

Y. Kenan Yilmaz et al., (2010) Probabilistic Latent Tensor Factorization, International Conference on Latent Variable Analysis and Signal Separation 346-353.

[11]

N. Fujita et al., (2018) Biomarker discovery by integrated joint non-negative matrix factorization and pathway signature analyses, Scientific Report.

[12]

https://rdrr.io/cran/nnTensor/man/JNMF.html

[13]

https://github.com/rikenbit/nnTensor

See also

JNMF, JNMFFeatureSelector

Example

>>> import numpy as np
>>> import pandas as pd
>>> from imml.impute import JNMFImputer
>>> Xs = [pd.DataFrame(np.random.default_rng(42).random((20, 10))) for i in range(3)]
>>> transformer = JNMFImputer(n_components = 5)
>>> labels = transformer.fit_transform(Xs)

transform(Xs)

Impute unseen data.

Parameters:

Xs (list of array-likes objects) --

• Xs length: n_mods

• Xs[i] shape: (n_samples, n_features_i)

A list of different modalities.

Returns:

transformed_Xs -- The transformed data with filled missing samples.

Return type:

list of array-likes objects, shape (n_samples, n_features_i)

fit_transform(Xs, y=None, **fit_params)

Fit to data, then impute them.

Parameters:

• Xs (list of array-likes objects) --

  • Xs length: n_mods

  • Xs[i] shape: (n_samples_i, n_features_i)

  A list of different mods.

• y (Ignored) -- Not used, present here for API consistency by convention.

• fit_params (Ignored) -- Not used, present here for API consistency by convention.

Returns:

transformed_X -- The transformed data with filled missing samples.

Return type:

array-likes objects of shape (n_samples, n_components)

set_fit_request(*, Xs: bool | None | str = '$UNCHANGED$') → JNMFImputer

Configure whether metadata should be requested to be passed to the fit method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:

Xs (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) -- Metadata routing for Xs parameter in fit.

Returns:

self -- The updated object.

Return type:

object

set_transform_request(*, Xs: bool | None | str = '$UNCHANGED$') → JNMFImputer

Configure whether metadata should be requested to be passed to the transform method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to transform.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:

Xs (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) -- Metadata routing for Xs parameter in transform.

Returns:

self -- The updated object.

Return type:

object

Missing Modality Indicator

class imml.impute.MissingModIndicator

Bases: FunctionTransformer

Binary indicators for missing modalities. Apply FunctionTransformer (from Scikit-learn) with get_missing_mod_indicator as a function.

Note that this component typically should not be used in a vanilla Pipeline consisting of preprocessing and an estimator.

See also

ObservedModIndicator, missing_mod_indicator, observed_mod_indicator

Example

>>> import numpy as np
>>> import pandas as pd
>>> from imml.impute import MissingModIndicator
>>> from imml.ampute import Amputer
>>> Xs = [pd.DataFrame(np.random.default_rng(42).random((20, 10))) for i in range(3)]
>>> Xs = Amputer(p= 0.2, random_state=42).fit_transform(Xs)
>>> indicator = MissingModIndicator()
>>> missing_mod = indicator.fit_transform()(Xs)

class imml.impute.get_missing_mod_indicator(Xs: list, y=None)

Return a binary indicator for missing modalities.

Parameters:

• Xs (list of array-likes objects) --

  • Xs length: n_mods

  • Xs[i] shape: (n_samples, n_features)

  A list of different modalities.

• y (Ignored) -- Not used, present here for API consistency by convention.

Returns:

transformed_X -- Binary indicator for missing modalities.

Return type:

array-likes objects, shape (n_samples, n_mods)

See also

MissingModIndicator, ObservedModIndicator, observed_mod_indicator

Example

>>> import numpy as np
>>> import pandas as pd
>>> from imml.impute import get_missing_mod_indicator
>>> from imml.ampute import Amputer
>>> Xs = [pd.DataFrame(np.random.default_rng(42).random((20, 10))) for i in range(3)]
>>> Xs = Amputer(p= 0.2, random_state=42).fit_transform(Xs)
>>> missing_mod = get_missing_mod_indicator()(Xs)

Observed Modality Indicator

class imml.impute.ObservedModIndicator

Bases: FunctionTransformer

Binary indicator for observed modalities. Apply FunctionTransformer (from Scikit-learn) with get_observed_mod_indicator as a function.

Note that this component typically should not be used in a vanilla Pipeline consisting of preprocessing and an estimator.

See also

MissingModIndicator, missing_mod_indicator, observed_mod_indicator

Example

>>> import numpy as np
>>> import pandas as pd
>>> from imml.impute import ObservedModIndicator
>>> from imml.ampute import Amputer
>>> Xs = [pd.DataFrame(np.random.default_rng(42).random((20, 10))) for i in range(3)]
>>> Xs = Amputer(p= 0.2, random_state=42).fit_transform(Xs)
>>> indicator = ObservedModIndicator()
>>> observed_mod = indicator.fit_transform()(Xs)

class imml.impute.get_observed_mod_indicator(Xs: list, y=None)

Return a binary indicator for observed modalities.

Parameters:

• Xs (list of array-likes objects) --

  • Xs length: n_mods

  • Xs[i] shape: (n_samples, n_features)

  A list of different modalities.

• y (Ignored) -- Not used, present here for API consistency by convention.

Returns:

transformed_X -- Binary indicator for observed modalities.

Return type:

array-likes objects, shape (n_samples, n_mods)

See also

MissingModIndicator, ObservedModIndicator, missing_mod_indicator

Example

>>> import numpy as np
>>> import pandas as pd
>>> from imml.impute import get_observed_mod_indicator
>>> from imml.ampute import Amputer
>>> Xs = [pd.DataFrame(np.random.default_rng(42).random((20, 10))) for i in range(3)]
>>> Xs = Amputer(p= 0.2, random_state=42).fit_transform(Xs)
>>> observed_mod = get_observed_mod_indicator()(Xs)