imputegap.algorithms.stmvl package

The imputegap.algorithms.stmvl package contains various imputation algorithms used for handling missing values in time series data. This package supports multiple imputation techniques like CDRec, MRNN, IIM, and more.

Submodules

Modules

imputegap.algorithms.stmvl.native_stmvl(__py_matrix, __py_window, __py_gamma, __py_alpha, __verbose=True)[source]

Perform matrix imputation using the STMVL algorithm with native C++ support.

Parameters

__py_matrixnumpy.ndarray

The input matrix with missing values (NaNs).

__py_windowint

The window size for the temporal component in the STMVL algorithm.

__py_gammafloat

The smoothing parameter for temporal weight (0 < gamma < 1).

__py_alphafloat

The power for the spatial weight.

__verbosebool, optional

Whether to display the contamination information (default is False).

Returns

numpy.ndarray

The recovered matrix after imputation.

Notes

The STMVL algorithm leverages temporal and spatial relationships to recover missing values in a matrix. The native C++ implementation is invoked for better performance.

Example

>>> recov_data = stmvl(incomp_data=incomp_data, window_size=2, gamma=0.85, alpha=7)
>>> print(recov_data)

References

Yi, X., Zheng, Y., Zhang, J., & Li, T. ST-MVL: Filling Missing Values in Geo-Sensory Time Series Data. School of Information Science and Technology, Southwest Jiaotong University; Microsoft Research; Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences.

imputegap.algorithms.stmvl.stmvl(incomp_data, window_size, gamma, alpha, logs=True, verbose=True)[source]

ST-MVL algorithm for imputation of missing data

Parameters

incomp_datanumpy.ndarray

The input matrix with contamination (missing values represented as NaNs).

window_sizeint

window size for temporal component

gammafloat

smoothing parameter for temporal weight

alphafloat

power for spatial weight

logsbool, optional

Whether to log the execution time (default is True).

verbosebool, optional

Whether to display the contamination information (default is False).

lib_pathstr, optional

Custom path to the shared library file (default is None).

verbosebool, optional

Whether to display the contamination information (default is True).

Returns

numpy.ndarray

The imputed matrix with missing values recovered.

Example

>>> recov_data = stmvl(incomp_data=incomp_data, window_size=7, gamma=0.85, alpha=7, logs=True)
>>> print(recov_data)

References

Yi, X., Zheng, Y., Zhang, J., & Li, T. ST-MVL: Filling Missing Values in Geo-Sensory Time Series Data. School of Information Science and Technology, Southwest Jiaotong University; Microsoft Research; Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences.