Skip to content

Releases: scikit-learn-contrib/qolmat

Version 0.1.8

13 Jun 21:29
f94aafd
Compare
Choose a tag to compare
Merge pull request #151 from scikit-learn-contrib/dev

Dev

Version 0.1.7

13 Jun 11:34
Compare
Choose a tag to compare
  • Little's test implemented in a new hole_characterization module
  • Documentation now includes an analysis section with a tutorial
  • Hole generators now provide reproducible outputs

Version 0.1.6

17 Apr 15:01
beb6c2a
Compare
Choose a tag to compare
  • Documentation patched

Version 0.1.5

17 Apr 13:51
41884b2
Compare
Choose a tag to compare
  • CICD now relies on Node.js 20
  • New tests for comparator.py and data.py

Version 0.1.4

15 Apr 15:19
Compare
Choose a tag to compare
  • ImputerMean, ImputerMedian and ImputerMode have been merged into ImputerSimple
  • File preprocessing.py added with classes new MixteHGBM, BinTransformer, OneHotEncoderProjector and WrapperTransformer providing tools to manage mixed types data
  • Tutorial plot_tuto_categorical showcasing mixed type imputation
  • Titanic dataset added
  • accuracy metric implemented
  • metrics.py rationalized, and split with algebra.py

Version 0.1.3

08 Mar 13:15
Compare
Choose a tag to compare

0.1.3 (2024-03-07)

  • RPCA algorithms now start with a normalizing scaler
  • The EM algorithms now include a gradient projection step to be more robust to colinearity
  • The EM algorithm based on the Gaussian model is now initialized using a robust estimation of the covariance matrix
  • A bug in the EM algorithm has been patched: the normalizing matrix gamma was creating a sampling biais
  • Speed up of the EM algorithm likelihood maximization, using the conjugate gradient method
  • The ImputeRegressor class now handles the nans by row by default
  • The metric frechet was not correctly called and has been patched
  • The EM algorithm with VAR(p) now fills initial holes in order to avoid exponential explosions

Version 0.1.2

28 Feb 13:05
Compare
Choose a tag to compare
  • RPCA Noisy now has separate fit and transform methods, allowing to impute efficiently new data without retraining
  • The class ImputerRPCA has been splitted between a class ImputerRpcaNoisy, which can fit then transform, and a class ImputerRpcaPcp which can only fit_transform
  • The class SoftImpute has been recoded to better fit the architecture, and is more tested
  • The class RPCANoisy now relies on sparse matrices for H, speeding it up for large instances

Version 0.1.1

06 Nov 17:43
ffdf737
Compare
Choose a tag to compare
  • Hotfix reference to tensorflow in the documentation, when it should be pytorch
  • Metrics KL forest has been removed from package
  • EM imputer made more robust to colinearity, and transform bug patched
  • CICD made faster with mamba and a quick test setting

Version 0.1.0

12 Oct 09:57
Compare
Choose a tag to compare
  • VAR(p) EM sampler implemented, founding on a VAR(p) modelization such as the one described in Lütkepohl (2005) New Introduction to Multiple Time Series Analysis
  • EM and RPCA matrices transposed in the low-level impelmentation, however the API remains unchanged
  • Sparse matrices introduced in the RPCA implementation so as to speed up the execution
  • Implementation of SoftImpute, which provides a fast but less robust alterantive to RPCA
  • Implementation of TabDDPM and TsDDPM, which are diffusion-based models for tabular data and time-series data, based on Denoising Diffusion Probabilistic Models. Their implementations follow the work of Tashiro et al., (2021) and Kotelnikov et al., (2023).
  • ImputerDiffusion is an imputer-wrapper of these two models TabDDPM and TsDDPM.
  • Docstrings and tests improved for the EM sampler
  • Fix ImputerPytorch
  • Update Benchmark Deep Learning

Version 0.0.15

03 Aug 14:13
8cfbfd3
Compare
Choose a tag to compare
  • Hyperparameters are now optimized in hyperparameters.py, with the maintained module hyperopt
  • The Imputer classes do not possess a dictionary attribute anymore, and all list attributes have
    been changed into tuple attributes so that all are not immutable
  • All the tests from scikit-learn's check_estimator now pass for the class Imputer
  • Fix MLP imputer, created a builder for MLP imputer
  • Switch tensorflow by pytorch. Change Test, environment, benchmark and imputers for pytorch
  • Add new datasets
  • Added dcor metrics with a pattern-wise computation on data with missing values