Mixedpower uses pilotdata and a linear mixed model fitted with lme4 to simulate new data sets. Power is computed separate for every effect in the model output as the relation of significant simulations to all simulations. More conservative simulations as a protection against a bias in the pilotdata are available as well as methods for plotting the results.

This package provides a library for Probabilistic Graphical Models. It can be used for learning (Structure and Parameter), inference (Probabilistic and Causal), and simulations in Bayesian Networks.

Nautilus is an pure-Python package for Bayesian posterior and evidence estimation. It utilizes importance sampling and efficient space exploration using neural networks. Compared to traditional MCMC and Nested Sampling codes, it often needs fewer likelihood calls and produces much larger posterior samples. Additionally, nautilus is highly accurate and produces Bayesian evidence estimates with percent precision. It is widely used in many areas of astrophysical research.

Similarity Weighted Nonnegative Embedding (SWNE) is a method for visualizing high dimensional datasets. SWNE uses Nonnegative Matrix Factorization to decompose datasets into latent factors, projects those factors onto 2 dimensions, and embeds samples and key features in 2 dimensions relative to the factors. SWNE can capture both the local and global dataset structure, and allows relevant features to be embedded directly onto the visualization, facilitating interpretation of the data.

MAPIE allows you to easily estimate prediction intervals (or prediction sets) using your favourite scikit-learn-compatible model for single-output regression or multi-class classification settings.

Prediction intervals output by MAPIE encompass both aleatoric and epistemic uncertainties and are backed by strong theoretical guarantees thanks to conformal prediction methods intervals.

SPAMS (SPArse Modeling Software) is an optimization toolbox for solving various sparse estimation problems. It includes tools for the following problems:

1. Dictionary learning and matrix factorization (NMF, sparse principle component analysis (PCA), ...)

2. Solving sparse decomposition problems with LARS, coordinate descent, OMP, SOMP, proximal methods

3. Solving structured sparse decomposition problems (l1/l2, l1/linf, sparse group lasso, tree-structured regularization, structured sparsity with overlapping groups,...).

This is a package for Non-Negative Linear Models (NNLM). It implements fast sequential coordinate descent algorithms for non-negative linear regression and non-negative matrix factorization (NMF). It supports mean square error and Kullback-Leibler divergence loss. Many other features are also implemented, including missing value imputation, domain knowledge integration, designable W and H matrices and multiple forms of regularizations.

Radian is an alternative console for the R program with multiline editing and rich syntax highlight. One would consider Radian as a IPython clone for R, though its design is more aligned to Julia.

libxls is a C library to read .xls spreadsheet files in the binary OLE BIFF8 format as created by Excel 97 and later versions. It cannot write them.

This package also provides xls2csv to export Excel files to CSV.

This package is a model building aid for nonlinear mixed-effects (population) model analysis using NONMEM, facilitating data set checkout, exploration and visualization, model diagnostics, candidate covariate identification and model comparison. The methods are described in Keizer et al. (2013) <doi:10.1038/psp.2013.24>, and Jonsson et al. (1999) <doi:10.1016/s0169-2607(98)00067-4>.

Did you ever wish you could make scatter plots with cat shaped points? Now you can!

This package provides a collection of (mostly simple) functions for generating and manipulating colors in R.

This package implements a functionality to optionally compute statistical test and add statistical annotations on plots generated with seaborn.

OpenTURNS is a scientific C++ and Python library including an internal data model and algorithms dedicated to the treatment of uncertainties. The main goal of this library is giving to specific applications all the functionalities needed to treat uncertainties in studies.

rpy2 is a redesign and rewrite of rpy. It is providing a low-level interface to R from Python, a proposed high-level interface, including wrappers to graphical libraries, as well as R-like structures and functions.

This package provides an implementation of Nested Sampling algorithms for evaluating Bayesian evidence.

This package enables survival analysis in Python, including Kaplan Meier, Nelson Aalen and regression.

dcor is distance correlation and energy statistics in Python.

E-statistics are functions of distances between statistical observations in metric spaces. Distance covariance and distance correlation are dependency measures between random vectors introduced in [SRB07] with a simple E-statistic estimator.

This package offers functions for calculating several E-statistics such as:

• estimator of the energy distance [SR13]

• biased and unbiased estimators of distance covariance and distance correlation [SRB07]

• estimators of the partial distance covariance and partial distance covariance [SR14]

rchitect provides access to R functionality from Python. Its main use is as the driver for radian, the R console.

GetDist is a Python package for analysing Monte Carlo samples, including correlated samples from Markov Chain Monte Carlo (MCMC).

This is a package to provide infrastructure for managing package parameters. Parameters are easy to get in relevant functions within a package, and error is thrown if a parameter is missing. Developers are able to register parameters and set their default value in a config file that is part of the package in YAML format, and users are able to override parameters using their own YAML. Users get an exception when trying to override a parameter that was not registered, and can load multiple parameters to the current environment.

This package performs KDE operations on multidimensional data to calculate estimated PDFs (probability distribution functions), and resample new data from those PDFs.

This package provides a resampling-based inference based on data resampling and permutation.

Features:

• Bootstrap resampling: ordinary or balanced with optional stratification

• Extended bootstrap resampling: also varies sample size

• Parametric resampling: Gaussian, Poisson, gamma, etc.)

• Jackknife estimates of bias and variance of any estimator

• Compute bootstrap confidence intervals (percentile or BCa) for any estimator

• Permutation-based variants of traditional statistical tests (USP test of independence and others)

• Tools for working with empirical distributions (CDF, quantile, etc.)

ArviZ is a Python package for exploratory analysis of Bayesian models. It includes functions for posterior analysis, data storage, model checking, comparison and diagnostics.