Filters








6 Hits in 3.6 sec

ArviZ a unified library for exploratory analysis of Bayesian models in Python

Ravin Kumar, Colin Carroll, Ari Hartikainen, Osvaldo Martin
2019 Journal of Open Source Software  
In the words of Persi Diaconis (Diaconis, 2011) "Exploratory data analysis seeks to reveal structure, or simple descriptions in data. We look at numbers or graphs and try to find patterns.  ...  of both model assumptions and model predictions • Comparison of models, including model selection or model averaging • Preparation of the results for a particular audience Successfully performing such  ...  We also would like to extend thanks to all the ArviZ contributors, and the contributors of the libraries used to build ArviZ -particularly xarray, matplotlib, pandas, and numpy.  ... 
doi:10.21105/joss.01143 fatcat:rolp4jvj5rg35cixwfc3wrc3pq

Increasing Interpretability of Bayesian Probabilistic Programming Models Through Interactive Representations

Evdoxia Taka, Sebastian Stein, John H. Williamson
2020 Frontiers in Computer Science  
However, the results of Bayesian inference are challenging for users to interpret in tasks like decision-making under uncertainty or model refinement.  ...  We present a concrete implementation that translates probabilistic programs to interactive graphical representations and show illustrative examples for a variety of Bayesian probabilistic models.  ...  Kumar et al. (2019) created ArviZ, a unified Python tool for exploratory analysis, processing and visualization of the inference results of probabilistic programming models.  ... 
doi:10.3389/fcomp.2020.567344 fatcat:aakzmrx3fncc3mbmhowcfbeemy

Enhancing analysis of NASA data with the open source Python Xarray library

Scott Henderson, Deepak Cherian, Jessica Scheick, Anderson Banihirwe, Alan Snow
2021 figshare.com  
Analysis toolkits in diverse domains (e.g. Arviz for Bayesian inference, MetPy for meteorology) build on top of Xarray, benefiting from vast array and storage capabilities without duplicating effort.  ...  In practice, discoveries happen through exploratory data analysis and iterative hypothesis testing.  ... 
doi:10.6084/m9.figshare.16689265.v1 fatcat:f4d3jygdwvgbtgyd5q5czluz54

tascCODA: Bayesian Tree-Aggregated Analysis of Compositional Amplicon and Single-Cell Data

Johannes Ostner, Salomé Carcy, Christian L. Müller
2021 Frontiers in Genetics  
We posit that tascCODA1 constitutes a valuable addition to the growing statistical toolbox for generative modeling and analysis of compositional changes in microbial or cell population data.  ...  To this end, we introduce a Bayesian model for tree-aggregated amplicon and single-cell compositional data analysis (tascCODA) that seamlessly integrates hierarchical information and experimental covariate  ...  ., Hartikainen, A., and Martin, O. (2019). ArviZ a Unified Library for Exploratory Analysis of Bayesian Models in python. Joss 4, 1143. CrossRef Full Text | Google Scholar Labus, J.  ... 
doi:10.3389/fgene.2021.766405 pmid:34950190 pmcid:PMC8689185 fatcat:62eypasbkjfrpl3jubvsnazfzq

Efectos protectores de los alimentos andinos contra el daño producido por el alcohol a nivel del epitelio intestinal, una aproximación estadística

Eduardo Heilbron, Osvaldo Martìn, Emiliano Fumagalli
2020 Ciencia, Docencia y Tecnología  
La barrera epitelial intestinal es altamente regulada y permite el pasaje selectivo de nutrientes, mientras que es impermeable a sustancias nocivas.  ...  The American Journal of Gastroenterology, 94(1), 200-207. kumar, R.; Carroll, C.; Hartikainen, A., y Martin, O. (2019). ArviZ a unified library for exploratory analysis of Bayesian models in Python.  ...  Pandas: A Foundational Python Library for Data Analysis and Statistics. medina, A. L.; Arévalo, N. M.; Beltrán, S. D.; Chavarro, Y. L.; Herazo, E., y Campo-Arias, A. (2015).  ... 
doi:10.33255//3161/747 fatcat:golv2hbt4bfsji52g2ccmwgite

Evaluating the Implicit Midpoint Integrator for Riemannian Manifold Hamiltonian Monte Carlo [article]

James A. Brofos, Roy R. Lederman
2021 arXiv   pre-print
We discuss advantages and disadvantages of the implicit midpoint integrator for Hamiltonian Monte Carlo, its theoretical properties, and an empirical assessment of the critical attributes of such an integrator  ...  and better reversibility, arguably yielding a more accurate sampling procedure.  ...  Acknowledgments The authors would like to thank Marcus A. Brubaker for helpful discussions.  ... 
arXiv:2102.07139v2 fatcat:5udo6rcauvgmbjpm7u6gnbyjfa