Index of all namespaces
Anglican, a probabilistic programming system
Beaver data from R
crp and DPmem implementation for easy porting of legacy Anglican code
Fisher's Iris dataset
Sample wrappers for OBSERVE and SAMPLE treating ERP and XRP in the same way
Annealed Importance Sampling Options: :number-of-steps (1 by default) - number of interpolation steps to perform :start-exponent (0 by default) - initial exponent in the cooling schedule :end-exponent (1 by default) - final exponent in the cooling schedule
+exploration-factor+ +not-a-predict+ +prior-choice-reward+ ->predict add-choice-predicts add-pending-choice award combined-predicts initial-state initialize-last-predicts log-entry-probability map->predict next-state prev-state select-entry state-update trace-weights ucb update-choice-count update-reward update-rewards utility
Bayesian ascent Monte Carlo Options: :predict-candidates (false by default) - output all samples rather than just those with increasing log-weight
REPL and command line
Top-level forms for Anglican programs
Interacting particle Markov chain Monte Carlo Options: :number-of-particles (2 by default) - Number of particles per sweep :number-of-nodes (32 by default) - Number of nodes running SMC and CSMC. :number-of-csmc-nodes (nil by default) - Number of nodes running as CSMC. Must be between 1 and (- :number-of-nodes 1). Defaults to (/ :number-of-nodes 2) when not specified. :all-particles? (true by default) - Return all particles, instead of 1 particle per sweep on each node. Note that even when :all-particles? is false, particles are still weighted due to Rao- Blackwellization of the Gibbs updates for the CSMC indices. :pool (:builtin by default) - Threadpool argument for pmap operation over nodes. Defaults to creating a pool of containing (+ (ncpus) 2) threads. See com.climate.claypoole/pmap for further info.
Parallel Adaptive LMH Options: :number-of-threads (2 by default) - number of threads
Parallel Cascade Options: :number-of-threads (16 by default) - number of threads :number-of-particles (number-of-threads/2 by default) - number of initial particles
Particle Filter Options: :number-of-particles (1 by default) - number of particles per sweep
Particle Gibbs (Iterated Conditional SMC) Options: :number-of-particles (2 by default) - number of particles per sweep
Particle Independent Metropolis-Hastings Options: :number-of-particles (2 by default) - number of particles per sweep
Replaying choices through the program
Parallel Lighweight Metropolis-Hastings Options: :number-of-threads ((min 4 #cores) by default) - number of threads to use
Alternative proposal distributions used in Random-walk Metropolis-Hastings (rmh)
C ->categorical-crp-distribution ->categorical-distribution ->categorical-dp-distribution ->chi-squared-distribution ->CRP-process categorical categorical-crp categorical-dp cbrt ceil chi-squared cos cosh cov covariance cps CRP
M ->multivariate-t-distribution ->mvn-distribution ->mvn-niw-process map->bernoulli-distribution map->beta-distribution map->binomial-distribution map->categorical-crp-distribution map->categorical-distribution map->categorical-dp-distribution map->chi-squared-distribution map->CRP-process map->dirichlet-discrete-process map->dirichlet-distribution map->discrete-distribution map->DP-process map->exponential-distribution map->flip-distribution map->gamma-distribution map->GP-process map->multivariate-t-distribution map->mvn-distribution map->mvn-niw-process map->normal-distribution map->poisson-distribution map->uniform-continuous-distribution map->uniform-discrete-distribution map->wishart-distribution mean multivariate-distribution multivariate-t mvn mvn-niw mvn-niw-posterior mvn-niw-predictive
MAP estimation via Simulated Annealing Options: :cooling-rate (0.99 by default) - cooling rate, should less than 1 :cooling-schedule (:exponential by default) - cooling schedule, :exponential or :lundy-mees :predict-candidates (false by default) - output all samples rather than those with increasing log-weight
Sequential Monte Carlo Options: :number-of-particles (1 by default) - number of particles per sweep
Statistics related functions
CPS transformation of Anglican program
C continue cps-of-and cps-of-application cps-of-apply cps-of-case cps-of-cond cps-of-do cps-of-expression cps-of-hash-map cps-of-if cps-of-let cps-of-loop cps-of-observe cps-of-opaque cps-of-or cps-of-predict cps-of-recur cps-of-retrieve cps-of-sample cps-of-set cps-of-store cps-of-vector cps-of-when
Translation of Anglican Scheme to Anglican Clojure, deprecated
Anglican is a probabilistic programming system implemented in Clojure, both the programming environment and the language. Introduction to anglican explains how to write and run programs in anglican.
Everyone is welcome to write programs which call inference, as well as to extend anglican with new algorithms, propose changes, and fix bugs. Consult the developer guide for details.
The code map provides a quick guide around the source code layout, as well as contains a list of implemented inference algorithms.
Copyright © 2014, 2015, 2016 Wood group
This file is part of Anglican, a probabilistic programming system.
Anglican is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
Anglican is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.