usptact

//
// Game of Ur
//

var roll = function() {
	return flip() + flip() + flip() + flip();
}

var game = function() {
	var n = sample(RandomInteger({n: 20}));

usptact

var person = function() {
        var is_female = sample(Bernoulli({p: 0.51}))
        if (is_female == false) {
                return {hgt: sample(Gaussian({mu: 178.0, sigma: 7.7})), is_female: is_female}
        } else {
                return {hgt: sample(Gaussian({mu: 163.0, sigma: 7.3})), is_female: is_female}
        }
}

var is_taller = function(h1, h2) {

usptact

// 4 dice: 4, 6, 8 and 12 sides
// each dice is perfect
var dice = [
	Categorical(
		{
			ps: [1/4, 1/4, 1/4, 1/4],
			vs: [0, 1, 2, 3]
		}
	),
	Categorical(

usptact

// Remove each element in array ys from array xs
var remove = function(xs, ys) {
  return _.without.apply(null, [xs].concat(ys));
};

var doors = [1, 2, 3];

// Monty chooses a door that is neither Alice's door
// nor the prize door
var monty = function(aliceDoor, prizeDoor) {

usptact

"""
Demo script to compute rank distributions given pairwise preference probabilities.

Data:
- pairwise preference probabilities (not including self)

Output:
- distribution over ranks

Michael Taylor, John Guiver, Stephen Robertson and Tom Minka, "SoftRank: Optimising Non-Smooth Rank Metrics", WSDM 2008.

usptact

# (C) Mathieu Blondel, November 2013
# License: BSD 3 clause

import numpy as np


def ranking_precision_score(y_true, y_score, k=10):
    """Precision at rank k

    Parameters

usptact

"""
Class to sort intervals.

A collection of intervals is sorted:
(1) in increasing order by start index
(2) in increasing order by end index

Sample data:
data = [(1, 2), (0, 1), (0, 3), (5, 9), (1, 4), (1, 2), (2, 5), (5, 8), (3, 6)]
"""

usptact

"""
Class to pack or tile segments.

Example data:
data = [
(0, 1), (0, 2), (0, 3), (1, 3), (2, 4), (0, 5)
]

NB: sorted segments are expected (cf. IntervalSort)!
"""

usptact

usptact

import seaborn as sns
import pymc3 as pm
import numpy as np
from scipy.stats import norm
import pandas as pd

import theano.tensor as T
from theano.compile.ops import as_op

# Generate True data