4 examples of 'np random binomial' in Python

144
def binomial(trials, p, shape=[]):
145
    """binomial(trials, p) or binomial(trials, p, [n, m, ...]) returns array of binomially distributed random integers.
146
147
           trials is the number of trials in the binomial distribution.
148
           p is the probability of an event in each trial of the binomial distribution."""
149
    if shape == []:
150
        shape = None
151
    return mt.binomial(trials, p, shape)

76
def test_binomial(self):
77
    assert self.rs.binomial(10, .5) >= 0
78
    assert self.rs.binomial(1000, .5) >= 0

123
def binomial(big, small):
124
    '''
125
    Get the binomial coefficient (big small).
126
    
127
    This is used in combinatorical calculations. More information:
128
    http://en.wikipedia.org/wiki/Binomial_coefficient
129
    '''
130
    if big == small:
131
        return 1
132
    if big < small:
133
        return 0
134
    else:
135
        return (math.factorial(big) // math.factorial(big - small)
136
                                                      // math.factorial(small))

658
def binomial_like(self, x, n, p, name='binomial', prior=False):
659
	"""Binomial log-likelihood"""
660
	
661
	if not shape(n) == shape(p): raise ParameterError, 'Parameters must have same dimensions'
662
	
663
	if ndim(n) > 1:
664
		
665
		return sum([self.binomial_like(y, _n, _p, name, prior) for y, _n, _p in zip(x, n, p)])
666
	
667
	else:
668
		
669
		# Ensure valid values of parameters
670
		self.constrain(p, 0, 1)
671
		self.constrain(n, lower=x)
672
		self.constrain(x, 0)
673
		
674
		# Enforce array type
675
		x = atleast_1d(x)
676
		p = resize(p, shape(x))
677
		n = resize(n, shape(x))
678
		
679
		# Goodness-of-fit
680
		if self._gof and not prior:
681
			
682
			try:
683
				self._like_names.append(name)
684
			except AttributeError:
685
				pass
686
			
687
			expval = p * n
688
			
689
			# Simulated values
690
			y = array([rbinomial(_n, _p) for _n, _p in zip(n, p)])
691
			
692
			# Generate GOF points
693
			gof_points = sum(transpose([self.loss(x, expval), self.loss(y, expval)]))
694
			
695
			self._gof_loss.append(gof_points)
696
		
697
		return sum([fbinomial(xx, nn, pp) for xx, nn, pp in zip(x, n, p)])