diff --git a/pymc3/distributions/discrete.py b/pymc3/distributions/discrete.py index d872bc0195..8121cadc78 100644 --- a/pymc3/distributions/discrete.py +++ b/pymc3/distributions/discrete.py @@ -117,7 +117,9 @@ def logp(self, value): p = self.p return bound( switch(value, log(p), log(1 - p)), - 0 <= p, p <= 1) + 0 <= p, p <= 1, + eq(value,0) | eq(value,1) + ) class Poisson(Discrete): diff --git a/pymc3/distributions/distribution.py b/pymc3/distributions/distribution.py index ea9b9aaecb..5774e588f9 100644 --- a/pymc3/distributions/distribution.py +++ b/pymc3/distributions/distribution.py @@ -2,7 +2,7 @@ import numpy as np from ..model import Model -__all__ = ['DensityDist', 'Distribution', 'Continuous', 'Discrete'] +__all__ = ['DensityDist', 'Distribution', 'Continuous', 'Discrete', 'NoDistribution', 'TensorType'] class Distribution(object): @@ -64,6 +64,10 @@ def get_test_val(self, val, defaults): def TensorType(dtype, shape): return t.TensorType(str(dtype), np.atleast_1d(shape) == 1) +class NoDistribution(Distribution): + def logp(self, x): + return 0 + class Discrete(Distribution): """Base class for discrete distributions""" def __init__(self, shape=(), dtype='int64', *args, **kwargs): diff --git a/pymc3/distributions/special.py b/pymc3/distributions/special.py index aa1fb27008..cb9d3bfc9a 100644 --- a/pymc3/distributions/special.py +++ b/pymc3/distributions/special.py @@ -30,10 +30,7 @@ def grad(self, inp, grads): def c_code(self, node, name, inp, out, sub): x, = inp z, = out - if node.inputs[0].type in [scalar.float32, scalar.float64]: - return """%(z)s = - lgamma(%(x)s);""" % locals() - raise NotImplementedError('only floatingpoint is implemented') + return """%(z)s = lgamma(%(x)s);""" % locals() def __eq__(self, other): return type(self) == type(other) @@ -119,10 +116,10 @@ def c_support_code(self): def c_code(self, node, name, inp, out, sub): x, = inp z, = out - if node.inputs[0].type in [scalar.float32, scalar.float64]: - return """%(z)s = - _psi(%(x)s);""" % locals() - raise NotImplementedError('only floatingpoint is implemented') + if node.inputs[0].type in scalar.complex_types: + raise NotImplementedError('type not supported', node.inputs[0].type) + + return """%(z)s = _psi(%(x)s);""" % locals() def __eq__(self, other): return type(self) == type(other) diff --git a/pymc3/examples/arbitrary_stochastic.py b/pymc3/examples/arbitrary_stochastic.py index 0f9e1df105..0b96640645 100644 --- a/pymc3/examples/arbitrary_stochastic.py +++ b/pymc3/examples/arbitrary_stochastic.py @@ -9,7 +9,7 @@ def logp(failure, value): return sum(failure * log(lam) - lam * value) - x = DensityDist('x', logp, observed=(failure, value)) + x = DensityDist('x', logp, observed={'failure':failure, 'value':value}) def run (n=3000): diff --git a/pymc3/examples/data/test_scores.csv b/pymc3/examples/data/test_scores.csv new file mode 100644 index 0000000000..c3f1da1856 --- /dev/null +++ b/pymc3/examples/data/test_scores.csv @@ -0,0 +1,17 @@ +score,male,sib,synd_or_disab,age_test,mother_hs,ident_by_3 +100,0,1,0,5,1,0 +35,0,2,0,9,0,1 +24,1,NaN,1,7,1,0 +75,1,0,0,2,1,1 +24,1,NaN,1,7,1,0 +81,1,0,0,2,1,1 +24,1,3,1,7,1,0 +64,1,0,NaN,11,1,0 +18,1,1,1,9,1,0 +24,1,1,0,8,0,0 +24,1,1,1,7,1,0 +76,0,2,0,2,NaN,0 +36,0,1,0,2,1,1 +16,1,1,0,2,0,0 +95,0,0,0,2,NaN,1 +65,1,2,0,2,0,1 diff --git a/pymc3/examples/disaster_model.ipynb b/pymc3/examples/disaster_model.ipynb new file mode 100644 index 0000000000..8687f5ac24 --- /dev/null +++ b/pymc3/examples/disaster_model.ipynb @@ -0,0 +1,199 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " [-----------------100%-----------------] 500 of 500 complete in 1.8 sec" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + ":0: FutureWarning: IPython widgets are experimental and may change in the future.\n" + ] + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1kAAALJCAYAAABGEOnEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "%matplotlib inline\n", + "\"\"\"\n", + "A model for coal mining disasters data with a changepoint\n", + "\n", + "switchpoint ~ U(0, 110)\n", + "early_mean ~ Exp(1.)\n", + "late_mean ~ Exp(1.)\n", + "disasters[t] ~ Po(early_mean if t <= switchpoint, late_mean otherwise)\n", + "\n", + "\"\"\"\n", + "\n", + "from pymc3 import *\n", + "\n", + "import theano.tensor as t\n", + "from numpy import arange, array, ones, concatenate\n", + "from numpy.random import randint\n", + "from numpy.ma import masked_values\n", + "\n", + "__all__ = ['disasters_data', 'switchpoint', 'early_mean', 'late_mean', 'rate',\n", + " 'disasters']\n", + "\n", + "# Time series of recorded coal mining disasters in the UK from 1851 to 1962\n", + "disasters_data = masked_values([4, 5, 4, 0, 1, 4, 3, 4, 0, -1, 3, 3, 4, 0, 2, 6,\n", + " 3, 3, 5, 4, 5, 3, 1, 4, 4, 1, 5, 5, 3, 4, 2, 5,\n", + " 2, 2, 3, 4, 2, 1, 3, 2, 2, 1, 1, 1, 1, 3, 0, 0,\n", + " 1, 0, 1, 1, 0, 0, 3, 1, 0, 3, 2, 2, 0, 1, 1, 1,\n", + " 0, 1, 0, 1, 0, 0, 0, -1, 1, 0, 0, 0, 1, 1, 0, 2,\n", + " 3, 3, 1, 1, 2, 1, 1, 1, 1, 2, 4, 2, 0, 0, 1, 4,\n", + " 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1], value=-1)\n", + "years = len(disasters_data)\n", + "\n", + "with Model() as model:\n", + "\n", + " # Prior for distribution of switchpoint location\n", + " switchpoint = DiscreteUniform('switchpoint', lower=0, upper=years)\n", + " # Priors for pre- and post-switch mean number of disasters\n", + " early_mean = Exponential('early_mean', lam=1.)\n", + " late_mean = Exponential('late_mean', lam=1.)\n", + "\n", + " # Allocate appropriate Poisson rates to years before and after current\n", + " # switchpoint location\n", + " idx = arange(years)\n", + " rate = switch(switchpoint >= idx, early_mean, late_mean)\n", + "\n", + " # Data likelihood\n", + " disasters = Poisson('disasters', rate, observed=disasters_data)\n", + " b = theano.tensor.set_subtensor(rate[5], 12) \n", + " a = Deterministic('a', disasters.sum())\n", + "\n", + "n =500\n", + "with model:\n", + "\n", + " # Initial values for stochastic nodes\n", + " start = {'early_mean': 2., 'late_mean': 3.}\n", + "\n", + " # Use slice sampler for means\n", + " step1 = Slice([early_mean, late_mean])\n", + " # Use Metropolis for switchpoint, since it accomodates discrete variables\n", + " \n", + " step2 = Metropolis([switchpoint] +model.missing_values)\n", + "\n", + " tr = sample(n, tune=500, start=start, step=[step1, step2])\n", + " traceplot(tr)\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "dtype('int32')" + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "model.test_point['disasters_missing'].dtype" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "'int64'" + ] + }, + "execution_count": 3, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "a.dtype" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "dtype('int32')" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "disasters_data.dtype" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.4.3" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/pymc3/examples/disaster_model_missing.py b/pymc3/examples/disaster_model_missing.py new file mode 100644 index 0000000000..fc4fde52f3 --- /dev/null +++ b/pymc3/examples/disaster_model_missing.py @@ -0,0 +1,68 @@ +""" +A model for coal mining disasters data with a changepoint + +switchpoint ~ U(0, 110) +early_mean ~ Exp(1.) +late_mean ~ Exp(1.) +disasters[t] ~ Po(early_mean if t <= switchpoint, late_mean otherwise) + +""" + +from pymc3 import * + +import theano.tensor as t +from numpy import arange, array, ones, concatenate +from numpy.random import randint +from numpy.ma import masked_values + +__all__ = ['disasters_data', 'switchpoint', 'early_mean', 'late_mean', 'rate', + 'disasters'] + +# Time series of recorded coal mining disasters in the UK from 1851 to 1962 +disasters_data = array([4, 5, 4, 0, 1, 4, 3, 4, 0, 6, 3, 3, 4, 0, 2, 6, + 3, 3, 5, 4, 5, 3, 1, -1, 4, 1, 5, 5, 3, 4, 2, 5, + 2, 2, 3, 4, 2, 1, 3, 2, 2, 1, 1, 1, 1, 3, 0, 0, + 1, 0, 1, 1, 0, 0, 3, 1, 0, 3, 2, 2, 0, 1, 1, 1, + 0, 1, 0, 1, -1, 0, 0, 2, 1, 0, 0, 0, 1, 1, 0, 2, + 3, 3, 1, 1, 2, 1, 1, 1, 1, 2, 4, 2, 0, 0, 1, 4, + 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1]) +years = len(disasters_data) + +masked_values = masked_values(disasters_data, value=-1) + +with Model() as model: + + # Prior for distribution of switchpoint location + switchpoint = DiscreteUniform('switchpoint', lower=0, upper=years) + # Priors for pre- and post-switch mean number of disasters + early_mean = Exponential('early_mean', lam=1.) + late_mean = Exponential('late_mean', lam=1.) + + # Allocate appropriate Poisson rates to years before and after current + # switchpoint location + idx = arange(years) + rate = switch(switchpoint >= idx, early_mean, late_mean) + + # Data likelihood + disasters = Poisson('disasters', rate, observed=masked_values) + + +def run(n=1000): + if n == "short": + n = 500 + with model: + + # Initial values for stochastic nodes + start = {'early_mean': 2., 'late_mean': 3.} + + # Use slice sampler for means + step1 = Slice([early_mean, late_mean]) + # Use Metropolis for switchpoint, since it accomodates discrete variables + step2 = Metropolis([switchpoint, disasters.missing_values ]) + + tr = sample(n, tune=500, start=start, step=[step1, step2]) + + summary(tr, vars=['disasters_missing']) + +if __name__ == '__main__': + run() diff --git a/pymc3/examples/lasso_missing.py b/pymc3/examples/lasso_missing.py new file mode 100644 index 0000000000..b707bff214 --- /dev/null +++ b/pymc3/examples/lasso_missing.py @@ -0,0 +1,53 @@ +from pymc3 import * +import numpy as np +import pandas as pd +from numpy.ma import masked_values + +# Import data, filling missing values with sentinels (-999) +test_scores = pd.read_csv(get_data_file('pymc3.examples', 'data/test_scores.csv')).fillna(-999) + +# Extract variables: test score, gender, number of siblings, previous disability, age, +# mother with HS education or better, hearing loss identified by 3 months of age +(score, male, siblings, disability, + age, mother_hs, early_ident) = test_scores[['score', 'male', 'sib', + 'synd_or_disab', 'age_test', + 'mother_hs', 'ident_by_3']].astype(float).values.T + +with Model() as model: + + # Impute missing values + sib_mean = Exponential('sib_mean', 1) + siblings_imp = Poisson('siblings_imp', sib_mean, observed=masked_values(siblings, value=-999)) + + p_disab = Beta('p_disab', 1, 1) + disability_imp = Bernoulli('disability_imp', p_disab, observed=masked_values(disability, value=-999)) + + p_mother = Beta('p_mother', 1, 1) + mother_imp = Bernoulli('mother_imp', p_mother, observed=masked_values(mother_hs, value=-999)) + + s = HalfCauchy('s', 5, testval=5) + beta = Laplace('beta', 0, 100, shape=7, testval=.1) + + expected_score = (beta[0] + beta[1]*male + beta[2]*siblings_imp + beta[3]*disability_imp + + beta[4]*age + beta[5]*mother_imp + beta[6]*early_ident) + + observed_score = Normal('observed_score', expected_score, s ** -2, observed=score) + + +with model: + start = find_MAP() + step1 = NUTS([beta, s, p_disab, p_mother, sib_mean], scaling=start) + + step2 = Metropolis([mother_imp.missing_values, + disability_imp.missing_values, + siblings_imp.missing_values]) + +def run(n=5000): + if n == 'short': + n = 100 + with model: + trace = sample(n, [step1, step2], start) + + +if __name__ == '__main__': + run() diff --git a/pymc3/model.py b/pymc3/model.py index 63295dfd14..05b1adb3c5 100644 --- a/pymc3/model.py +++ b/pymc3/model.py @@ -90,6 +90,7 @@ def __init__(self): self.observed_RVs = [] self.deterministics = [] self.potentials = [] + self.missing_values = [] self.model = self @property @@ -101,7 +102,7 @@ def logpt(self): @property def vars(self): - """List of unobserved random variables the model is defined in terms of (which excludes deterministics).""" + """List of unobserved random variables used as inputs to the model (which excludes deterministics).""" return self.free_RVs @property @@ -112,7 +113,7 @@ def basic_RVs(self): @property def unobserved_RVs(self): """List of all random variable, including deterministic ones.""" - return self.free_RVs + self.deterministics + return self.vars + self.deterministics @property @@ -142,9 +143,20 @@ def Var(self, name, dist, data=None): if data is None: var = FreeRV(name=name, distribution=dist, model=self) self.free_RVs.append(var) - else: + + elif isinstance(data, dict): + var = MultiObservedRV(name=name, data=data, distribution=dist, model=self) + self.observed_RVs.append(var) + if var.missing_values: + self.free_RVs += var.missing_values + self.missing_values += var.missing_values + else: var = ObservedRV(name=name, data=data, distribution=dist, model=self) self.observed_RVs.append(var) + if var.missing_values: + self.free_RVs.append(var.missing_values) + self.missing_values.append(var.missing_values) + self.add_random_variable(var) return var @@ -322,8 +334,76 @@ def __init__(self, type=None, owner=None, index=None, name=None, distribution=No self.logp_elemwiset = distribution.logp(self) self.model = model -class ObservedRV(Factor): - """Observed random variable that a model is specified in terms of.""" +def pandas_to_array(data): + if hasattr(data, 'values'): #pandas + if data.isnull().any().any(): #missing values + return np.ma.MaskedArray(data.values, data.isnull().values) + else: + return data.values + elif hasattr(data, 'mask'): + return data + else: + return np.asarray(data) + + +def as_tensor(data, name,model, dtype): + data = pandas_to_array(data).astype(dtype) + + if hasattr(data, 'mask'): + from .distributions import NoDistribution + fakedist = NoDistribution.dist(shape=data.mask.sum(), dtype=dtype, testval=data.mean().astype(dtype)) + missing_values = FreeRV(name=name + '_missing', distribution=fakedist, model=model) + + constant = t.as_tensor_variable(data.filled()) + + dataTensor = theano.tensor.set_subtensor(constant[data.mask.nonzero()], missing_values) + dataTensor.missing_values = missing_values + return dataTensor + else: + data = t.as_tensor_variable(data, name=name) + data.missing_values = None + return data + +class ObservedRV(Factor, TensorVariable): + """Observed random variable that a model is specified in terms of. + Potentially partially observed. + """ + def __init__(self, type=None, owner=None, index=None, name=None, data=None, distribution=None, model=None): + """ + Parameters + ---------- + + type : theano type (optional) + owner : theano owner (optional) + + name : str + distribution : Distribution + model : Model + """ + from .distributions import TensorType + if type is None: + data = pandas_to_array(data) + type = TensorType(distribution.dtype, data.shape) + + super(TensorVariable, self).__init__(type, None, None, name) + + if distribution is not None: + data = as_tensor(data, name,model,distribution.dtype) + self.missing_values = data.missing_values + + self.logp_elemwiset = distribution.logp(data) + self.model = model + self.distribution = distribution + + #make this RV a view on the combined missing/nonmissing array + theano.gof.Apply(theano.compile.view_op, inputs=[data], outputs=[self]) + + self.tag.test_value = theano.compile.view_op(data).tag.test_value + +class MultiObservedRV(Factor): + """Observed random variable that a model is specified in terms of. + Potentially partially observed. + """ def __init__(self, name, data, distribution, model): """ Parameters @@ -337,17 +417,11 @@ def __init__(self, name, data, distribution, model): model : Model """ self.name = name - data = getattr(data, 'values', data) #handle pandas - args = as_iterargs(data) - if len(args) > 1: - params = getargspec(distribution.logp).args - args = [t.constant(d, name=name + "_" + param) - for d,param in zip(args,params) ] - else: - args = [t.constant(args[0], name=name)] + self.data = { name : as_tensor(data, name, model, distribution.dtype) for name, data in data.items()} - self.logp_elemwiset = distribution.logp(*args) + self.missing_values = [ data.missing_values for data in self.data.values() if data.missing_values is not None] + self.logp_elemwiset = distribution.logp(**self.data) self.model = model self.distribution = distribution @@ -385,8 +459,6 @@ def Potential(name, var, model=None): def as_iterargs(data): if isinstance(data, tuple): return data - if hasattr(data, 'columns'): # data frames - return [np.asarray(data[c]) for c in data.columns] else: return [data] diff --git a/pymc3/tests/test_missing.py b/pymc3/tests/test_missing.py new file mode 100644 index 0000000000..dfbd27088b --- /dev/null +++ b/pymc3/tests/test_missing.py @@ -0,0 +1,27 @@ +from pymc3 import Model, Normal, Metropolis, MvNormal +from numpy import ma +import numpy +import pandas as pd + +def test_missing(): + data = ma.masked_values([1,2,-1,4,-1], value=-1) + with Model() as model: + x = Normal('x', 1, 1) + y = Normal('y', x, 1, observed=data) + + y_missing, = model.missing_values + assert y_missing.tag.test_value.shape == (2,) + + model.logp(model.test_point) + + +def test_missing_pandas(): + data = pd.DataFrame([1,2,numpy.nan,4,numpy.nan]) + with Model() as model: + x = Normal('x', 1, 1) + y = Normal('y', x, 1, observed=data) + + y_missing, = model.missing_values + assert y_missing.tag.test_value.shape == (2,) + + model.logp(model.test_point) diff --git a/pymc3/theanof.py b/pymc3/theanof.py index 69963b3ff2..489e47b7a4 100644 --- a/pymc3/theanof.py +++ b/pymc3/theanof.py @@ -1,5 +1,5 @@ from .vartypes import typefilter, continuous_types -from theano import theano, tensor as t +from theano import theano, scalar, tensor as t from theano.gof.graph import inputs from .memoize import memoize @@ -43,12 +43,16 @@ def gradient1(f, v): return t.flatten(t.grad(f, v, disconnected_inputs='warn')) +empty_gradient = t.zeros(0, dtype='float32') @memoize def gradient(f, vars=None): if vars is None: vars = cont_inputs(f) - return t.concatenate([gradient1(f, v) for v in vars], axis=0) + if vars: + return t.concatenate([gradient1(f, v) for v in vars], axis=0) + else: + return empty_gradient def jacobian1(f, v): @@ -67,7 +71,10 @@ def jacobian(f, vars=None): if vars is None: vars = cont_inputs(f) - return t.concatenate([jacobian1(f, v) for v in vars], axis=1) + if vars: + return t.concatenate([jacobian1(f, v) for v in vars], axis=1) + else: + return empty_gradient @memoize @@ -91,7 +98,10 @@ def hessian_diag(f, vars=None): if vars is None: vars = cont_inputs(f) - return -t.concatenate([hessian_diag1(f, v) for v in vars], axis=0) + if vars: + return -t.concatenate([hessian_diag1(f, v) for v in vars], axis=0) + else: + return empty_gradient def makeiter(a): @@ -99,3 +109,31 @@ def makeiter(a): return a else: return [a] + + +class IdentityOp(scalar.UnaryScalarOp): + @staticmethod + def st_impl(x): + return x + + def impl(self, x): + return x + + def grad(self, inp, grads): + x, = inp + gz, = grads + return grads + + def c_code(self, node, name, inp, out, sub): + x, = inp + z, = out + return """%(z)s = %(x)s;""" % locals() + + def __eq__(self, other): + return type(self) == type(other) + + def __hash__(self): + return hash(type(self)) + +scalar_identity = IdentityOp(scalar.upgrade_to_float, name='scalar_identity') +identity = t.Elemwise(scalar_identity, name='identity') diff --git a/pymc3/vartypes.py b/pymc3/vartypes.py index cfe046828a..24a7225a59 100644 --- a/pymc3/vartypes.py +++ b/pymc3/vartypes.py @@ -22,4 +22,4 @@ def typefilter(vars, types): # Returns variables of type `types` from `vars` - return filter(lambda v: v.dtype in types, vars) + return [v for v in vars if v.dtype in types]