def map(prob, modulename = 'model' ):
try:
mystr = "from " + modulename + " import model"
exec(mystr)
except:
print 'cannot import', modulename
M = MAP( model(prob) )
M.fit()
return M
def map(prob, modulename='model'):
try:
mystr = "from " + modulename + " import model"
exec(mystr)
except:
print('cannot import', modulename)
M = MAP(model(prob))
M.fit()
return M
def sample(self,
model,
mapstart=False,
step_methods=None,
iters=80000,
burn=20000,
num_chains=4,
doplot=True,
showplots=False,
force=False,
progress_bar=False):
print('MCMC for %s' % self.name)
if self.is_done():
if not force:
print('\tAlready done, skipping...')
return self.db_file
else:
print(
'\tWARNING: recomputing, there might be spurious files from previous runs...'
) # Not a good idea
# Let's graph the model
graph = pymc.graph.dag(pymc.Model(model),
name=self.name,
path=self.model_dir)
graph.write_png(op.join(self.model_dir, self.name + '.png'))
start = time.time()
if mapstart:
# See http://stronginference.com/post/burn-in-and-other-mcmc-folklore
# BUT WARNING, WOULD THIS MAKE MULTIPLE CHAIN START BE OVERLY CORRELATED?
try:
from pymc import MAP
print('\tFinding MAP estimates...')
M = MAP(model)
M.fit()
model = M.variables
print('\tMAP estimates found...')
except Exception, e:
print('\tMAP Failed...', str(e))
from matplotlib import pylab as plt import dive_model import no_ba_model import asocial_model import pymc from pymc import MCMC, MAP from pymc.Matplot import plot as mcplot MASOC = MAP(asocial_model) MASOC.fit(method="fmin", iterlim=100000, tol=0.000001) print(MASOC.AIC) print(MASOC.BIC) MNOBA = MAP(no_ba_model) MNOBA.fit(method="fmin", iterlim=100000, tol=0.000001) print(MNOBA.AIC) print(MNOBA.BIC) print(MNOBA.dist.value) print(MNOBA.lag.value) print(MNOBA.intrinsic_rate.value) print(MNOBA.social_rate.value) MFULL = MAP(dive_model) MFULL.fit(method="fmin", iterlim=100000, tol=0.000001) print(MFULL.AIC) print(MFULL.BIC) print(MFULL.dist.value) print(MFULL.lag.value) print(MFULL.blind_angle.value) print(MFULL.intrinsic_rate.value)
import corRandomWalk import environment import constantModel import constantModelAlign import networkModel #import hybridModel #import hybridModelv2 import pymc from pymc import MAP from pymc.Matplot import plot as mcplot import networkModel NM = MAP(networkModel) NM.fit(method ='fmin', iterlim=100000, tol=.000001) print(NM.AIC) print(NM.BIC) #print(NM.decay_exponent.value) print(NM.interaction_length.value) print(NM.interaction_angle.value) # #import decayModelAlign # #DMA = MAP(decayModelAlign) #DMA.fit(method ='fmin', iterlim=100000, tol=.000001) #print(DMA.AIC) #print(DMA.BIC) #
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# Author: Yuki Furuta <*****@*****.**>
import numpy as np
from pymc import Dirichlet, Categorical, MAP, Model
if __name__ == '__main__':
alpha = np.array([30.,25.,20.,25.])
prob_dist = Dirichlet('prob_dist', alpha)
exp_data = np.array([1,1,3,2,2,1,0])
f_x = Categorical("cat", prob_dist, value=exp_data, observed=True)
model = Model({'f_x': f_x,
'prob_dist': prob_dist
})
m = MAP(model)
m.fit()
print m.prob_dist.value