if __name__ == "__main__":
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
### MPI map
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
try:
from SimpleMPI.MPI_map import MPI_map, is_master_process
except ImportError:
MPI_map = map
is_master_process = lambda: True
allret = MPI_map(run, map(lambda x: [x], [0.01, 0.1, 1.0] * 100 ))
if is_master_process():
allfbs = FiniteBestSet(max=True)
allfbs.merge(allret)
H = allfbs.get_all()
for h in H:
h.likelihood_temperature = 0.01 # on what set of data we want?
h.compute_posterior(data)
# show the *average* ll for each hypothesis
for h in sorted(H, key=lambda h: h.posterior_score):
print h.posterior_score, h.prior, h.likelihood, h.likelihood_temperature
print h
""" # initialize the data data = generate_data(data_size) # starting hypothesis -- here this generates at random h0 = NumberExpression(G) hyps = FiniteBestSet(max=True, N=options.TOP_COUNT, key="posterior_score") hyps.add( mh_sample(h0, data, options.STEPS, trace=False) ) return hyps # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # Main running if is_master_process(): display_option_summary(options) # choose the appropriate map function allret = MPI_map(run, map(lambda x: [x], options.DATA_AMOUNTS * options.CHAINS)) # Handle all of the output allfs = FiniteBestSet(max=True) allfs.merge(allret) allfs.save(options.OUT_PATH) # save this in a file ## If we want to print the summary with the "large" data size (average posterior score computed empirically) if options.LARGE_DATA_SIZE > 0 and is_master_process(): #now evaluate on different amounts of data too: huge_data = generate_data(options.LARGE_DATA_SIZE)
from LOTlib.Inference.MetropolisHastings import mh_sample
for h in lot_iter(mh_sample(h0, data, SAMPLES)):
fbs.add(h, h.posterior_score)
return fbs
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
### MPI map
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
from SimpleMPI.MPI_map import MPI_map, is_master_process
allret = MPI_map(run, map(lambda x: [x], [0.01, 0.1, 1.0] * 100 ))
if is_master_process():
allfbs = FiniteBestSet(max=True)
allfbs.merge(allret)
H = allfbs.get_all()
for h in H:
h.likelihood_temperature = 0.01 # on what set of data we want?
h.compute_posterior(data)
# show the *average* ll for each hypothesis
for h in sorted(H, key=lambda h: h.posterior_score):
print h.posterior_score, h.prior, h.likelihood, h.likelihood_temperature
print h
"""
# initialize the data
data = generate_data(data_size)
# starting hypothesis -- here this generates at random
h0 = NumberExpression(grammar)
hyps = FiniteBestSet(max=True, N=options.TOP_COUNT, key="posterior_score")
hyps.add( mh_sample(h0, data, options.STEPS, trace=False) )
return hyps
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
# Main running
if is_master_process():
display_option_summary(options)
# choose the appropriate map function
allret = MPI_map(run, map(lambda x: [x], options.DATA_AMOUNTS * options.CHAINS))
# Handle all of the output
allfs = FiniteBestSet(max=True)
allfs.merge(allret)
import pickle
with open(options.OUT_PATH, 'w') as f:
pickle.dump(allfs, f)
## If we want to print the summary with the "large" data size (average posterior score computed empirically)
if options.LARGE_DATA_SIZE > 0 and is_master_process():
