def main2(gdname, gdfunction, training, regression, learningrate):
'''Perform gradient descent with a learner and analyze the results.'''
# learning rate
try:
initiallr, reducelr = learningrate
suffix = 'dynamic' + str(initiallr)
except TypeError:
initiallr = learningrate
reducelr = lambda x: x
suffix = initiallr
# learn
weights = GradientDescent.loop(
len(training[0].features) * [0.0],
training,
regression.model,
regression.gradient,
initiallr,
reducelr,
gdfunction,
0.99)
# test
terror = stats.rmse(
[regression.model(weights, dp.features) for dp in testing],
[dp.label for dp in testing])
print INDENT * 2 + 'Testing RMSE:', terror
## produce a result set
results = [resultset.DataResult(dp.label,
regression.model(weights, dp.features)) \
for dp in testing]
## ## find a good operating point
## op = resultset.minerrop(results)
## print INDENT * 2 + 'Operating Point:', op
## ## assign predictions
## results = resultset.applyop(op, results)
## output roc data
roc = resultset.rocdata(results)
auc = resultset.auc(roc)
with open('{}-{}_lambda={}_auc={}'.format(regression.__name__, gdname,
suffix, auc).lower(),
mode='wb') as fd:
for fpr, tpr in roc:
fd.write('{}, {}\n'.format(fpr, tpr))
# curry the stump to a classifier function [dp --> -1 or 1]
curried = (lambda s: lambda dp: Stump.query(s, dp))(stump)
#
# boost new datapoint weights
wv = boost.round(curried)
#
# output status
print mistakect(training, H) / float(len(training)),
testerr = mistakect(testing, H) / float(len(testing))
print testerr,
#
# calculate auc
rv = [
resultset.DataResult(int(dp.label > 0), boost.model(dp))
for dp in testing
]
print resultset.auc(resultset.rocdata(rv)),
#
# done
print
print 'Boosting time:', str(datetime.datetime.now() - start)
rocfn = 'boostROC-{}.log'.format(datetime.datetime.now()).replace(':', '-')
print 'Writing ROC data to "{}"'.format(rocfn)
with open(rocfn, mode='wb') as fd:
for fpr, tpr in resultset.rocdata(rv):
fd.write('{} {}\n'.format(fpr, tpr))
#raw_input('[Press Enter to Quit]')
###############################################################################
stump = svpick(wv)
print stump[0],
#
# curry the stump to a classifier function [dp --> -1 or 1]
curried = (lambda s: lambda dp: Stump.query(s, dp))(stump)
#
# boost new datapoint weights
wv = boost.round(curried)
#
# output status
print mistakect(training, H) / float(len(training)),
testerr = mistakect( testing, H) / float(len( testing))
print testerr,
#
# calculate auc
rv = [resultset.DataResult(int(dp.label > 0), boost.model(dp)) for dp in testing]
print resultset.auc(resultset.rocdata(rv)),
#
# done
print
print 'Boosting time:', str(datetime.datetime.now() - start)
rocfn = 'boostROC-{}.log'.format(datetime.datetime.now()).replace(':', '-')
print 'Writing ROC data to "{}"'.format(rocfn)
with open(rocfn, mode='wb') as fd:
for fpr, tpr in resultset.rocdata(rv):
fd.write('{} {}\n'.format(fpr, tpr))
#raw_input('[Press Enter to Quit]')
###############################################################################