totalvalid = totaltest = 0
for name in names:
net = NNet.load(filepath = join(BASE_DIR, 'results', 'nets', name))
for nm, val in net.get_params().iteritems():
print '{0:s} = {1:}'.format(nm, val)
#for nm, data in [('val', valid), ('tst', test)]:
#probs = net.predict_proba(data)
#save(join(SUBMISSIONS_DIR, '{0}_{1}_raw.npy'.format(name, nm)), probs)
#makeSubmission(probs, fname = join(SUBMISSIONS_DIR, '{0}_{1}_rescale.csv'.format(name, nm)), digits = 8)
probs = net.predict_proba(valid)
probs = scale_to_priors(probs, priors = PRIORS)
save(join(SUBMISSIONS_DIR, '{0}_valid.npy'.format(name)), probs)
totalvalid += probs
probs = net.predict_proba(test)
probs = scale_to_priors(probs, priors = PRIORS)
save(join(SUBMISSIONS_DIR, '{0}_test.npy'.format(name)), probs)
makeSubmission(probs, fname = join(SUBMISSIONS_DIR, '{0}_test.csv'.format(name)), digits = 8)
totaltest += probs
save(join(SUBMISSIONS_DIR, 'total_valid.npy'), totalvalid)
save(join(SUBMISSIONS_DIR, 'total_valid.npy'), totaltest)
makeSubmission(totaltest, fname = join(SUBMISSIONS_DIR, 'total_test.csv'), digits = 8)
print 'saved predictions'
'nn__weight_decay': [1.0e-05, 0],
},
fit_params = {
},
n_iter = 10,
n_jobs = 10,
scoring = log_loss_scorer,
iid = False,
refit = True,
#pre_dispatch = cpus + 2,
cv = ShuffleSplit(
n = train.shape[0],
n_iter = 1,
test_size = 0.2,
#random_state = random,
),
#random_state = random,
verbose = bool(VERBOSITY),
error_score = -1000000,
)
opt.fit(train, labels)
with open(join(LOGS_DIR, 'debug_{0:.4f}.json'.format(-opt.best_score_)), 'w+') as fh:
print 'saving results (no scaling to priors) for top score {0:.4f}:'.format(-opt.best_score_), opt.best_params_
dump(opt.best_params_, fp = fh, indent = 4)
probs = opt.best_estimator_.predict_proba(test)
makeSubmission(probs, fname = join(SUBMISSIONS_DIR, 'debug_{0:.4f}.csv'.format(-opt.best_score_)), digits = 8)
testmodels = (foresttest1, gradienttest1, boostedtest1, foresttest2,
gradienttest2, boostedtest2, foresttest3, gradienttest3,
boostedtest3, svmtest, nntest, knn4test, knn32test,
knn256test, treeboosttest)
ptest = np.hstack(testmodels)
ptest = stack_predictions(ptest, len(testmodels))
print "ptest", np.shape(ptest)
weights = [1, 24, 28, 0, 0, 31, 38, 1, 34, 18, 0, 0, 4]
#weights = bestfitness[1]
trainfeat = np.load('data/testmat.npy')
testfeat, _ = get_testing_data()
classi = classifierEnsemble(p, trueclasses, ptest, trainfeat[:, 1:],
testfeat)
#mean = mean_ensemble(ptest, weights)
from utils.ioutil import makeSubmission
import os
if os.path.isfile('classsub.csv'):
os.remove('classsub.csv')
makeSubmission(classi, 'classsub.csv')
#if os.path.isfile('meansub.csv'):
# os.remove('meansub.csv')
#makeSubmission(mean, 'meansub.csv')
print "Done making submission"