def m_svm():
if ARGV.one_vs:
svm_model = svm.svm_to_binary(svm.svm_raw())
else:
svm_model = multi.one_against_one(svm.svm_to_binary(svm.svm_raw()))
# return milk.defaultclassifier(mode='slow', multi_strategy='1-vs-1')
learner = milk.supervised.classifier.ctransforms(
# remove nans
supervised.normalise.chkfinite(),
# normalize to [-1,1]
supervised.normalise.interval_normalise(),
# feature selection
featureselection.featureselector(
featureselection.linear_independent_features),
# sda filter
featureselection.sda_filter(),
# same parameter range as 'medium'
supervised.gridsearch(
svm_model,
params = {
'C': 2.0 ** np.arange(-2, 4),
'kernel': [ svm.rbf_kernel(2.0 ** i) for i in xrange(-4, 4) ]
}
)
)
return learner
def metric(self):
weak = milk.supervised.tree.stump_learner()
model = milk.supervised.adaboost.boost_learner(weak)
model = one_against_one(model)
totalTimer = Timer()
with totalTimer:
model = model.train(self.data_split[0], self.data_split[1])
metric = {}
metric["runtime"] = totalTimer.ElapsedTime()
return metric
def test_too_many_boolean_indices_regression():
import milk.supervised.randomforest
import milk.supervised.adaboost
import milksets.wine
from milk.supervised.multi import one_against_one
weak = milk.supervised.randomforest.rf_learner()
learner = milk.supervised.adaboost.boost_learner(weak)
learner = one_against_one(learner)
features, labels = milksets.wine.load()
# sample features so that the test is faster (still gives error):
learner.train(features[::16], labels[::16])
def test_too_many_boolean_indices_regression():
import milk.supervised.randomforest
import milk.supervised.adaboost
import milksets.wine
from milk.supervised.multi import one_against_one
weak = milk.supervised.randomforest.rf_learner()
learner = milk.supervised.adaboost.boost_learner(weak)
learner = one_against_one(learner)
features, labels = milksets.wine.load()
# sample features so that the test is faster (still gives error):
learner.train(features[::16], labels[::16])
def leave_one_out(y, x, param=None, n=None):
results = []
for i, test in enumerate(zip(y, x)):
training_y = y[:i] + y[i+1:]
training_x = x[:i] + x[i+1:]
training_y = np.array(training_y)
training_x = np.array([np.array(tx) for tx in training_x])
learner = mil_rf.rf_learner()
learner = mil_multi.one_against_one(learner)
model = learner.train(training_x, training_y)
result = model.apply(np.array(x[i:i+1][0]))
results.append((result,) + (test[0],
make_d.decode(x[i], DECODE_DIC)))
return results
def random_forest(X, Y):
from milk.supervised import randomforest
from milk.supervised.multi import one_against_one
import milk.nfoldcrossvalidation
features = np.array(X)
labels = np.array(Y)
rf_learner = randomforest.rf_learner()
# rf is a binary learner, so we transform it into a multi-class classifier
learner = one_against_one(rf_learner)
# learner = rf_learner
# result
# cross validate with this learner and return predictions on left-out elements
cmat, names, preds = milk.nfoldcrossvalidation(features, labels, nfolds=4, classifier=learner, return_predictions=1)
print 'cross-validation accuracy:', cmat.trace() / float(cmat.sum())
return
def random_forest(X, Y):
from milk.supervised import randomforest
from milk.supervised.multi import one_against_one
import milk.nfoldcrossvalidation
features = np.array(X)
labels = np.array(Y)
rf_learner = randomforest.rf_learner()
# rf is a binary learner, so we transform it into a multi-class classifier
learner = one_against_one(rf_learner)
# learner = rf_learner
# result
# cross validate with this learner and return predictions on left-out elements
cmat, names, preds = milk.nfoldcrossvalidation(features,
labels,
nfolds=4,
classifier=learner,
return_predictions=1)
print 'cross-validation accuracy:', cmat.trace() / float(cmat.sum())
return
def m_svm():
if ARGV.one_vs:
svm_model = svm.svm_to_binary(svm.svm_raw())
else:
svm_model = multi.one_against_one(svm.svm_to_binary(svm.svm_raw()))
# return milk.defaultclassifier(mode='slow', multi_strategy='1-vs-1')
learner = milk.supervised.classifier.ctransforms(
# remove nans
supervised.normalise.chkfinite(),
# normalize to [-1,1]
supervised.normalise.interval_normalise(),
# feature selection
featureselection.featureselector(
featureselection.linear_independent_features),
# sda filter
featureselection.sda_filter(),
# same parameter range as 'medium'
supervised.gridsearch(
svm_model,
params={
'C': 2.0**np.arange(-2, 4),
'kernel': [svm.rbf_kernel(2.0**i) for i in xrange(-4, 4)]
}))
return learner
from milk.supervised import randomforest
from milk.supervised.multi import one_against_one
import milk.nfoldcrossvalidation
import milk.unsupervised
import pylab
from milksets import wine
# Load 'wine' dataset
features, labels = wine.load()
# random forest learner
rf_learner = randomforest.rf_learner()
# rf is a binary learner, so we transform it into a multi-class classifier
learner = one_against_one(rf_learner)
# cross validate with this learner and return predictions on left-out elements
cmat,names, preds = milk.nfoldcrossvalidation(features, labels, classifier=learner, return_predictions=1)
print('cross-validation accuracy:', cmat.trace()/float(cmat.sum()))
# dimensionality reduction for display
x,v = milk.unsupervised.pca(features)
colors = "rgb" # predicted colour
marks = "xo" # whether the prediction was correct
for (y,x),p,r in zip(x[:,:2], preds, labels):
c = colors[p]
m = marks[p == r]
pylab.plot(y,x,c+m)
pylab.show()
def BuildModel(self):
# Create and train the classifier.
rf_learner = randomforest.rf_learner(**self.opts)
learner = one_against_one(rf_learner)
return learner
def BuildModel(self): # Create and train the classifier. weak = milk.supervised.tree.stump_learner() learner = milk.supervised.adaboost.boost_learner(weak) learner = one_against_one(learner) return learner
def BuildModel(self): # Create and train the classifier. rf_learner = randomforest.rf_learner(**self.opts) learner = one_against_one(rf_learner) return learner
