def split(self, x_data, y_data):
Xt, Yt, Xv, Yv = super(SMOTESplitter, self).split(x_data, y_data)
Xt_smote, Yt_smote = SMOTE(**self._smote_params).fit_transform(
Xt.as_matrix(), Yt.as_matrix())
Xt_smote, Yt_smote = UnderSampler(
ratio=self._under_sample).fit_transform(Xt_smote, Yt_smote)
return Xt_smote, Yt_smote, Xv, Yv
def sampling():
verbose = False
y = np.bincount(target_train1)
print y
ratio = float(y[2]) / float(y[1])
# 'Random over-sampling'
OS = OverSampler(ratio=ratio, verbose=verbose)
osx, osy = OS.fit_transform(data_train1, target_train1)
random_methods(osx,osy)
# 'SMOTE'
smote = SMOTE(ratio=ratio, verbose=verbose, kind='regular')
smox, smoy = smote.fit_transform(data_train1, target_train1)
random_methods(smox,smoy)
# 'SMOTE bordeline 1'
bsmote1 = SMOTE(ratio=ratio, verbose=verbose, kind='borderline1')
bs1x, bs1y = bsmote1.fit_transform(data_train, target_train)
random_methods(bs1x,bs1y)
# 'SMOTE bordeline 2'
bsmote2 = SMOTE(ratio=ratio, verbose=verbose, kind='borderline2')
bs2x, bs2y = bsmote2.fit_transform(data_train1, target_train1)
random_methods(bs2x,bs2y)
# 'SMOTE SVM'
svm_args={'class_weight' : 'auto'}
svmsmote = SMOTE(ratio=ratio, verbose=verbose, kind='svm', **svm_args)
svsx, svsy = svmsmote.fit_transform(data_train1, target_train1)
random_methods(svsx,svsy)
# 'SMOTE Tomek links'
STK = SMOTETomek(ratio=ratio, verbose=verbose)
stkx, stky = STK.fit_transform(data_train1, target_train1)
random_methods(stkx,stky)
# 'SMOTE ENN'
SENN = SMOTEENN(ratio=ratio, verbose=verbose)
ennx, enny = SENN.fit_transform(data_train1, target_train1)
random_methods(ennx,enny)
# 'EasyEnsemble'
EE = EasyEnsemble(verbose=verbose)
eex, eey = EE.fit_transform(data_train1, target_train1)
random_methods(eex,eey)
# 'BalanceCascade'
BS = BalanceCascade(verbose=verbose)
bsx, bsy = BS.fit_transform(data_train1, target_train1)
random_methods(bsx,bsy)
def resample(self, X, y, t, fold):
if not self.resample_method:
return X, y
else:
start = time.time()
if self.verbose:
ptf('> Resampling for timestep %d, fold %d' % (t, fold),
self.logfile)
# create resampler
if self.resample_method == 'under':
print 'UNDER SAMPLING is not implemented yet'
return X, y
elif self.resample_method == 'over':
if self.oversample_method.lower() == 'smote':
resampler = SMOTE(**self.oversample_arguments)
else:
print 'Your resampling method is not implemented yet'
return X, y
print type(X), type(y)
print X.shape, y[0].shape
Xsmote, ysmote = resampler.fit_transform(X, y[0])
# resample
ysmote_tuple = self.build_smoted_label_tuple(ysmote, y, fold)
# ysmote_df = self.build_smoted_label_df(ysmote, y, fold)
# # find new folds
# folds, ynewdf = self.find_new_folds(Xsmote, ysmote, y)
if self.debug:
print np.sum(y[0] == 0), np.sum(ysmote == 0)
print np.sum(y[0] == 1), np.sum(ysmote == 1)
if self.on_disk:
self.pickle_time_step(ysmote_tuple,
'trigger_resample_labels',
fold=fold,
t=t)
self.pickle_time_step(Xsmote,
'trigger_resample_features',
fold=fold,
t=t)
else:
self.trigger_resample_labels[fold][t] = ysmote_tuple
self.trigger_resample_features[fold][t] = Xsmote
end = time.time()
if self.verbose:
ptf('... %d s' % (end - start), self.logfile)
return Xsmote, ysmote_tuple
def apply_sampling(X_data, Y_data, sampling, n_states, maxlen):
ratio = float(np.count_nonzero(Y_data == 1)) / \
float(np.count_nonzero(Y_data == 0))
X_data = np.reshape(X_data, (len(X_data), n_states * maxlen))
# 'Random over-sampling'
if sampling == 'OverSampler':
OS = OverSampler(ratio=ratio, verbose=True)
# 'Random under-sampling'
elif sampling == 'UnderSampler':
OS = UnderSampler(verbose=True)
# 'Tomek under-sampling'
elif sampling == 'TomekLinks':
OS = TomekLinks(verbose=True)
# Oversampling
elif sampling == 'SMOTE':
OS = SMOTE(ratio=1, verbose=True, kind='regular')
# Oversampling - Undersampling
elif sampling == 'SMOTETomek':
OS = SMOTETomek(ratio=ratio, verbose=True)
# Undersampling
elif sampling == 'OneSidedSelection':
OS = OneSidedSelection(verbose=True)
# Undersampling
elif sampling == 'CondensedNearestNeighbour':
OS = CondensedNearestNeighbour(verbose=True)
# Undersampling
elif sampling == 'NearMiss':
OS = NearMiss(version=1, verbose=True)
# Undersampling
elif sampling == 'NeighbourhoodCleaningRule':
OS = NeighbourhoodCleaningRule(verbose=True)
# ERROR: WRONG SAMPLER, TERMINATE
else:
print('Wrong sampling variable you have set... Exiting...')
sys.exit()
# print('shape ' + str(X.shape))
X_data, Y_data = OS.fit_transform(X_data, Y_data)
return X_data, Y_data
def _sample_values(X, y, method=None, ratio=1, verbose=False):
"""Perform any kind of sampling(over and under).
Parameters
----------
X : array, shape = [n_samples, n_features]
Data.
y : array, shape = [n_samples]
Target.
method : str, optional default: None
Over or under smapling method.
ratio: float
Unbalanced class ratio.
Returns
-------
X, y : tuple
Sampled X and y.
"""
if method == 'SMOTE':
sampler = SMOTE(ratio=ratio, verbose=verbose)
elif method == 'SMOTEENN':
ratio = ratio * 0.3
sampler = SMOTEENN(ratio=ratio, verbose=verbose)
elif method == 'random_over_sample':
sampler = OverSampler(ratio=ratio, verbose=verbose)
elif method == 'random_under_sample':
sampler = UnderSampler(verbose=verbose)
elif method == 'TomekLinks':
sampler = TomekLinks(verbose=verbose)
return sampler.fit_transform(X, y)
y = pd.read_csv(tain_path,
header=None,
index_col=False,
names=colnames,
skiprows=[0],
usecols=[8])
y = y['violation'].values
# X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.333, random_state=0)
main_x = X.values
main_y = y
verbose = False
ratio = float(np.count_nonzero(y == 1)) / float(np.count_nonzero(y == 0))
# 'SMOTE'
smote = SMOTE(ratio=ratio, verbose=verbose, kind='regular')
x, y = smote.fit_transform(main_x, main_y)
ratio = float(np.count_nonzero(y == 1)) / float(np.count_nonzero(y == 0))
X_train, X_test, y_train, y_test = train_test_split(x,
y,
test_size=.333,
random_state=0)
from sklearn.ensemble import RandomForestClassifier
from sklearn.cross_validation import cross_val_score
clf = RandomForestClassifier(n_estimators=10)
scores = cross_val_score(clf, X_test, y_test)
y_pred = clf.fit(X_train, y_train).predict(X_test)
n_source, n_target, n_samples = None, None, None
for f in args.read_train:
npzfile = np.load(f)
print "Read %d instances from %s" \
% (npzfile['feature_matrix'].shape[0], f.name)
assert npzfile['targets'].size == npzfile['feature_matrix'].shape[0]
tgt, fm = npzfile['targets'], npzfile['feature_matrix']
print "target size: ", tgt.shape
print "positive examples: ", sum(sum(tgt))
tgt = tgt.reshape(tgt.size)
if args.smote:
ratio = float(np.count_nonzero(tgt == 0)) / \
float(np.count_nonzero(tgt == 1))
OS = SMOTE(ratio=ratio, kind='regular')
fm, tgt = OS.fit_transform(fm, tgt)
if targets is None:
targets = tgt
m = fm
else:
print "Before concat: ", targets.shape, tgt.shape
targets = np.concatenate((targets, tgt), axis=0)
m = np.concatenate((m, fm), axis=0)
print "After concat: ", targets.shape, tgt.shape
assert targets.size == m.shape[0]
assert m.shape[0] == targets.shape[0]
print "Sum of targets: ", sum(targets)
label_index = 11
folds = os.listdir(location)
delimiter = '\t'
folds[:] = [location + i for i in folds]
verbose = True
## LOAD RAW DATA ##
raw_data = []
feat_content = []
labels = []
for fold_nr in range(0, len(folds)):
file = open(folds[fold_nr], 'r')
raw_data.append([])
feat_content.append([])
labels.append([])
for line in file.readlines():
raw_data[fold_nr].append(line.split(delimiter))
feat_content[fold_nr] = [
row[content_index] for row in raw_data[fold_nr]
]
labels[fold_nr] = [row[label_index] for row in raw_data[fold_nr]]
labels[fold_nr] = [
0 if label == 'f' else 1 for label in labels[fold_nr]
]
file.close()
## EVALUATE ##
resampler = SMOTE(verbose=verbose)
pip = Pipeline(feat_content, labels, resampler, verbose)
f1_complete = pip.validation()
y = pd.read_csv(tain_path,
header=None,
index_col=False,
names=colnames,
skiprows=[0],
usecols=[8])
y = y['violation'].values
# X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.333, random_state=0)
main_x = X.values
main_y = y
verbose = False
ratio = float(np.count_nonzero(y == 1)) / float(np.count_nonzero(y == 0))
# 'SMOTE bordeline 2'
bsmote2 = SMOTE(ratio=ratio, verbose=verbose, kind='borderline2')
x, y = bsmote2.fit_transform(main_x, main_y)
ratio = float(np.count_nonzero(y == 1)) / float(np.count_nonzero(y == 0))
X_train, X_test, y_train, y_test = train_test_split(x,
y,
test_size=.333,
random_state=0)
from sklearn.ensemble import RandomForestClassifier
from sklearn.cross_validation import cross_val_score
clf = RandomForestClassifier(n_estimators=10)
scores = cross_val_score(clf, X_test, y_test)
y_pred = clf.fit(X_train, y_train).predict(X_test)
def f6(x):
if x == 7:
return 2
elif x > 7:
return x - 1
else:
return x
#y = map(f6, y)
y = numpy.array(y)
sm = SMOTE(kind='regular')
for i in xrange(20):
x_metrics, y = sm.fit_transform(x_metrics, y)
clf = RandomForestClassifier(n_estimators=100, class_weight='auto')
pr = cross_validation.cross_val_predict(clf, x_metrics, y, cv=10)
#print metrics.accuracy_score(y, pr)
#print metrics.confusion_matrix(y, pr)
delete_rows_indexes = [i for i, y_i in enumerate(pr) if y_i == 2 and y[i] == 7]
x_metrics = numpy.delete(x_metrics, delete_rows_indexes, axis=0)
y = numpy.delete(y, delete_rows_indexes, axis=0)
#clf.fit(x_metrics,y)
#joblib.dump(clf, 'rand_forest_model_3.pkl')
pr = cross_validation.cross_val_predict(clf, x_metrics, y, cv=10)
usecols=[3, 4, 5, 6, 7])
y = pd.read_csv(tain_path,
header=None,
index_col=False,
names=colnames,
skiprows=[0],
usecols=[8])
y = y['violation'].values
# X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.333, random_state=0)
main_x = X.values
main_y = y
verbose = False
ratio = float(np.count_nonzero(y == 1)) / float(np.count_nonzero(y == 0))
# 'SMOTE bordeline 1'
bsmote1 = SMOTE(ratio=ratio, verbose=verbose, kind='borderline1')
x, y = bsmote1.fit_transform(main_x, main_y)
ratio = float(np.count_nonzero(y == 1)) / float(np.count_nonzero(y == 0))
X_train, X_test, y_train, y_test = train_test_split(x,
y,
test_size=.333,
random_state=0)
from sklearn.ensemble import RandomForestClassifier
from sklearn.cross_validation import cross_val_score
clf = RandomForestClassifier(n_estimators=10)
scores = cross_val_score(clf, X_test, y_test)
y_pred = clf.fit(X_train, y_train).predict(X_test)
orig_X, orig_y = read_data()
skf = StratifiedKFold(orig_y, n_folds=4, shuffle=True)
while True:
scores = []
for train_index, test_index in skf:
X, X_cv = orig_X[train_index], orig_X[test_index]
y, y_cv = orig_y[train_index], orig_y[test_index]
# Fraction of majority samples to draw with respect to samples of
# minority class.
sampled_X,sampled_y = X,y
# Oversample data from the minority class.
if P['is_smote']:
sampled_X, sampled_y = SMOTE(k=P['k'], m=P['m'], ratio=P['ratio'], verbose=False, kind='regular').fit_transform(sampled_X, sampled_y)
# Undersample samples from the majority class.
sampled_X, sampled_y = UnderSampler(1.0).fit_transform(sampled_X, sampled_y)
# Fit a scaler only for the sampled data.
scaler = Scaler(sampled_X, sampled_y)
sampled_X = scaler.getOriginalTransformedData()
#model = RandomForestClassifier(n_estimators=100).fit(sampled_X, sampled_y)
#model = RandomForestClassifier(n_estimators=P['n_estimators'], criterion=P['criterion'], max_depth=P['max_depth'], min_samples_split=P['min_samples_split'], min_samples_leaf=P['min_samples_leaf'], min_weight_fraction_leaf=P['min_weight_fraction_leaf'], max_features=P['max_features'], max_leaf_nodes=P['max_leaf_nodes'], bootstrap=P['bootstrap'], oob_score=P['oob_score'], n_jobs=8, random_state=None, verbose=0, warm_start=False, class_weight=None).fit(sampled_X, sampled_y)
model = MLPClassifier(activation=P['activation'], algorithm=P['algorithm'], alpha=P['alpha'], hidden_layer_sizes=P['layer'], learning_rate=P['learning_rate'], tol=P['tol'], random_state=1).fit(sampled_X, sampled_y)
#model = xgb.XGBClassifier(max_depth=P['max_depth'], n_estimators=P['n_estimators'], learning_rate=P['learning_rate'], nthread=8, subsample=P['subsample'], colsample_bylevel=P['colsample_bylevel']).fit(sampled_X, sampled_y, eval_metric=P['eval_metric'])
prediction_cv = model.predict_proba(scaler.transform(X_cv))
auc_score = roc_auc_score(y_cv, prediction_cv[:,1])
scores.append(auc_score)
log("***roc_auc_score:%f" % auc_score)
