def haupt():
train, test = explore('./Data')
clf = RandomForestClassifier(n_estimators=100, n_jobs=2)
"""
Cluster using FASTMAP
"""
# Training data
train_DF = createDF(train[1])
# Testing data
test_df = createDF(test[1])
# set_trace()
features = train_DF.columns[3:-2]
klass = train_DF[train_DF.columns[-1]]
clf.fit(train_DF[features], klass)
preds = clf.predict(test_df[features]).tolist()
# print preds
# def isdefective(data):
label = set(train_DF.columns[-1])
_id = list(set(train_DF[train_DF.columns[-1]]))
dfct = {lbl: str(np.mean(list(train_DF[train_DF['klass'] == lbl]['$<bug'])) >= 0.3) \
for lbl in _id}
# print label
predictions = [dfct[i] for i in preds]
actuals = [str(not i == 0) for i in test_df[test_df.columns[-2]].tolist()]
return _runAbcd(train=actuals, test=predictions, verbose=False)
def f1(rows): indep = rows[1:-1]; case = 0 # set_trace() whereParm, tree = update(indep) [test, train] = tdivPrec(where = None, dtree = tree, train = trainDat[1], test = testDat[1]); g = _runAbcd(train = train, test = test, verbose = False) return g
def haupt():
train, test = explore('./Data')
clf = RandomForestClassifier(n_estimators = 100, n_jobs = 2)
"""
Cluster using FASTMAP
"""
# Training data
train_DF = createDF(train[1])
# Testing data
test_df = createDF(test[1])
# set_trace()
features = train_DF.columns[3:-2]
klass = train_DF[train_DF.columns[-1]];
clf.fit(train_DF[features], klass)
preds = clf.predict(test_df[features]).tolist()
# print preds
# def isdefective(data):
label = set(train_DF.columns[-1]);
_id = list(set(train_DF[train_DF.columns[-1]]))
dfct = {lbl: str(np.mean(list(train_DF[train_DF['klass'] == lbl]['$<bug'])) >= 0.3) \
for lbl in _id}
# print label
predictions = [dfct[i] for i in preds]
actuals = [str(not i == 0) for i in test_df[test_df.columns[-2]].tolist()]
return _runAbcd(train = actuals, test = predictions, verbose = False)
def main(dir = None):
whereParm, tree = None, None # _de()
G = []; G1 = []; reps = 1;
trainDat, testDat = explore(dir = 'Data/')
for _ in xrange(reps):
print reps
[test, train] = tdivPrec(whereParm, tree, train = trainDat[1], test = testDat[0]);
g = _runAbcd(train = train, test = test, verbose = False)
G.append(g)
G.insert(0, 'DT ')
for _ in xrange(reps):
print reps
[test, train] = tdivPrec1(whereParm, tree, train = trainDat[1], test = testDat[0]);
g = _runAbcd(train = train, test = test, verbose = False)
G1.append(g)
G1.insert(0, 'C4.5')
return [G, G1]
