def getLastMonthData():
trainingFeatures, trainTarget, test = pipeline.load_data()
pipeline.digitizeMatrix(test)
print("Finding last month account statuses")
lastMonthData = trainingFeatures[trainingFeatures.fecha_dato == 15]
lastMonthLabels = trainTarget[trainingFeatures.fecha_dato == 15]
#lastMonthData= lastMonthData.groupby(lastMonthData.columns, axis = 1).transform(lambda x: x.fillna(x.mean()))
lastMonthData = lastMonthData.fillna(lastMonthData.mean())
return (trainingFeatures, trainTarget, test, lastMonthData,
lastMonthLabels)
def link_reg(sub):
import os
import pipeline as p
d = p.load_data("%s/info.txt" % sub)
if BaleenHP in d:
os.chdir("%s/BaleenHP" % sub)
src = "%s/register.dof6.dat" % d["BaleenHP"][Run1XXX]
dst = "register.dat"
print("%s -> %s" % (src dst))
if os.path.isfile(src) and not os.path.exists(dst):
try:
os.symlink(src dst)
except:
pass
def get_data(subjects):
all_data = {}
for sub in subjects:
eve_dat = op.join('/%s/kuperberg/SemPrMM/MEG/data/%s/eve' % (pre, sub), 'info.txt')
try:
dat = load_data(eve_dat)
# reorient dat into all_dat
for k,v in dat.items():
if k not in all_data:
all_data[k] = {}
all_data[k][sub] = v
except IOError:
pass
return all_data
def get_data(subjects):
all_data = {}
for sub in subjects:
eve_dat = op.join('/%s/kuperberg/SemPrMM/MEG/data/%s/eve' % (pre, sub),
'info.txt')
try:
dat = load_data(eve_dat)
# reorient dat into all_dat
for k, v in dat.items():
if k not in all_data:
all_data[k] = {}
all_data[k][sub] = v
except IOError:
pass
return all_data
def main():
data_path = '../data/techcrunch.csv'
result = load_data(data_path)
def main():
result = pp.load_data(FILE_NAME)
for dict_ in result:
print("=> ", dict_)
def main():
data_path = "../data/techcrunch.csv"
result = load_data(data_path)
print("Result: ", result)
def main():
data_path = '../data/techcrunch.csv'
result = load_data(data_path)
print("Result", result)
import pipeline
import submitData
from sklearn.decomposition import PCA
from sklearn.ensemble import GradientBoostingClassifier
trainingFeatures, trainTarget, test = pipeline.load_data()
pipeline.digitizeMatrix(test)
print("====================Cleaning Test Data====================")
del test['index']
del test['prev_fecha_dato']
test = test.fillna(test.mean())
print("====================DONE Cleaning Test Data====================")
print(
"====================Filling Training Feature NA's with means...==================="
)
trainingFeatures.fillna(trainingFeatures.mean())
print(
"====================DONE Filling Training Feature NA's with means...==================="
)
print("===============Running PCA==================")
pca = PCA(n_components=8)
training = pca.fit_transform(trainingFeatures)
testing = pca.fit_transform(test)
print("===============DONE Runnning PCA============")
print("==============Classifiying================")
clf = GradientBoostingClassifier(n_estimators=70, verbose=True)
predictions = submitData.fitClassifier(train,
def main():
data_path = '../data/techcrunch.csv' # Angabe des Datenpfades
resultat = load_data(data_path)
print("Resultat: ", resultat)
