def save_to_file(self, dict, name):
file_path = Extension.get_data_dir(self.config) / ("%s.pkl" % name)
try:
with file_path.open("wb") as f:
pickle.dump(dict, f, pickle.HIGHEST_PROTOCOL)
pickle.close()
except Exception:
return False
def save_to_file(self, dict, name):
file_path = Path(self.config["iris"]["data_dir"]) / ("%s.pkl" % name)
try:
with file_path.open("wb") as f:
pickle.dump(dict, f, pickle.HIGHEST_PROTOCOL)
pickle.close()
except Exception:
return False
def train(trainingDataFolder, featureFile, labelFile):
y_train = genfromtxt(labelFile)
y_train1 = np.zeros(y_train.shape, dtype=np.int32)
for items in range(y_train.shape[0]):
y_train1[items] = y_train[items]
x_train1 = genfromtxt(featureFile, delimiter=',')
# Create a svm Classifier
# clf = svm.SVC(kernel='linear') # Linear Kernel
clf = svm.SVC(C=1.0,
cache_size=200,
class_weight=None,
coef0=0.0,
decision_function_shape='ovr',
degree=1,
gamma='auto',
kernel='linear',
max_iter=-1,
probability=False,
random_state=None,
shrinking=True,
tol=0.0001,
verbose=False)
X_train, X_test, y_train, y_test = train_test_split(x_train1,
y_train1,
test_size=0.3,
random_state=109)
clf.fit(x_train1, y_train1)
y_pred = clf.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)
y_pred1 = clf.predict(x_train1)
accuracy = accuracy_score(y_train1, y_pred1)
featuresTestfile = trainingDataFolder + '/featuresTest.csv'
x_test = genfromtxt(featuresTestfile, delimiter=',')
y_pred = np.zeros(x_test.shape[0], dtype=np.int32)
# Predict the response for test dataset
y_pred = clf.predict(x_test)
# Save model file
modelFile = trainingDataFolder + '/SVMModel.bin'
pickle.dump(clf, open(modelFile, 'wb'))
pickle.close()
modelFile1 = trainingDataFolder + '/SVMModel1.bin'
joblib.dump(clf, open(modelFile1, 'wb'))
return modelFile1
"""
#call functions of the file and this file only for ruleGeneration
#the use of a hash table superdict which was created in gen_powerset.py which has the support of all frequent candidates has been made to find support in O(1) time.
#we also use the fact that all subsets of a frequent candidate shall be frequent to avoid hitting the data set
import pickle
import copy
pkl_file = open("../pkl_files/superdict.pkl","rb")
superdict = pickle.load(pkl_file)
item = []
pkl_file.close()
pkl_file = open("../pkl_files/redundant_item.pkl","rb")
redundant_item = pickle.load(pkl_file)
pickle.close()
pkl_file = open("../pkl_files/hash_table.pkl","rb")
hash_table = pickle.load(pkl_file)
pkl_file.close()
#it is a recurssive functions which takes in lv1 rules as input and generate rules for rest of generations
def generateAllRules(x,y,used,item):
x = list(x)
y = list(y)
if len(x) > 1:
inum = 0
for i in x:
x_made = []
x_made = x[:]
y_made = y[:]
y_made.append(x[inum])
y_made.sort()
def createMonthlyDataFile(self):
path = "/home/prashant/pythonGeneratedFiles/"
year = "2015"
powerIndex = 10
for m in range(1, 13):
totalEnergy = 0
temp = datetime(1900, 1, 1, 0, 0, 0)
duration = datetime(1900, 1, 1, 0, 0, 0)
peak = 0
mFile = year
if m < 10:
mFile = mFile + "-0" + str(m)
else:
mFile = mFile + "-" + str(m)
fm = open(mFile + ".csv", "w")
fm.write("Total energy,Peak,Duration\n")
print "Creating Month :", m, " File..............."
if m == 1 or m == 3 or m == 5 or m == 7 or m == 8 or m == 10 or m == 12:
dateMax = 31
elif m == 2:
dateMax = 28
else:
dateMax = 30
for d in range(1, dateMax + 1):
fileName = year
if m < 10:
fileName = fileName + "-0" + str(m) + "-"
else:
fileName = fileName + "-" + str(m) + "-"
if d < 10:
fileName = fileName + "0" + str(d)
else:
fileName = fileName + str(d)
if os.path.exists(path + fileName + ".csv"):
fp = open(path + fileName + ".csv", "r")
print fp.readline().split(",")
startTime = datetime.strptime(fp.readline().split(",")[0],
"%H:%M")
for fileData in fp:
singleData = fileData.split(",")
if int(singleData[powerIndex]) > peak:
peak = singleData[powerIndex]
endTime = datetime.strptime(singleData[0], "%H:%M")
temp = endTime - startTime
duration = duration + temp
fp.close()
totalEnergy = totalEnergy + int(singleData[11])
print "Month :" + str(m) + "-" + year
print "Total Energy=", totalEnergy
print "Duration :", duration
print "Peak :", peak
fm.write(str(totalEnergy))
fm.write(",")
fm.write(str(peak))
fm.write(",")
fm.write(str(duration))
fm.close()