def train_model(self, test_split=0.4):
# split training and test data
feature_train, feature_test, lable_train, lable_test = train_test_split(
data.get_features(), data.get_lables(), test_size=test_split
)
self.classifier.fit(feature_train, lable_train["lable"].values)
self.accuracy = self.get_accuracy(feature_test, lable_test["lable"].values)
def get_n_fold_validation_score(self, fold=10):
features = data.get_features()
lables = data.get_lables()
length = len(features)
jump = length / fold
index = 0
k = 0
scores = list()
while k < fold:
feature_test = features.iloc[index : (index + jump), :]
lable_test = lables.iloc[index : (index + jump), :]
feature_train_1, feature_train_2 = (
features.iloc[0 : index - 1, :] if index != 0 else pd.DataFrame(),
features.iloc[index + jump + 1 : length - 1],
)
feature_train = pd.concat([feature_train_1, feature_train_2])
lable_train_1, lable_train_2 = (
lables.iloc[0 : index - 1, :] if index != 0 else pd.DataFrame(),
lables.iloc[index + jump + 1 : length - 1],
)
lable_train = pd.concat([lable_train_1, lable_train_2])
index += jump
k += 1
classifier = GradientBoostingClassifier()
classifier.fit(feature_train, lable_train["lable"].values)
scores.append(accuracy_score(lable_test, classifier.predict(feature_test)))
return sum(scores) / float(len(scores))
def get_n_fold_validation_score(self, fold=10):
features = data.get_features()
lables = data.get_lables()
length = len(features)
jump = length / fold
index = 0
k = 0
scores = list()
while k < fold:
feature_test = features.iloc[index:(index + jump), :]
lable_test = lables.iloc[index:(index + jump), :]
feature_train_1, feature_train_2 = features.iloc[
0:index - 1, :] if index != 0 else pd.DataFrame(
), features.iloc[index + jump + 1:length - 1]
feature_train = pd.concat([feature_train_1, feature_train_2])
lable_train_1, lable_train_2 = lables.iloc[
0:index -
1, :] if index != 0 else pd.DataFrame(), lables.iloc[index +
jump +
1:length -
1]
lable_train = pd.concat([lable_train_1, lable_train_2])
index += jump
k += 1
classifier = GradientBoostingClassifier()
classifier.fit(feature_train, lable_train['lable'].values)
scores.append(
accuracy_score(lable_test, classifier.predict(feature_test)))
return sum(scores) / float(len(scores))
def train_model(self, test_split=0.4):
# split training and test data
feature_train, feature_test, lable_train, lable_test = train_test_split(
data.get_features(), data.get_lables(), test_size=test_split)
self.classifier.fit(feature_train, lable_train['lable'].values)
self.accuracy = self.get_accuracy(feature_test,
lable_test['lable'].values)
def train_model(self, test_split=0.1):
# split training and test data
feature_train, feature_test, lable_train, lable_test = train_test_split(data.get_features(), data.get_lables()
, test_size=test_split)
print "Start - Training Model"
self.classifier.fit(feature_train, lable_train)
print "Done - Training Model"
joblib.dump(self.classifier, MODEL_FILE)
def get_n_fold_validation_score(self, fold=10):
features = data.get_features()
lables = data.get_lables()
length = len(features)
jump = length/fold
index = 0
k = 0
scores = list()
while k < fold:
feature_test = features.iloc[index:(index + jump), :]
lable_test = lables.iloc[index: (index + jump), :]
feature_train_1, feature_train_2 = features.iloc[0: index-1, :] if index != 0 else pd.DataFrame(), features.iloc[index+jump+1: length-1]
feature_train = pd.concat([feature_train_1, feature_train_2])
lable_train_1, lable_train_2 = lables.iloc[0: index-1, :] if index != 0 else pd.DataFrame(), lables.iloc[index+jump+1: length-1]
lable_train = pd.concat([lable_train_1, lable_train_2])
index += jump
k += 1
classifier = RandomForestClassifier(criterion="gini", n_estimators=10)
classifier.fit(feature_train, lable_train['lable'].values)
scores.append(accuracy_score(lable_test, classifier.predict(feature_test)))
return sum(scores)/float(len(scores))
def predict_activities(self):
records = db.get_records_to_predict(helper.last_read())
helper.write_last_read()
user_activities = []
for record in records:
user_activities.append({
'user_id': record.pop('user_id'),
'distance': round(float(record.pop('distance')), 2),
'start_datetime': int(record['start_timestamp']),
'end_datetime': int(record.pop('start_timestamp')) + 5000,
'workout_type_id': int(float(self.classifier.predict(pd.DataFrame(record, index=[0]))[0]))
})
return user_activities
def fetch_predict_load_activities(self):
user_activities = self.predict_activities()
for user_activity in user_activities:
db.insert_activity(user_activity)
