def main(): predictor = MFUPredictor() db_helper = ApeicDBHelper() users = db_helper.get_users() for user in users: logs = db_helper.get_logs(user) sessions = db_helper.get_sessions(user) training_logs, testing_logs = split(sessions, aggregated=True) predictor.train(training_logs) launches = map(lambda x: x['application'], testing_logs) predictions = map(lambda x: predictor.predict(x), testing_logs) hr, mrr = predictor.test(launches, predictions) print hr, mrr
def main(): predictor = MFUPredictor() db_helper = ApeicDBHelper() users = db_helper.get_users() for user in users: logs = db_helper.get_logs(user) sessions = db_helper.get_sessions(user) training_logs, testing_logs = split(sessions, aggregated=True) predictor.train(training_logs) launches = map(lambda x: x['application'], testing_logs) predictions = map(lambda x: predictor.predict(x), testing_logs) hr, mrr = predictor.test(launches, predictions) print hr, mrr
def main(): predictor = LUPredictor() db_helper = ApeicDBHelper() users = db_helper.get_users() for user in users: sessions = db_helper.get_sessions(user) training_logs, testing_logs = split(sessions, aggregated=True) predictor.train(training_logs) launches = map(lambda x: x['application'], testing_logs[2:]) predictions = map(lambda i: predictor.predict(\ {'lu1': testing_logs[i-1]['application'], 'lu2': testing_logs[i-2]['application']}), \ xrange(2, len(testing_logs))) hr, mrr = predictor.test(launches, predictions) print hr, mrr
def main(): predictor = LUPredictor() db_helper = ApeicDBHelper() users = db_helper.get_users() for user in users: sessions = db_helper.get_sessions(user) training_logs, testing_logs = split(sessions, aggregated=True) predictor.train(training_logs) launches = map(lambda x: x['application'], testing_logs[2:]) predictions = map(lambda i: predictor.predict(\ {'lu1': testing_logs[i-1]['application'], 'lu2': testing_logs[i-2]['application']}), \ xrange(2, len(testing_logs))) hr, mrr = predictor.test(launches, predictions) print hr, mrr
from sklearn.datasets.samples_generator import make_blobs from sklearn.preprocessing import StandardScaler ############################################################################## # Generate sample data centers = [[1, 1], [-1, -1], [1, -1]] X, labels_true = make_blobs(n_samples=100, centers=centers, cluster_std=0.4, random_state=0) X = StandardScaler().fit_transform(X) from numpy import array from predictor.predictor import Predictor, split db_helper = ApeicDBHelper() for user in db_helper.get_users()[7:]: sessions = db_helper.get_sessions(user) training_logs, testing_logs = split(sessions, aggregated=True) training_logs = filter( lambda x: x['latitude'] != 0 and x['longitude'] != 0, training_logs) latlng_pairs = list( set(map(lambda x: (x['latitude'], x['longitude']), training_logs))) print latlng_pairs print len(latlng_pairs) # X = array(latlng_pairs) # print X.size # print X result = [] for la1, ln1 in latlng_pairs: dists = []
from predictor.mfu_predictor import MFUPredictor from sklearn.naive_bayes import MultinomialNB, BernoulliNB from sklearn.naive_bayes import GaussianNB from sklearn.feature_extraction import DictVectorizer from sklearn import tree def split(sessions, ratio=0.8): split_index = int(len(sessions) * ratio) return sessions[:split_index], sessions[split_index:] import sys if __name__ == '__main__': db_helper = ApeicDBHelper() users = db_helper.get_users() accuracies = [] for user in users: print colored(user, attrs=['blink']) sessions = db_helper.get_sessions(user) training_sessions, testing_sessions = split(sessions, 0.8) preprocessor = Preprocessor([]) start = 0 tesiting_sessions = filter(lambda x: len(x) > start, testing_sessions) logs = preprocessor.aggregate_sessions(\ training_sessions + \ map(lambda x: [x[start]], tesiting_sessions)) l = len(logs) - len(tesiting_sessions)
from predictor.mfu_predictor import MFUPredictor from sklearn.naive_bayes import MultinomialNB, BernoulliNB from sklearn.naive_bayes import GaussianNB from sklearn.feature_extraction import DictVectorizer from sklearn import tree def split(sessions, ratio=0.8): split_index = int(len(sessions)*ratio) return sessions[:split_index], sessions[split_index:] import sys if __name__ == '__main__': db_helper = ApeicDBHelper() users = db_helper.get_users() accuracies = [] for user in users: print colored(user, attrs=['blink']) sessions = db_helper.get_sessions(user) training_sessions, testing_sessions = split(sessions, 0.8) preprocessor = Preprocessor([]) start = 0 tesiting_sessions = filter(lambda x: len(x) > start, testing_sessions) logs = preprocessor.aggregate_sessions(\ training_sessions + \ map(lambda x: [x[start]], tesiting_sessions)) l = len(logs) - len(tesiting_sessions)
from sklearn import metrics from sklearn.datasets.samples_generator import make_blobs from sklearn.preprocessing import StandardScaler ############################################################################## # Generate sample data centers = [[1, 1], [-1, -1], [1, -1]] X, labels_true = make_blobs(n_samples=100, centers=centers, cluster_std=0.4, random_state=0) X = StandardScaler().fit_transform(X) from numpy import array from predictor.predictor import Predictor, split db_helper = ApeicDBHelper() for user in db_helper.get_users()[7:]: sessions = db_helper.get_sessions(user) training_logs, testing_logs = split(sessions, aggregated=True) training_logs = filter(lambda x: x['latitude'] != 0 and x['longitude']!= 0, training_logs) latlng_pairs = list(set(map(lambda x: (x['latitude'], x['longitude']), training_logs))) print latlng_pairs print len(latlng_pairs) # X = array(latlng_pairs) # print X.size # print X result = [] for la1, ln1 in latlng_pairs: dists = [] for la2, ln2 in latlng_pairs: