def setup_method(self):
latitude = constants.LATITUDE
longitude = constants.LONGITUDE
date_time = constants.DATETIME
user_id = constants.UID
lat_lons = np.array(
[
[43.8430139, 10.5079940],
[43.5442700, 10.3261500],
[43.7085300, 10.4036000],
[43.7792500, 11.2462600],
[43.8430139, 10.5079940],
[43.7085300, 10.4036000],
[43.8430139, 10.5079940],
[43.5442700, 10.3261500],
[43.5442700, 10.3261500],
[43.7085300, 10.4036000],
[43.8430139, 10.5079940],
[43.7792500, 11.2462600],
[43.7085300, 10.4036000],
[43.5442700, 10.3261500],
[43.7792500, 11.2462600],
[43.7085300, 10.4036000],
[43.7792500, 11.2462600],
[43.8430139, 10.5079940],
[43.8430139, 10.5079940],
[43.5442700, 10.3261500],
]
)
traj = pd.DataFrame(lat_lons, columns=[latitude, longitude])
traj[date_time] = pd.to_datetime(
[
"20110203 8:34:04",
"20110203 9:34:04",
"20110203 10:34:04",
"20110204 10:34:04",
"20110203 8:34:04",
"20110203 9:34:04",
"20110204 10:34:04",
"20110204 11:34:04",
"20110203 8:34:04",
"20110203 9:34:04",
"20110204 10:34:04",
"20110204 11:34:04",
"20110204 10:34:04",
"20110204 11:34:04",
"20110204 12:34:04",
"20110204 10:34:04",
"20110204 11:34:04",
"20110205 12:34:04",
"20110204 10:34:04",
"20110204 11:34:04",
]
)
traj[user_id] = (
[1 for _ in range(4)]
+ [2 for _ in range(4)]
+ [3 for _ in range(4)]
+ [4 for _ in range(3)]
+ [5 for _ in range(3)]
+ [6 for _ in range(2)]
)
self.unique_points = [
(43.544270, 10.326150),
(43.708530, 10.403600),
(43.779250, 11.246260),
(43.843014, 10.507994),
]
self.traj = traj.sort_values([user_id, date_time])
self.trjdat = TrajDataFrame(traj, user_id=user_id)
[43.7085300, 10.4036000],
[43.5442700, 10.3261500],
[43.7085300, 10.4036000],
[43.5442700, 10.3261500],
[43.7085300, 10.4036000],
[43.5442700, 10.3261500]])
trj2 = pd.DataFrame(lat_lons_2, columns=[latitude, longitude])
trj2[date_time] = pd.to_datetime(['20110203 8:34:04' for _ in range(36)])
trj2[user_id] = [1 for _ in range(11)] + [2 for _ in range(4)] + \
[3 for _ in range(9)] + [4 for _ in range(12)]
traj = traj.sort_values([user_id, date_time])
trjdat = TrajDataFrame(traj, user_id=user_id)
trjdat_2 = TrajDataFrame(trj2, user_id=user_id)
trj_freq = frequency_vector(trjdat_2)
trj_prob = probability_vector(trjdat_2)
first_instance = trjdat[:2].values
second_instance = pd.concat([trjdat[0:1], trjdat[3:4]]).values
third_instance = trj_freq[1:3].values
fourth_instance = trj_prob[1:3].values
@pytest.mark.parametrize('traj,instance,output',
[(trjdat, first_instance, 1.0 / 4.0), (trjdat, second_instance, 1.0 / 3.0)])
def test_location_match(traj, instance, output):
at = attacks.LocationAttack(knowledge_length=1)
results = []
def test_compress(self):
output = compression.compress(self.trjdat)
expected = self.trjdat.copy()
output.reset_index(inplace=True)
output.drop(columns=["index"], inplace=True)
expected.reset_index(inplace=True)
expected.drop(columns=["index"], inplace=True)
# assert
pd.testing.assert_frame_equal(output, expected)
output = compression.compress(self.trjdat, spatial_radius_km=20)
expected = TrajDataFrame(
{
"lat": {
"0": 43.8430139,
"1": 43.6264,
"2": 43.77925,
"3": 43.8430139,
"4": 43.54427,
"5": 43.6264,
"6": 43.8430139,
"7": 43.77925,
"8": 43.6264,
"9": 43.77925,
"10": 43.70853,
"11": 43.77925,
"12": 43.8430139,
"13": 43.8430139,
"14": 43.54427,
},
"lng": {
"0": 10.507994,
"1": 10.364875,
"2": 11.24626,
"3": 10.507994,
"4": 10.32615,
"5": 10.364875,
"6": 10.507994,
"7": 11.24626,
"8": 10.364875,
"9": 11.24626,
"10": 10.4036,
"11": 11.24626,
"12": 10.507994,
"13": 10.507994,
"14": 10.32615,
},
"datetime": {
"0": 1296722044,
"1": 1296725644,
"2": 1296815644,
"3": 1296722044,
"4": 1296819244,
"5": 1296722044,
"6": 1296815644,
"7": 1296819244,
"8": 1296815644,
"9": 1296822844,
"10": 1296815644,
"11": 1296819244,
"12": 1296909244,
"13": 1296815644,
"14": 1296819244,
},
"uid": {
"0": 1,
"1": 1,
"2": 1,
"3": 2,
"4": 2,
"5": 3,
"6": 3,
"7": 3,
"8": 4,
"9": 4,
"10": 5,
"11": 5,
"12": 5,
"13": 6,
"14": 6,
},
},
timestamp=True,
)
output.reset_index(inplace=True)
output.drop(columns=["index"], inplace=True)
expected.reset_index(inplace=True)
expected.drop(columns=["index"], inplace=True)
# assert
pd.testing.assert_frame_equal(output, expected, check_dtype=False)