-
Notifications
You must be signed in to change notification settings - Fork 117
/
location_smoothing.py
269 lines (239 loc) · 12.6 KB
/
location_smoothing.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
# Standard imports
import numpy as np
import json
import logging
from dateutil import parser
import math
import pandas as pd
import attrdict as ad
import datetime as pydt
import time as time
import pytz
import geojson as gj
# Our imports
import emission.analysis.point_features as pf
import emission.analysis.intake.cleaning.cleaning_methods.speed_outlier_detection as eaico
import emission.analysis.intake.cleaning.cleaning_methods.jump_smoothing as eaicj
import emission.storage.pipeline_queries as epq
import emission.storage.decorations.analysis_timeseries_queries as esda
import emission.storage.timeseries.abstract_timeseries as esta
import emission.core.wrapper.entry as ecwe
import emission.core.wrapper.metadata as ecwm
import emission.core.wrapper.smoothresults as ecws
import emission.core.common as ecc
import emission.storage.decorations.useful_queries as taug
import emission.storage.decorations.location_queries as lq
import emission.core.get_database as edb
np.set_printoptions(suppress=True)
# This is what we use in the segmentation code to see if the points are "the same"
DEFAULT_SAME_POINT_DISTANCE = 100
def recalc_speed(points_df):
"""
The input dataframe already has "speed" and "distance" columns.
Drop them and recalculate speeds from the first point onwards.
The speed column has the speed between each point and its previous point.
The first row has a speed of zero.
"""
stripped_df = points_df.drop("speed", axis=1).drop("distance", axis=1)
point_list = [ad.AttrDict(row) for row in points_df.to_dict('records')]
zipped_points_list = zip(point_list, point_list[1:])
distances = [pf.calDistance(p1, p2) for (p1, p2) in zipped_points_list]
distances.insert(0, 0)
with_speeds_df = pd.concat([stripped_df, pd.Series(distances, index=points_df.index, name="distance")], axis=1)
speeds = [pf.calSpeed(p1, p2) for (p1, p2) in zipped_points_list]
speeds.insert(0, 0)
with_speeds_df = pd.concat([with_speeds_df, pd.Series(speeds, index=points_df.index, name="speed")], axis=1)
return with_speeds_df
def add_dist_heading_speed(points_df):
# type: (pandas.DataFrame) -> pandas.DataFrame
"""
Returns a new dataframe with an added "speed" column.
The speed column has the speed between each point and its previous point.
The first row has a speed of zero.
"""
point_list = [ad.AttrDict(row) for row in points_df.to_dict('records')]
zipped_points_list = zip(point_list, point_list[1:])
distances = [pf.calDistance(p1, p2) for (p1, p2) in zipped_points_list]
distances.insert(0, 0)
speeds = [pf.calSpeed(p1, p2) for (p1, p2) in zipped_points_list]
speeds.insert(0, 0)
headings = [pf.calHeading(p1, p2) for (p1, p2) in zipped_points_list]
headings.insert(0, 0)
with_distances_df = pd.concat([points_df, pd.Series(distances, name="distance")], axis=1)
with_speeds_df = pd.concat([with_distances_df, pd.Series(speeds, name="speed")], axis=1)
with_headings_df = pd.concat([with_speeds_df, pd.Series(headings, name="heading")], axis=1)
return with_headings_df
def add_heading_change(points_df):
"""
Returns a new dataframe with an added "heading_change" column.
The heading change column has the heading change between this point and the
two points preceding it. The first two rows have a speed of zero.
"""
point_list = [ad.AttrDict(row) for row in points_df.to_dict('records')]
zipped_points_list = zip(point_list, point_list[1:], point_list[2:])
hcs = [pf.calHC(p1, p2, p3) for (p1, p2, p3) in zipped_points_list]
hcs.insert(0, 0)
hcs.insert(1, 0)
with_hcs_df = pd.concat([points_df, pd.Series(hcs, name="heading_change")], axis=1)
return with_hcs_df
def filter_current_sections(user_id):
time_query = epq.get_time_range_for_smoothing(user_id)
try:
sections_to_process = esda.get_entries(esda.RAW_SECTION_KEY, user_id,
time_query)
for section in sections_to_process:
logging.info("^" * 20 + ("Smoothing section %s for user %s" % (section.get_id(), user_id)) + "^" * 20)
filter_jumps(user_id, section.get_id())
if len(sections_to_process) == 0:
# Didn't process anything new so start at the same point next time
last_section_processed = None
else:
last_section_processed = sections_to_process[-1]
epq.mark_smoothing_done(user_id, last_section_processed)
except:
logging.exception("Marking smoothing as failed")
epq.mark_smoothing_failed(user_id)
def filter_jumps(user_id, section_id):
"""
filters out any jumps in the points related to this section and stores a entry that lists the deleted points for
this trip and this section.
:param user_id: the user id to filter the trips for
:param section_id: the section_id to filter the trips for
:return: none. saves an entry with the filtered points into the database.
"""
logging.debug("filter_jumps(%s, %s) called" % (user_id, section_id))
outlier_algo = eaico.BoxplotOutlier()
tq = esda.get_time_query_for_trip_like(esda.RAW_SECTION_KEY, section_id)
ts = esta.TimeSeries.get_time_series(user_id)
section_points_df = ts.get_data_df("background/filtered_location", tq)
is_ios = section_points_df["filter"].dropna().unique().tolist() == ["distance"]
if is_ios:
logging.debug("Found iOS section, filling in gaps with fake data")
section_points_df = _ios_fill_fake_data(section_points_df)
filtering_algo = eaicj.SmoothZigzag(is_ios, DEFAULT_SAME_POINT_DISTANCE)
logging.debug("len(section_points_df) = %s" % len(section_points_df))
points_to_ignore_df = get_points_to_filter(section_points_df, outlier_algo, filtering_algo)
if points_to_ignore_df is None:
# There were no points to delete
return
points_to_ignore_df_filtered = points_to_ignore_df._id.dropna()
logging.debug("after filtering ignored points, %s -> %s" %
(len(points_to_ignore_df), len(points_to_ignore_df_filtered)))
# We shouldn't really filter any fuzzed points because they represent 100m in 60 secs
# but let's actually check for that
# assert len(points_to_ignore_df) == len(points_to_ignore_df_filtered)
deleted_point_id_list = list(points_to_ignore_df_filtered)
logging.debug("deleted %s points" % len(deleted_point_id_list))
filter_result = ecws.Smoothresults()
filter_result.section = section_id
filter_result.deleted_points = deleted_point_id_list
filter_result.outlier_algo = "BoxplotOutlier"
filter_result.filtering_algo = "SmoothZigzag"
result_entry = ecwe.Entry.create_entry(user_id, "analysis/smoothing", filter_result)
ts.insert(result_entry)
def get_points_to_filter(section_points_df, outlier_algo, filtering_algo):
"""
From the incoming dataframe, filter out large jumps using the specified outlier detection algorithm and
the specified filtering algorithm.
:param section_points_df: a dataframe of points for the current section
:param outlier_algo: the algorithm used to detect outliers
:param filtering_algo: the algorithm used to determine which of those outliers need to be filtered
:return: a dataframe of points that need to be stripped, if any.
None if none of them need to be stripped.
"""
with_speeds_df = add_dist_heading_speed(section_points_df)
logging.debug("section_points_df.shape = %s, with_speeds_df.shape = %s" %
(section_points_df.shape, with_speeds_df.shape))
# if filtering algo is none, there's nothing that can use the max speed
if outlier_algo is not None and filtering_algo is not None:
maxSpeed = outlier_algo.get_threshold(with_speeds_df)
# TODO: Is this the best way to do this? Or should I pass this in as an argument to filter?
# Or create an explicit set_speed() method?
# Or pass the outlier_algo as the parameter to the filtering_algo?
filtering_algo.maxSpeed = maxSpeed
logging.debug("maxSpeed = %s" % filtering_algo.maxSpeed)
if filtering_algo is not None:
try:
filtering_algo.filter(with_speeds_df)
to_delete_mask = np.logical_not(filtering_algo.inlier_mask_)
return with_speeds_df[to_delete_mask]
except Exception as e:
logging.info("Caught error %s while processing section, skipping..." % e)
return None
else:
logging.debug("no filtering algo specified, returning None")
return None
def get_filtered_points(section_df, outlier_algo, filtering_algo):
"""
Filter the points that correspond to the section object that is passed in.
The section object is an AttrDict with the startTs and endTs fields.
Returns a filtered df with the index after the initial filter for accuracy
TODO: Switch this to the section wrapper object going forward
TODO: Note that here, we assume that the data has already been chunked into sections.
But really, we need to filter (at least for accuracy) before segmenting in
order to avoid issues like https://github.com/e-mission/e-mission-data-collection/issues/45
"""
with_speeds_df = add_dist_heading_speed(section_df)
# if filtering algo is none, there's nothing that can use the max speed
if outlier_algo is not None and filtering_algo is not None:
maxSpeed = outlier_algo.get_threshold(with_speeds_df)
# TODO: Is this the best way to do this? Or should I pass this in as an argument to filter?
# Or create an explicit set_speed() method?
# Or pass the outlier_algo as the parameter to the filtering_algo?
filtering_algo.maxSpeed = maxSpeed
if filtering_algo is not None:
try:
filtering_algo.filter(with_speeds_df)
return with_speeds_df[filtering_algo.inlier_mask_]
except Exception as e:
logging.info("Caught error %s while processing section, skipping..." % e)
return with_speeds_df
else:
return with_speeds_df
def _ios_fill_fake_data(locs_df):
diff_ts = locs_df.ts.diff()
fill_ends = diff_ts[diff_ts > 60].index.tolist()
if len(fill_ends) == 0:
logging.debug("No large gaps found, no gaps to fill")
return locs_df
else:
logging.debug("Found %s large gaps, filling them all" % len(fill_ends))
filled_df = locs_df
for end in fill_ends:
logging.debug("Found large gap ending at %s, filling it" % end)
assert end > 0
start = end - 1
start_point = locs_df.iloc[start]["loc"]["coordinates"]
end_point = locs_df.iloc[end]["loc"]["coordinates"]
if ecc.calDistance(start_point, end_point) > DEFAULT_SAME_POINT_DISTANCE:
logging.debug("Distance between %s and %s = %s, adding noise is not enough, skipping..." %
(start_point, end_point, ecc.calDistance(start_point, end_point)))
continue
# else
# Design from https://github.com/e-mission/e-mission-server/issues/391#issuecomment-247246781
logging.debug("start = %s, end = %s, generating entries between %s and %s" %
(start, end, locs_df.ts[start], locs_df.ts[end]))
ts_fill = np.arange(locs_df.ts[start] + 60, locs_df.ts[end], 60)
# We only pick entries that are *greater than* 60 apart
assert len(ts_fill) > 0
dist_fill = np.random.uniform(low=0, high=100, size=len(ts_fill))
angle_fill = np.random.uniform(low=0, high=2 * np.pi, size=len(ts_fill))
# Formula from http://gis.stackexchange.com/questions/5821/calculating-latitude-longitude-x-miles-from-point
lat_fill = locs_df.latitude[end] + np.multiply(dist_fill, np.sin(
angle_fill) / 111111)
cl = np.cos(locs_df.latitude[end])
lng_fill = locs_df.longitude[end] + np.multiply(dist_fill, np.cos(
angle_fill) / cl / 111111)
logging.debug("Fill lengths are: dist %s, angle %s, lat %s, lng %s" %
(len(dist_fill), len(angle_fill), len(lat_fill), len(lng_fill)))
# Unsure if this is needed, but lets put it in just in case
loc_fill = [gj.Point(l) for l in zip(lng_fill, lat_fill)]
fill_df = pd.DataFrame(
{"ts": ts_fill, "longitude": lng_fill, "latitude": lat_fill,
"loc": loc_fill})
filled_df = pd.concat([filled_df, fill_df])
sorted_filled_df = filled_df.sort("ts").reset_index()
logging.debug("after filling, returning head = %s, tail = %s" %
(sorted_filled_df[["fmt_time", "ts", "latitude", "longitude", "metadata_write_ts"]].head(),
sorted_filled_df[["fmt_time", "ts", "latitude", "longitude", "metadata_write_ts"]].tail()))
return sorted_filled_df