-
Notifications
You must be signed in to change notification settings - Fork 5
/
Copy pathgsdc.py
463 lines (373 loc) · 19.5 KB
/
gsdc.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
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
# -*- coding: utf-8 -*-
# encoding = utf-8
'''
gsdc.py
author：alvin
create dayno: 20210716
Function: Excute all the postprocess.
1. Position Shift
2. Stop Detection and Postprocess
3. Snap to Grid
4. Gaussian Filter
5. Kalman Filter
6. Phone Mean
7. Position Shift
功能: 汇总后处理过程。
1. 位置偏移
2. 停车检测和后处理
3. 路网匹配
4. 高斯滤波
5. 卡尔曼滤波
6. 平均路径
7. 位置偏移
History:
version contributor comment
v1.0 alvin 第一版
v2.0 alvin 第二版
v3.0 alvin 第三版: 加入movingpandas停车检测代码
v4.0 alvin 第四版: 分路段场景调参优化
v5.0 liu 第五版：对2021-04-22-US-SJC-1 GT数据中某交叉路口的双行道进行平滑,
去掉训练集中downtown路段：2021-04-28–US-SJC1
'''
import numpy as np
import pandas as pd
import glob
import os
import matplotlib.pyplot as plt
from tqdm import tqdm
from pathlib import Path
import plotly.express as px
from shapely.geometry import Point
from scipy import spatial
import geopandas as gpd
import optuna
import warnings
warnings.filterwarnings("ignore")
import postprocess_car_stop as pcs
import snap_to_grid as stg
import ro_kf_mean as rkm
import gf_mean as gm
import position_shift as ps
# Load Dataset. 读取数据
data_dir = Path("../data")
trn_df = pd.read_csv(data_dir / "baseline_locations_train.csv")
tst_df = pd.read_csv(data_dir / "baseline_locations_test.csv")
sub_df = pd.read_csv(data_dir / 'sample_submission.csv')
gt_df = pd.DataFrame()
for (collection_name, phone_name), df in pd.read_csv(data_dir / "baseline_locations_train.csv").groupby(["collectionName", "phoneName"]):
path = data_dir / f"train/{collection_name}/{phone_name}/ground_truth.csv"
df = pd.read_csv(path)
gt_df = pd.concat([gt_df, df]).reset_index(drop=True)
gt_df['phone'] = gt_df['collectionName'] + '_' + gt_df['phoneName']
# If you want to repeat again, please open the below comment
# trn_df0 = pd.read_csv(data_dir / "baseline_locations_train.csv")
# trn_df = pd.read_csv("../submit/baseline_locations_train_pcs_stg_gf_kf_mean_shift_sub_0803.csv")
# trn_df = trn_df.merge(trn_df0[['phone', 'millisSinceGpsEpoch', 'heightAboveWgs84EllipsoidM']], on=['phone', 'millisSinceGpsEpoch'], how='left')
# trn_df['collectionName'] = trn_df['phone'].apply(lambda x:x.split('_')[0])
# trn_df['phoneName'] = trn_df['phone'].apply(lambda x:x.split('_')[1])
# tst_df0 = pd.read_csv(data_dir / "baseline_locations_test.csv")
# tst_df = pd.read_csv("../submit/baseline_locations_test_pcs_stg_gf_kf_mean_shift_sub_0803.csv")
# tst_df = tst_df.merge(tst_df0[['phone', 'millisSinceGpsEpoch', 'heightAboveWgs84EllipsoidM']], on=['phone', 'millisSinceGpsEpoch'], how='left')
# tst_df['collectionName'] = tst_df['phone'].apply(lambda x:x.split('_')[0])
# tst_df['phoneName'] = tst_df['phone'].apply(lambda x:x.split('_')[1])
# Visualization
def visualize_trafic(df, center, zoom=10):
df_copy = df.copy()
df_copy['index'] = df_copy.index
fig = px.scatter_mapbox(df_copy,
# Here, plotly gets, (x,y) coordinates
lat="latDeg",
lon="lngDeg",
hover_name='index',
#Here, plotly detects color of series
color="phoneName",
labels="phoneName",
zoom=zoom,
center=center,
height=600,
width=800)
fig.update_layout(mapbox_style='stamen-terrain')
fig.update_layout(margin={"r": 0, "t": 0, "l": 0, "b": 0})
fig.update_layout(title_text="GPS trafic")
fig.show()
def visualize_collection(df, collection):
target_df = df[df['collectionName']==collection].copy()
lat_center = target_df['latDeg'].mean()
lng_center = target_df['lngDeg'].mean()
center = {"lat":lat_center, "lon":lng_center}
visualize_trafic(target_df, center)
##############################
# Parameter setting 参数设置
##############################
save_dir = Path("../submit")
save_ver = '0803v2' # 版本号
stop_dist_threshold = 3
stop_window_size = 3
stop_verbose = True
trn_tgt_cols = ['2021-04-22-US-SJC-1', '2021-04-28-US-SJC-1', '2021-04-29-US-SJC-2']
tst_tgt_cols = ['2021-04-22-US-SJC-2', '2021-04-29-US-SJC-3']
grid_cols = '2021-04-22-US-SJC-1'
grid_phone = 'Pixel4'
dist_th = 50
vel_th = 18
window_size = 3
window_thresh = 2
sort_window_size = 20
# Since 2021-04-22-US-SJC-1 is used to be the grid, we don't add it into the train dataset otherwise it will overfit.
# Since 2021-04-28-US-SJC-1 has much more noise, we don't add it into the train dataset.
trn_col_cat_dict = {
'2020-05-14-US-MTV-1':'highway',
'2020-05-14-US-MTV-2':'highway',
'2020-05-21-US-MTV-1':'highway',
'2020-05-21-US-MTV-2':'highway',
'2020-05-29-US-MTV-1':'highway',
'2020-05-29-US-MTV-2':'highway',
'2020-06-04-US-MTV-1':'highway',
'2020-06-05-US-MTV-1':'highway',
'2020-06-05-US-MTV-2':'highway',
'2020-06-11-US-MTV-1':'highway',
'2020-07-08-US-MTV-1':'highway',
'2020-07-17-US-MTV-1':'highway',
'2020-07-17-US-MTV-2':'highway',
'2020-08-03-US-MTV-1':'highway',
'2020-08-06-US-MTV-2':'highway',
'2020-09-04-US-SF-1':'highway',
'2020-09-04-US-SF-2':'highway',
'2021-01-04-US-RWC-1':'highway',
'2021-01-04-US-RWC-2':'highway',
'2021-01-05-US-SVL-1':'highway',
'2021-01-05-US-SVL-2':'highway',
'2021-03-10-US-SVL-1':'street',
'2021-04-15-US-MTV-1':'street',
# '2021-04-22-US-SJC-1':'downtown',
'2021-04-26-US-SVL-1':'street',
'2021-04-28-US-MTV-1':'street',
# '2021-04-28-US-SJC-1':'downtown',
'2021-04-29-US-MTV-1':'street',
'2021-04-29-US-SJC-2':'downtown'
}
tst_col_cat_dict = {
'2020-05-15-US-MTV-1':'highway',
'2020-05-28-US-MTV-1':'highway',
'2020-05-28-US-MTV-2':'highway',
'2020-06-04-US-MTV-2':'highway',
'2020-06-10-US-MTV-1':'highway',
'2020-06-10-US-MTV-2':'highway',
'2020-08-03-US-MTV-2':'highway',
'2020-08-13-US-MTV-1':'highway',
'2021-03-16-US-MTV-2':'highway',
'2021-03-16-US-RWC-2':'street',
'2021-03-25-US-PAO-1':'street',
'2021-04-02-US-SJC-1':'street',
'2021-04-08-US-MTV-1':'street',
'2021-04-21-US-MTV-1':'street',
'2021-04-22-US-SJC-2':'downtown',
'2021-04-26-US-SVL-2':'street',
'2021-04-28-US-MTV-2':'street',
'2021-04-29-US-MTV-2':'street',
'2021-04-29-US-SJC-3':'downtown'
}
##############################
# 1. Position Shift 位置偏移
##############################
def objective(trial):
a = trial.suggest_uniform('a', -1, 1)
trn_shift0 = ps.position_shift(trn_df, a)
score, scores = ps.compute_dist(trn_shift0, gt_df)
return score
def find_best_param(n_trials):
study = optuna.create_study()
study.optimize(objective, n_trials)
return study.best_params
best_param = find_best_param(25)
print('The best params (Position Shift): ', best_param['a'])
trn_shift = ps.position_shift(trn_df, best_param['a'])
tst_shift = ps.position_shift(tst_df, best_param['a'])
# Collect the result of different road types. 收集各路段类型的结果
trn_sub = []
tst_sub = []
trn_df_copy = trn_shift.copy()
tst_df_copy = tst_shift.copy()
ground_truth_finetuned = gt_df.copy()
ground_truth = gt_df.copy()
# GT Smoothing Manually. 手动平滑GT路网
# 2021-04-22-US-SJC-1
# pixel4
ground_truth_finetuned['latDeg'].iloc[103937 + 2446] = ground_truth['latDeg'].iloc[103937 + 2340]
ground_truth_finetuned['lngDeg'].iloc[103937 + 2446] = ground_truth['lngDeg'].iloc[103937 + 2340]
ground_truth_finetuned['latDeg'].iloc[103937 + 2445] = ground_truth['latDeg'].iloc[103937 + 2339]
ground_truth_finetuned['lngDeg'].iloc[103937 + 2445] = ground_truth['lngDeg'].iloc[103937 + 2339]
ground_truth_finetuned['latDeg'].iloc[103937 + 2444] = ground_truth['latDeg'].iloc[103937 + 2338]
ground_truth_finetuned['lngDeg'].iloc[103937 + 2444] = ground_truth['lngDeg'].iloc[103937 + 2338]
ground_truth_finetuned['latDeg'].iloc[103937 + 2443] = ground_truth['latDeg'].iloc[103937 + 2333]
ground_truth_finetuned['lngDeg'].iloc[103937 + 2443] = ground_truth['lngDeg'].iloc[103937 + 2333]
ground_truth_finetuned['latDeg'].iloc[103937 + 2442] = ground_truth['latDeg'].iloc[103937 + 2331]
ground_truth_finetuned['lngDeg'].iloc[103937 + 2442] = ground_truth['lngDeg'].iloc[103937 + 2331]
ground_truth_finetuned['latDeg'].iloc[103937 + 2441] = ground_truth['latDeg'].iloc[103937 + 2329]
ground_truth_finetuned['lngDeg'].iloc[103937 + 2441] = ground_truth['lngDeg'].iloc[103937 + 2329]
ground_truth_finetuned['latDeg'].iloc[103937 + 2440] = ground_truth['latDeg'].iloc[103937 + 2322]
ground_truth_finetuned['lngDeg'].iloc[103937 + 2440] = ground_truth['lngDeg'].iloc[103937 + 2322]
ground_truth_finetuned['latDeg'].iloc[103937 + 2439] = ground_truth['latDeg'].iloc[103937 + 2320]
ground_truth_finetuned['lngDeg'].iloc[103937 + 2439] = ground_truth['lngDeg'].iloc[103937 + 2320]
# Samsung
ground_truth_finetuned['latDeg'].iloc[106827 + 2446] = ground_truth['latDeg'].iloc[106827 + 2340]
ground_truth_finetuned['lngDeg'].iloc[106827 + 2446] = ground_truth['lngDeg'].iloc[106827 + 2340]
ground_truth_finetuned['latDeg'].iloc[106827 + 2445] = ground_truth['latDeg'].iloc[106827 + 2339]
ground_truth_finetuned['lngDeg'].iloc[106827 + 2445] = ground_truth['lngDeg'].iloc[106827 + 2339]
ground_truth_finetuned['latDeg'].iloc[106827 + 2444] = ground_truth['latDeg'].iloc[106827 + 2338]
ground_truth_finetuned['lngDeg'].iloc[106827 + 2444] = ground_truth['lngDeg'].iloc[106827 + 2338]
ground_truth_finetuned['latDeg'].iloc[106827 + 2443] = ground_truth['latDeg'].iloc[106827 + 2333]
ground_truth_finetuned['lngDeg'].iloc[106827 + 2443] = ground_truth['lngDeg'].iloc[106827 + 2333]
ground_truth_finetuned['latDeg'].iloc[106827 + 2442] = ground_truth['latDeg'].iloc[106827 + 2331]
ground_truth_finetuned['lngDeg'].iloc[106827 + 2442] = ground_truth['lngDeg'].iloc[106827 + 2331]
ground_truth_finetuned['latDeg'].iloc[106827 + 2441] = ground_truth['latDeg'].iloc[106827 + 2329]
ground_truth_finetuned['lngDeg'].iloc[106827 + 2441] = ground_truth['lngDeg'].iloc[106827 + 2329]
ground_truth_finetuned['latDeg'].iloc[106827 + 2440] = ground_truth['latDeg'].iloc[106827 + 2322]
ground_truth_finetuned['lngDeg'].iloc[106827 + 2440] = ground_truth['lngDeg'].iloc[106827 + 2322]
ground_truth_finetuned['latDeg'].iloc[106827 + 2439] = ground_truth['latDeg'].iloc[106827 + 2320]
ground_truth_finetuned['lngDeg'].iloc[106827 + 2439] = ground_truth['lngDeg'].iloc[106827 + 2320]
gt_df_copy = ground_truth_finetuned.copy()
for col_type in ['highway', 'street', 'downtown']:
print("Now, let's postprocess the road type：", col_type)
# Get the collections under the given road type. 收集给定col_type下的collectionName
tgt_trn_cols = []
for col_key, col_val in trn_col_cat_dict.items():
if col_val == col_type:
tgt_trn_cols.append(col_key)
tgt_tst_cols = []
for col_key, col_val in tst_col_cat_dict.items():
if col_val == col_type:
tgt_tst_cols.append(col_key)
# Get the dataset under the given road type. 获取特定的数据集
trn_df = trn_df_copy[trn_df_copy['collectionName'].isin(tgt_trn_cols)]
tst_df = tst_df_copy[tst_df_copy['collectionName'].isin(tgt_tst_cols)]
gt_df = gt_df_copy.copy()
trn_df = trn_df.reset_index(drop = True)
tst_df = tst_df.reset_index(drop = True)
gt_df = gt_df.reset_index(drop = True)
##############################
# 2. Stop Detection and Postprocess 停车检测和后处理
##############################
def objective(trial):
stop_window_size = trial.suggest_int('stop_window_size', 2, 5)
stop_dist_threshold = trial.suggest_int('stop_dist_threshold', 1, 10)
tmp_trn_df2 = pcs.process_car_stop_for_trn(trn_df, stop_dist_threshold, stop_window_size, stop_verbose)
score, scores = gm.compute_dist(tmp_trn_df2, gt_df)
return score
def pcs_find_best_param(n_trials):
study = optuna.create_study()
study.optimize(objective, n_trials)
return study.best_params
best_param = pcs_find_best_param(10)
print('The best params (Stop Detection and Postprocess): ', best_param)
trn_df2 = pcs.process_car_stop_for_trn(trn_df, best_param['stop_dist_threshold'], best_param['stop_window_size'], stop_verbose)
tst_df1 = pcs.process_car_park_for_tst(tst_df, stop_verbose)
tst_df2 = pcs.process_car_stop_for_tst(tst_df1, best_param['stop_dist_threshold'], best_param['stop_window_size'])
trn_df2.to_csv(save_dir / "baseline_locations_train_pcs_{}_{}.csv".format(col_type, save_ver), index=False)
tst_df2.to_csv(save_dir / "baseline_locations_test_pcs_{}_{}.csv".format(col_type, save_ver), index=False)
##############################
# 3. Snap to Grid 路网匹配
##############################
def objective(trial):
dist_th = trial.suggest_int('dist_th', 40, 60)
vel_th = trial.suggest_int('vel_th', 15, 30)
window_th = trial.suggest_int('window_th', 1, 5)
window_size = trial.suggest_int('window_size', 1, 5)
tmp_trn_df3 = stg.snap_to_grid_for_trn(trn_df2, gt_df, trn_tgt_cols, grid_cols, grid_phone, dist_th, vel_th, window_th, window_size)
score, scores = gm.compute_dist(tmp_trn_df3, gt_df)
return score
def stg_find_best_param(n_trials):
study = optuna.create_study()
study.optimize(objective, n_trials)
return study.best_params
best_param = {'dist_th':45, 'vel_th':22, 'window_size':3, 'window_th':2}
print('The best params (Snap to Grid): ', best_param)
trn_df3 = stg.snap_to_grid_for_trn(trn_df2, gt_df, trn_tgt_cols, grid_cols, grid_phone, best_param['dist_th'], best_param['vel_th'], best_param['window_th'], best_param['window_size'])
tst_df3 = stg.snap_to_grid_for_tst(tst_df2, gt_df, tst_tgt_cols, grid_cols, grid_phone, best_param['dist_th'], best_param['vel_th'], best_param['window_th'], best_param['window_size'])
trn_df3.to_csv(save_dir / "baseline_locations_train_pcs_stg_{}_{}.csv".format(col_type, save_ver), index=False)
tst_df3.to_csv(save_dir / "baseline_locations_test_pcs_stg_{}_{}.csv".format(col_type, save_ver), index=False)
##############################
# 4-6. Gaussian Filter + Kalman Filter + Phone Mean 高斯滤波 + 卡尔曼滤波 + 平均路径
##############################
trn_ro = trn_df3.copy()
tst_ro = tst_df3.copy()
# Gaussian Filter + Kalman Filter. 高斯滤波 + 卡尔曼滤波
def objective(trial):
sz_1 = trial.suggest_uniform('sz_1', 0.2, 0.8)
sz_2 = trial.suggest_uniform('sz_2', 5, 7)
sz_crit = trial.suggest_uniform('sz_crit', 0.5, 1.5)
T = trial.suggest_uniform('T', 1, 3)
size = trial.suggest_int('size', 4, 8)
noise = trial.suggest_uniform('noise', 1e-06, 3e-06)
obs_noise = trial.suggest_uniform('obs_noise', 1e-06, 3e-06)
# GF
gf_df = gm.apply_gauss_smoothing(trn_ro, {'sz_1' : sz_1, 'sz_2' :sz_2, 'sz_crit' : sz_crit})
# KF
tmp_kf = rkm.make_kalman_filter(T, size, noise, obs_noise)
trn_gf_kf = rkm.apply_kf_smoothing(gf_df, tmp_kf)
mean_df = gm.mean_with_other_phones(trn_gf_kf)
score, scores = gm.compute_dist(mean_df, gt_df)
return score.values[0]
def gf_find_best_param(n_trials):
study = optuna.create_study()
study.optimize(objective, n_trials)
return study.best_params
best_param = gf_find_best_param(30)
print('The best params (Gaussian + KF + Lerp + Mean): ', best_param)
trn_gf = gm.apply_gauss_smoothing(trn_ro, {'sz_1':best_param['sz_1'],
'sz_2':best_param['sz_2'],
'sz_crit':best_param['sz_crit']})
best_kf = rkm.make_kalman_filter(best_param['T'], best_param['size'], best_param['noise'], best_param['obs_noise'])
trn_gf_kf = rkm.apply_kf_smoothing(trn_gf, best_kf)
trn_mean = gm.mean_with_other_phones(trn_gf_kf)
tst_gf = gm.apply_gauss_smoothing(tst_ro, {'sz_1':best_param['sz_1'],
'sz_2':best_param['sz_2'],
'sz_crit':best_param['sz_crit']})
tst_gf_kf = rkm.apply_kf_smoothing(tst_gf, best_kf)
tst_mean = gm.mean_with_other_phones(tst_gf_kf)
trn_mean.to_csv(save_dir / "baseline_locations_train_pcs_stg_gf_kf_mean_{}_{}.csv".format(col_type, save_ver), index=False)
tst_mean.to_csv(save_dir / "baseline_locations_test_pcs_stg_gf_kf_mean_{}_{}.csv".format(col_type, save_ver), index=False)
##############################
# 7. Position Shift 位置偏移
##############################
# Add heightAboveWgs84EllipsoidM. 加入大地高特征
trn_mean = trn_mean.merge(trn_df[['collectionName', 'phoneName', 'millisSinceGpsEpoch', 'heightAboveWgs84EllipsoidM']], on=['collectionName', 'phoneName', 'millisSinceGpsEpoch'], how='left')
tst_mean = tst_mean.merge(tst_df[['collectionName', 'phoneName', 'millisSinceGpsEpoch', 'heightAboveWgs84EllipsoidM']], on=['collectionName', 'phoneName', 'millisSinceGpsEpoch'], how='left')
def objective(trial):
a = trial.suggest_uniform('a', -1, 1)
trn_shift = ps.position_shift(trn_mean, a)
score, scores = ps.compute_dist(trn_shift, gt_df)
return score
def find_best_param(n_trials):
study = optuna.create_study()
study.optimize(objective, n_trials)
return study.best_params
best_param = find_best_param(25)
print('The best params (Position Shift): ', best_param['a'])
trn_shift = ps.position_shift(trn_mean, best_param['a'])
tst_shift = ps.position_shift(tst_mean, best_param['a'])
# Prepare Submission. 准备提交结果
trn_shift['phone'] = trn_shift['collectionName'] + '_' + trn_shift['phoneName']
tst_shift['phone'] = tst_shift['collectionName'] + '_' + tst_shift['phoneName']
trn_shift = trn_shift[['phone', 'millisSinceGpsEpoch', 'latDeg', 'lngDeg']]
tst_shift = tst_shift[['phone', 'millisSinceGpsEpoch', 'latDeg', 'lngDeg']]
# Save Submission.保存提交结果
trn_shift.to_csv(save_dir / "baseline_locations_train_pcs_stg_gf_kf_mean_shift_{}_{}.csv".format(col_type, save_ver), index=False)
tst_shift.to_csv(save_dir / "baseline_locations_test_pcs_stg_gf_kf_mean_shift_{}_{}.csv".format(col_type, save_ver), index=False)
trn_sub.append(trn_shift)
tst_sub.append(tst_shift)
tst_shift['collectionName'] = tst_shift['phone'].apply(lambda x: x.split('_')[0])
tst_shift['phoneName'] = tst_shift['phone'].apply(lambda x: x.split('_')[1])
# Concatenate the result of different road types. 合并各个路段类型的处理结果
trn_sub = pd.concat(trn_sub, axis = 0)
tst_sub = pd.concat(tst_sub, axis = 0)
trn_sub = trn_sub.reset_index(drop = True)
tst_sub = tst_sub.reset_index(drop = True)
sub_df = sub_df[['phone', 'millisSinceGpsEpoch']].merge(tst_sub, on=['phone', 'millisSinceGpsEpoch'], how='left')
sub_df.drop(['collectionName', 'phoneName'], axis = 1, inplace = True)
trn_sub.to_csv(save_dir / "baseline_locations_train_pcs_stg_gf_kf_mean_shift_sub_{}.csv".format(save_ver), index=False)
sub_df.to_csv(save_dir / "baseline_locations_test_pcs_stg_gf_kf_mean_shift_sub_{}.csv".format(save_ver), index=False)
# Print score
score, scores = gm.compute_dist(trn_sub, gt_df_copy)
print("Score：", score)
print(scores)