1# Copyright (C) 2018-2023 Mark McIntyre
2
3import pandas as pd
4
5
6def _findNearDuplicates():
7 """ not yet implemented """
8 cols=['_mjd','_localtime','_lat1','_lng1','_lat2','_lng2','numstats','stations']
9 df = pd.read_parquet('matched/matches-full-2022.parquet.snap', columns=cols)
10 df['l1diff']=df._lat1.diff()*60
11 df['l2diff']=df._lat2.diff()*60
12 df['g1diff']=df._lng1.diff()*60
13 df['g2diff']=df._lng2.diff()*60
14 df['dtdiff']=df._mjd.diff()*86400
15 df['stdiff']=df.numstats.diff()
16 df['bl1diff']=df._lat1.diff(periods=-1)*60
17 df['bl2diff']=df._lat2.diff(periods=-1)*60
18 df['bg1diff']=df._lng1.diff(periods=-1)*60
19 df['bg2diff']=df._lng2.diff(periods=-1)*60
20 df['bdtdiff']=df._mjd.diff(periods=-1)*86400
21 df['bstdiff']=df.numstats.diff(periods=-1)
22 df = df.fillna(999)
23
24 df2 = df[(abs(df.dtdiff) <= 1.0) | (abs(df.bdtdiff) <= 1.0)]
25 df2 = df2[(abs(df2.l1diff) <= 2) | (abs(df2.bl1diff) <= 2)]
26 df2 = df2[(abs(df2.l2diff) <= 2) | (abs(df2.bl2diff) <= 2)]
27 df2 = df2[(abs(df2.g1diff) <= 2) | (abs(df2.bg1diff) <= 2)]
28 df2 = df2[(abs(df2.g2diff) <= 2) | (abs(df2.bg2diff) <= 2)]
29
30 numsame=0
31 for i,rw in df2.iterrows():
32 if df.iloc[i-1].stations != rw.stations:
33 numsame +=1
34 print(rw._localtime)
35 print(f' number with the same cameras: {numsame}')
36 print(f' number with different cameras: {len(df2)-numsame}')