XY

 1 import pandas as pd
 2 import copy
 3 
 4 class BenchMark:
 5     def __init__(self):
 6         self.MIN = 10000
 7         self.data = 0
 8     def Reset(self):
 9         self.MIN = 10000
10         self.data = 0
11 
12 dictCounts = {}
13 dictTop10_D2C = {}
14 BENCH_MARK = BenchMark()
15 LAST_BENCH_MARK = BenchMark()
16 run_count1 = 0
17 run_count2 = 0
18 
19 def FindTop10(data):
20     global BENCH_MARK, LAST_BENCH_MARK,run_count1,run_count2
21     if(data in dictCounts):
22         dictCounts[data] += 1
23     else:
24         dictCounts[data] = 1
25 
26     temp = dictCounts[data]
27     
28     #just record run times
29     run_count1 += 1
30     
31     if LAST_BENCH_MARK.MIN != 10000 and temp< LAST_BENCH_MARK.MIN:
32         return
33 
34     dictTop10_D2C[data] = temp
35 
36     if len(dictTop10_D2C)>10:
37         BENCH_MARK.Reset()
38         for item in dictTop10_D2C:
39             
40             #just record run times
41             run_count2+=1
42             
43             if dictTop10_D2C[item] < BENCH_MARK.MIN:
44                 BENCH_MARK.MIN = dictTop10_D2C[item]
45                 BENCH_MARK.data = item
46         LAST_BENCH_MARK = copy.deepcopy(BENCH_MARK)
47         dictTop10_D2C.pop(BENCH_MARK.data)
48 
49 def PrintData2Count(aDict):
50     for key in aDict:
51         print('%.1f:%d' % (key, aDict[key]))
52 
53 if __name__ == '__main__':
54     df = pd.read_csv('D:/data/ctp_data/rb/201709/rb1801_20170905.csv')
55     for data in df['LastPx']:
56         FindTop10(data)
57 
58     PrintData2Count(dictCounts)
59     print("==============dictCounts length:", len(dictCounts))
60     PrintData2Count(dictTop10_D2C)
61 
62     print("run_count1:%d,run_count2:%d" %(run_count1,run_count2))
63