Geek Answers Handbook

Why use pandas qcut return ValueError: bit edges must be unique?

I have a dataset:

recency;frequency;monetary
21;156;41879955
13;88;16850284
8;74;79150488
2;74;26733719
9;55;16162365
...;...;...

detailed data -> http://pastebin.com/beiEeS80 and I put in DataFrame, and here is my full code:

df = pd.DataFrame(datas, columns=['userid', 'recency', 'frequency', 'monetary'])
df['recency'] = df['recency'].astype(float)
df['frequency'] = df['frequency'].astype(float)
df['monetary'] = df['monetary'].astype(float)

df['recency'] = pd.qcut(df['recency'].values, 5).codes + 1
df['frequency'] = pd.qcut(df['frequency'].values, 5).codes + 1
df['monetary'] = pd.qcut(df['monetary'].values, 5).codes + 1

but it returns an error

df['frequency'] = pd.qcut(df['frequency'].values, 5).codes + 1
ValueError: Bin edges must be unique: array([   1.,    1.,    2.,    4.,    9.,  156.])

How to solve this?

+4
source share
2 answers

I ran this in Jupyter and put the exampledata.txt file in the same directory as the laptop.

Note that the first line:

df = pd.DataFrame(datas, columns=['userid', 'recency', 'frequency', 'monetary'])

loads colors 'userid'when it is not defined in the data file. I deleted this column name.

Decision

import pandas as pd

def pct_rank_qcut(series, n):
    edges = pd.Series([float(i) / n for i in range(n + 1)])
    f = lambda x: (edges >= x).argmax()
    return series.rank(pct=1).apply(f)

datas = pd.read_csv('./exampledata.txt', delimiter=';')

df = pd.DataFrame(datas, columns=['recency', 'frequency', 'monetary'])

df['recency'] = df['recency'].astype(float)
df['frequency'] = df['frequency'].astype(float)
df['monetary'] = df['monetary'].astype(float)

df['recency'] = pct_rank_qcut(df.recency, 5)
df['frequency'] = pct_rank_qcut(df.frequency, 5)
df['monetary'] = pct_rank_qcut(df.monetary, 5)

Explanation

, , pd.qcut, 5 . , , 'frequency' 28% 1. qcut.

pct_rank_qcut, 1 .

    edges = pd.Series([float(i) / n for i in range(n + 1)])

, n. n = 5 [0.0, 0.2, 0.4, 0.6, 0.8, 1.0]

    f = lambda x: (edges >= x).argmax()

, . edges >= x , , edges, True False , x . x = 0.14 (edges >= x) [False, True, True, True, True, True]. argmax(), , True, 1.

    return series.rank(pct=1).apply(f)

series . , , apply(f). , , 1 n. - , :

pd.qcut(df['recency'].values, 5).codes + 1

, bin 1 2. - . , .

print df.head()

   recency  frequency  monetary
0        3          5         5
1        2          5         5
2        2          5         5
3        1          5         5
4        2          5         5

Update

pd.Series.argmax() . pd.Series.values.argmax()() !

def pct_rank_qcut(series, n):
    edges = pd.Series([float(i) / n for i in range(n + 1)])
    f = lambda x: (edges >= x).values.argmax()
    return series.rank(pct=1).apply(f)
+6

, :

pandas, >= 0.20.0, duplicates = 'raise' | 'drop', , , , , ( ), .

pandas :

pd.qcut(df['frequency'].rank(method='first').values, 5).codes + 1

, . ( , , , , pandas.cut, , pandas.qcut , )

+1

More articles:


All Articles