Python Pandas:使用地图按范围创建组 [英] Python Pandas: Create Groups by Range using map

问题描述

我有一个大数据集，我希望根据总数的累计百分比来创建组.我已经通过使用 map 函数完成了这个工作，请参见下面的代码.如果我想让我的小组更加细化，有没有更好的方法来做到这一点?例如，现在我正在查看 5% 的增量……如果想要查看 1% 的增量怎么办.想知道是否有另一种方法可以让我不必将它们明确输入到我的codethem"函数中.

I have a large data set where I am looking to create groups based upon cumulative sum percent of the total. I have gotten this to work by using the map function see below code. Is there a better way to do this say if I wanted to make my groups even more granular? So for example now am looking at 5% increments...what if want to look at 1 % increments. Wondering if there is another way where I don't have to explicitly enter them into my "codethem" function.

def codethem(dl):
if  dl < .05 : return '5'
elif .05 < dl <= .1: return '10'
elif .1 < dl <= .15: return '15'
elif .15 < dl <= .2: return '20'
elif .2 < dl <= .25: return '25'
elif .25 < dl <= .3: return '30'
elif  .3 < dl <= .35: return '35'
elif .35 < dl <= .4: return '40'
elif .4 < dl <= .45: return '45'
elif .45 < dl <= .5: return '50'
elif .5 < dl <= .55: return '55'
elif .55 < dl <= .6: return '60'
elif .6 < dl <= .65: return '65'
elif .65 < dl <= .7: return '70'
elif .7 < dl <= .75: return '75'
elif .75 < dl <= .8: return '80'
elif  .8 < dl <= .85: return '85'
elif .85 < dl <= .9: return '90'
elif .9 < dl <= .95: return '95'
elif .95 < dl <= 1: return '100'
else: return 'None'

my_df['code'] = my_df['sales_csum_aspercent'].map(code them)

谢谢！

推荐答案

有一个特殊的方法 - pd.cut()

there is a special method for that - pd.cut()

演示:

创建随机DF:

In [393]: df = pd.DataFrame({'a': np.random.rand(10)})

In [394]: df
Out[394]:
          a
0  0.860256
1  0.399267
2  0.209185
3  0.773647
4  0.294845
5  0.883161
6  0.985758
7  0.559730
8  0.723033
9  0.126226

我们应该在调用pd.cut()时指定bins:

we should specify bins when calling pd.cut():

In [404]: np.linspace(0, 1, 11)
Out[404]: array([ 0. ,  0.1,  0.2,  0.3,  0.4,  0.5,  0.6,  0.7,  0.8,  0.9,  1. ])

In [395]: pd.cut(df.a, bins=np.linspace(0, 1, 11))
Out[395]:
0    (0.8, 0.9]
1    (0.3, 0.4]
2    (0.2, 0.3]
3    (0.7, 0.8]
4    (0.2, 0.3]
5    (0.8, 0.9]
6      (0.9, 1]
7    (0.5, 0.6]
8    (0.7, 0.8]
9    (0.1, 0.2]
Name: a, dtype: category
Categories (10, object): [(0, 0.1] < (0.1, 0.2] < (0.2, 0.3] < (0.3, 0.4] ... (0.6, 0.7] < (0.7, 0.8] < (0.8, 0.9] < (0.9, 1]]

如果我们想要自定义标签，我们应该明确指定它们:

if we want to have a custom labels, we should explicitly specify them:

In [401]: bins = np.linspace(0,1, 11)

注意:bin 标签必须比 bin 边的数量少 1

NOTE: bin labels must be one fewer than the number of bin edges

In [402]: labels = (bins[1:]*100).astype(int)

In [412]: labels
Out[412]: array([ 10,  20,  30,  40,  50,  60,  70,  80,  90, 100])

In [403]: pd.cut(df.a, bins=bins, labels=labels)
Out[403]:
0     90
1     40
2     30
3     80
4     30
5     90
6    100
7     60
8     80
9     20
Name: a, dtype: category
Categories (10, int64): [10 < 20 < 30 < 40 ... 70 < 80 < 90 < 100]

让我们用 5% 步骤来做

In [419]: bins = np.linspace(0, 1, 21)

In [420]: bins
Out[420]: array([ 0.  ,  0.05,  0.1 ,  0.15,  0.2 ,  0.25,  0.3 ,  0.35,  0.4 ,  0.45,  0.5 ,  0.55,  0.6 ,  0.65,  0.7 ,  0.75,  0.8 ,  0.8
5,  0.9 ,  0.95,  1.  ])

In [421]: labels = (bins[1:]*100).astype(int)

In [422]: labels
Out[422]: array([  5,  10,  15,  20,  25,  30,  35,  40,  45,  50,  55,  60,  65,  70,  75,  80,  85,  90,  95, 100])

In [423]: pd.cut(df.a, bins=bins, labels=labels)
Out[423]:
0     90
1     40
2     25
3     80
4     30
5     90
6    100
7     60
8     75
9     15
Name: a, dtype: category
Categories (20, int64): [5 < 10 < 15 < 20 ... 85 < 90 < 95 < 100]

这篇关于Python Pandas:使用地图按范围创建组的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！