4 Pandas--重复值&nan值 清洗

import pandas as pd
from pandas import DataFrame
import numpy as np

处理丢失数据

• 有两种丢失数据：

  • None
  • np.nan(NaN)

• 两种丢失数据的区别

type(None)

NoneType

type(np.nan)

float

• 为什么在数据分析中需要用到的是浮点类型的空而不是对象类型？
  • 数据分析中会常常使用某些形式的运算来处理原始数据，如果原数数据中的空值为NAN的形式，则不会干扰或者中断运算。
  • NAN可以参与运算的
  • None是不可以参与运算

np.nan + 1

nan

None + 1

---------------------------------------------------------------------------

TypeError                                 Traceback (most recent call last)

<ipython-input-5-3fd8740bf8ab> in <module>
----> 1 None + 1


TypeError: unsupported operand type(s) for +: 'NoneType' and 'int'

• 在pandas中如果遇到了None形式的空值则pandas会将其强转成NAN的形式。

df = DataFrame(data=np.random.randint(0,100,size=(7,5)))
df.iloc[2,3] = None
df.iloc[4,2] = np.nan
df.iloc[5,4] = None
df

	0	1	2	3	4
0	23	67	72.0	72.0	78.0
1	37	26	28.0	44.0	19.0
2	0	69	90.0	NaN	28.0
3	76	67	15.0	15.0	50.0
4	54	52	NaN	69.0	53.0
5	38	85	0.0	80.0	NaN
6	86	98	88.0	70.0	84.0

pandas处理空值操作

• isnull

• notnull

• any

• all

• dropna

• fillna

• 方式1：对空值进行过滤（删除空所在的行数据）

  • 技术：isnull，notnull，any，all

df.isnull()

	0	1	2	3	4
0	False	False	False	False	False
1	False	False	False	False	False
2	False	False	False	True	False
3	False	False	False	False	False
4	False	False	True	False	False
5	False	False	False	False	True
6	False	False	False	False	False

#哪些行中有空值
#any(axis=1)检测哪些行中存有空值
df.isnull().any(axis=1) #any会作用isnull返回结果的每一行
#true对应的行就是存有缺失数据的行

0    False
1    False
2     True
3    False
4     True
5     True
6    False
dtype: bool

df.notnull()

	0	1	2	3	4
0	True	True	True	True	True
1	True	True	True	True	True
2	True	True	True	False	True
3	True	True	True	True	True
4	True	True	False	True	True
5	True	True	True	True	False
6	True	True	True	True	True

df.notnull().all(axis=1)

0     True
1     True
2    False
3     True
4    False
5    False
6     True
dtype: bool

#将布尔值作为源数据的行索引
df.loc[df.notnull().all(axis=1)]

	0	1	2	3	4
0	23	67	72.0	72.0	78.0
1	37	26	28.0	44.0	19.0
3	76	67	15.0	15.0	50.0
6	86	98	88.0	70.0	84.0

#获取空对应的行数据
df.loc[df.isnull().any(axis=1)]
#获取空对应行数据的行索引
indexs = df.loc[df.isnull().any(axis=1)].index
indexs

Int64Index([2, 4, 5], dtype='int64')

df.drop(labels=indexs,axis=0)

	0	1	2	3	4
0	23	67	72.0	72.0	78.0
1	37	26	28.0	44.0	19.0
3	76	67	15.0	15.0	50.0
6	86	98	88.0	70.0	84.0

• 方式2：
  • dropna：可以直接将缺失的行或者列进行删除

df.dropna(axis=0)

	0	1	2	3	4
0	23	67	72.0	72.0	78.0
1	37	26	28.0	44.0	19.0
3	76	67	15.0	15.0	50.0
6	86	98	88.0	70.0	84.0

• 对缺失值进行覆盖
  • fillna

df.fillna(value=999) #使用指定值将源数据中所有的空值进行填充

	0	1	2	3	4
0	23	67	72.0	72.0	78.0
1	37	26	28.0	44.0	19.0
2	0	69	90.0	999.0	28.0
3	76	67	15.0	15.0	50.0
4	54	52	999.0	69.0	53.0
5	38	85	0.0	80.0	999.0
6	86	98	88.0	70.0	84.0

#使用空的近邻值进行填充
#method=ffill向前填充，bfill向后填充 axis=o 列填充
df.fillna(axis=0,method='bfill')

	0	1	2	3	4
0	23	67	72.0	72.0	78.0
1	37	26	28.0	44.0	19.0
2	0	69	90.0	15.0	28.0
3	76	67	15.0	15.0	50.0
4	54	52	0.0	69.0	53.0
5	38	85	0.0	80.0	84.0
6	86	98	88.0	70.0	84.0

• 什么时候用dropna什么时候用fillna

  • 尽量使用dropna，如果删除成本比较高，则使用fillna

• 使用空值对应列的均值进行空值填充

for col in df.columns:
    #检测哪些列中存有空值
    if df[col].isnull().sum() > 0:#说明df[col]中存有空值
        mean_value = df[col].mean()
        df[col] = df[col].fillna(value=mean_value)

实例：

• 数据说明：

  • 数据是1个冷库的温度数据，1-7对应7个温度采集设备，1分钟采集一次。

• 数据处理目标：

  • 用1-4对应的4个必须设备，通过建立冷库的温度场关系模型，预估出5-7对应的数据。
  • 最后每个冷库中仅需放置4个设备，取代放置7个设备。
  • f(1-4) --> y(5-7)

• 数据处理过程：

  • 1、原始数据中有丢帧现象，需要做预处理；
  • 2、matplotlib 绘图；
  • 3、建立逻辑回归模型。

• 无标准答案，按个人理解操作即可，请把自己的操作过程以文字形式简单描述一下，谢谢配合。

• 测试数据为testData.xlsx

data = pd.read_excel('./data/testData.xlsx').drop(labels=['none','none1'],axis=1)
data

	time	1	2	3	4	5	6	7
0	2019-01-27 17:00:00	-24.8	-18.2	-20.8	-18.8	NaN	NaN	NaN
1	2019-01-27 17:01:00	-23.5	-18.8	-20.5	-19.8	-15.2	-14.5	-16.0
2	2019-01-27 17:02:00	-23.2	-19.2	NaN	NaN	-13.0	NaN	-14.0
3	2019-01-27 17:03:00	-22.8	-19.2	-20.0	-20.5	NaN	-12.2	-9.8
4	2019-01-27 17:04:00	-23.2	-18.5	-20.0	-18.8	-10.2	-10.8	-8.8
5	2019-01-27 17:05:00	NaN	NaN	-19.0	-18.2	-10.0	-10.5	-10.8
6	2019-01-27 17:06:00	NaN	-18.5	-18.2	-17.5	NaN	NaN	NaN
7	2019-01-27 17:07:00	-24.8	-18.0	-17.5	-17.2	-14.2	-14.0	-12.5
8	2019-01-27 17:08:00	-25.2	-17.8	NaN	NaN	-16.2	NaN	-14.5
9	2019-01-27 17:09:00	-24.8	-18.2	NaN	-17.5	NaN	-15.5	-16.0
10	2019-01-27 17:10:00	-24.5	-18.5	-16.0	-18.5	-17.5	-16.5	-17.2
11	2019-01-27 17:11:00	NaN	NaN	-16.0	-18.5	-17.8	-16.8	-12.0
12	2019-01-27 17:12:00	NaN	-18.5	-15.8	-18.8	NaN	NaN	NaN
13	2019-01-27 17:13:00	-23.8	-18.5	NaN	NaN	4.5	NaN	0.0
14	2019-01-27 17:14:00	-23.2	-18.2	NaN	-19.0	NaN	5.8	6.8
15	2019-01-27 17:15:00	-23.5	-17.8	-15.0	-18.0	10.5	10.5	10.8
16	2019-01-27 17:16:00	NaN	NaN	-14.2	-17.2	14.0	13.5	13.0
17	2019-01-27 17:17:00	NaN	-18.2	-13.8	-17.8	15.8	15.2	14.2
18	2019-01-27 17:18:00	-23.2	-19.0	-13.8	-18.2	NaN	NaN	NaN
19	2019-01-27 17:19:00	-23.2	-19.5	NaN	NaN	17.8	NaN	15.2
20	2019-01-27 17:20:00	-23.2	-19.8	NaN	-19.0	18.2	17.2	15.8
21	2019-01-27 17:21:00	-23.5	-20.0	-13.8	-19.5	NaN	17.8	16.0
22	2019-01-27 17:22:00	NaN	NaN	-14.0	-19.5	18.8	18.0	16.2
23	2019-01-27 17:23:00	-23.2	-20.2	-14.0	-19.5	19.0	18.2	16.5
24	2019-01-27 17:24:00	NaN	-20.2	-14.2	-19.5	NaN	NaN	NaN
25	2019-01-27 17:25:00	-22.8	-20.5	-14.5	-19.5	19.2	NaN	16.5
26	2019-01-27 17:26:00	-22.8	-20.8	-15.0	-16.8	NaN	17.2	16.8
27	2019-01-27 17:27:00	-22.0	-16.0	NaN	-16.0	18.8	17.2	16.2
28	2019-01-27 17:28:00	-22.8	-15.2	-14.8	-15.2	18.8	17.2	16.2
29	2019-01-27 17:29:00	-22.5	-15.0	-14.8	-15.2	18.8	17.2	16.5
...	...	...	...	...	...	...	...	...
1030	2019-01-28 10:10:00	-30.5	-27.5	-29.5	-27.8	-3.8	-3.5	-8.2
1031	2019-01-28 10:11:00	-30.8	-27.0	-29.2	-27.8	-3.8	-3.2	-8.5
1032	2019-01-28 10:12:00	-30.5	-26.2	-29.0	-26.8	-3.5	-3.0	-8.8
1033	2019-01-28 10:13:00	-28.8	-25.2	-28.2	-26.2	-3.5	-3.0	-8.8
1034	2019-01-28 10:14:00	-25.2	-25.2	-28.2	-25.8	-3.0	-2.5	-8.8
1035	2019-01-28 10:15:00	-25.2	-25.8	-28.5	-26.2	-3.0	-2.2	-8.5
1036	2019-01-28 10:16:00	-25.8	-26.2	-28.8	-26.8	-2.8	-2.0	-8.2
1037	2019-01-28 10:17:00	-26.2	-26.8	-29.0	-27.2	-2.5	-1.8	-8.2
1038	2019-01-28 10:18:00	-26.5	-27.0	-29.2	-27.5	NaN	NaN	NaN
1039	2019-01-28 10:19:00	-27.0	-27.2	-29.5	-28.0	-2.2	-1.5	-7.8
1040	2019-01-28 10:20:00	-26.5	-26.8	-29.0	-28.0	-2.2	-1.5	-7.5
1041	2019-01-28 10:21:00	-25.0	-25.8	-28.5	-27.2	-2.2	-1.5	-7.5
1042	2019-01-28 10:22:00	-24.0	-25.2	-28.2	-26.5	-2.0	-1.5	-7.2
1043	2019-01-28 10:23:00	-23.8	-25.0	-28.0	-26.0	-2.0	-1.5	-7.2
1044	2019-01-28 10:24:00	-24.0	-25.2	-28.0	-25.5	-2.0	-1.5	-7.0
1045	2019-01-28 10:25:00	-25.0	-26.0	-28.2	-26.2	-2.2	-1.5	-6.8
1046	2019-01-28 10:26:00	-25.8	-26.5	-28.8	-26.8	-2.2	-1.5	-6.5
1047	2019-01-28 10:27:00	-26.2	-26.5	-28.8	-27.2	-2.2	-1.5	-6.5
1048	2019-01-28 10:28:00	-25.0	-25.8	-28.5	-27.0	-2.2	-1.8	-6.2
1049	2019-01-28 10:29:00	-24.8	-25.2	-28.0	-26.2	-2.2	-1.8	-6.0
1050	2019-01-28 10:30:00	-24.5	-24.8	-27.8	-25.8	-2.0	-2.0	-6.0
1051	2019-01-28 10:31:00	-24.0	-24.8	-27.8	-25.5	-2.0	-2.0	-5.8
1052	2019-01-28 10:32:00	-24.2	-25.5	-28.0	-26.0	-2.0	-2.0	-5.5
1053	2019-01-28 10:33:00	-25.0	-26.2	-28.2	-26.8	-2.0	-2.0	-5.2
1054	2019-01-28 10:34:00	-25.8	-26.8	-28.5	-27.0	-2.0	-2.2	-5.2
1055	2019-01-28 10:35:00	-26.2	-27.2	-28.8	-27.5	-2.0	NaN	-5.0
1056	2019-01-28 10:36:00	-26.8	-27.5	-29.0	-27.8	-2.2	NaN	-5.0
1057	2019-01-28 10:37:00	-27.2	-27.8	-29.0	-28.0	-2.2	NaN	-5.0
1058	2019-01-28 10:38:00	-27.5	-27.0	-29.0	-28.0	-3.5	-3.2	-5.8
1059	2019-01-28 10:39:00	-27.0	-27.2	-29.0	-27.8	-5.0	NaN	-7.0

1060 rows × 8 columns

data.shape

(1060, 8)

#删除空对应的行数据
data.dropna(axis=0).shape

(927, 8)

#填充
data.fillna(method='ffill',axis=0).fillna(method='bfill',axis=0)

	time	1	2	3	4	5	6	7
0	2019-01-27 17:00:00	-24.8	-18.2	-20.8	-18.8	-15.2	-14.5	-16.0
1	2019-01-27 17:01:00	-23.5	-18.8	-20.5	-19.8	-15.2	-14.5	-16.0
2	2019-01-27 17:02:00	-23.2	-19.2	-20.5	-19.8	-13.0	-14.5	-14.0
3	2019-01-27 17:03:00	-22.8	-19.2	-20.0	-20.5	-13.0	-12.2	-9.8
4	2019-01-27 17:04:00	-23.2	-18.5	-20.0	-18.8	-10.2	-10.8	-8.8
5	2019-01-27 17:05:00	-23.2	-18.5	-19.0	-18.2	-10.0	-10.5	-10.8
6	2019-01-27 17:06:00	-23.2	-18.5	-18.2	-17.5	-10.0	-10.5	-10.8
7	2019-01-27 17:07:00	-24.8	-18.0	-17.5	-17.2	-14.2	-14.0	-12.5
8	2019-01-27 17:08:00	-25.2	-17.8	-17.5	-17.2	-16.2	-14.0	-14.5
9	2019-01-27 17:09:00	-24.8	-18.2	-17.5	-17.5	-16.2	-15.5	-16.0
10	2019-01-27 17:10:00	-24.5	-18.5	-16.0	-18.5	-17.5	-16.5	-17.2
11	2019-01-27 17:11:00	-24.5	-18.5	-16.0	-18.5	-17.8	-16.8	-12.0
12	2019-01-27 17:12:00	-24.5	-18.5	-15.8	-18.8	-17.8	-16.8	-12.0
13	2019-01-27 17:13:00	-23.8	-18.5	-15.8	-18.8	4.5	-16.8	0.0
14	2019-01-27 17:14:00	-23.2	-18.2	-15.8	-19.0	4.5	5.8	6.8
15	2019-01-27 17:15:00	-23.5	-17.8	-15.0	-18.0	10.5	10.5	10.8
16	2019-01-27 17:16:00	-23.5	-17.8	-14.2	-17.2	14.0	13.5	13.0
17	2019-01-27 17:17:00	-23.5	-18.2	-13.8	-17.8	15.8	15.2	14.2
18	2019-01-27 17:18:00	-23.2	-19.0	-13.8	-18.2	15.8	15.2	14.2
19	2019-01-27 17:19:00	-23.2	-19.5	-13.8	-18.2	17.8	15.2	15.2
20	2019-01-27 17:20:00	-23.2	-19.8	-13.8	-19.0	18.2	17.2	15.8
21	2019-01-27 17:21:00	-23.5	-20.0	-13.8	-19.5	18.2	17.8	16.0
22	2019-01-27 17:22:00	-23.5	-20.0	-14.0	-19.5	18.8	18.0	16.2
23	2019-01-27 17:23:00	-23.2	-20.2	-14.0	-19.5	19.0	18.2	16.5
24	2019-01-27 17:24:00	-23.2	-20.2	-14.2	-19.5	19.0	18.2	16.5
25	2019-01-27 17:25:00	-22.8	-20.5	-14.5	-19.5	19.2	18.2	16.5
26	2019-01-27 17:26:00	-22.8	-20.8	-15.0	-16.8	19.2	17.2	16.8
27	2019-01-27 17:27:00	-22.0	-16.0	-15.0	-16.0	18.8	17.2	16.2
28	2019-01-27 17:28:00	-22.8	-15.2	-14.8	-15.2	18.8	17.2	16.2
29	2019-01-27 17:29:00	-22.5	-15.0	-14.8	-15.2	18.8	17.2	16.5
...	...	...	...	...	...	...	...	...
1030	2019-01-28 10:10:00	-30.5	-27.5	-29.5	-27.8	-3.8	-3.5	-8.2
1031	2019-01-28 10:11:00	-30.8	-27.0	-29.2	-27.8	-3.8	-3.2	-8.5
1032	2019-01-28 10:12:00	-30.5	-26.2	-29.0	-26.8	-3.5	-3.0	-8.8
1033	2019-01-28 10:13:00	-28.8	-25.2	-28.2	-26.2	-3.5	-3.0	-8.8
1034	2019-01-28 10:14:00	-25.2	-25.2	-28.2	-25.8	-3.0	-2.5	-8.8
1035	2019-01-28 10:15:00	-25.2	-25.8	-28.5	-26.2	-3.0	-2.2	-8.5
1036	2019-01-28 10:16:00	-25.8	-26.2	-28.8	-26.8	-2.8	-2.0	-8.2
1037	2019-01-28 10:17:00	-26.2	-26.8	-29.0	-27.2	-2.5	-1.8	-8.2
1038	2019-01-28 10:18:00	-26.5	-27.0	-29.2	-27.5	-2.5	-1.8	-8.2
1039	2019-01-28 10:19:00	-27.0	-27.2	-29.5	-28.0	-2.2	-1.5	-7.8
1040	2019-01-28 10:20:00	-26.5	-26.8	-29.0	-28.0	-2.2	-1.5	-7.5
1041	2019-01-28 10:21:00	-25.0	-25.8	-28.5	-27.2	-2.2	-1.5	-7.5
1042	2019-01-28 10:22:00	-24.0	-25.2	-28.2	-26.5	-2.0	-1.5	-7.2
1043	2019-01-28 10:23:00	-23.8	-25.0	-28.0	-26.0	-2.0	-1.5	-7.2
1044	2019-01-28 10:24:00	-24.0	-25.2	-28.0	-25.5	-2.0	-1.5	-7.0
1045	2019-01-28 10:25:00	-25.0	-26.0	-28.2	-26.2	-2.2	-1.5	-6.8
1046	2019-01-28 10:26:00	-25.8	-26.5	-28.8	-26.8	-2.2	-1.5	-6.5
1047	2019-01-28 10:27:00	-26.2	-26.5	-28.8	-27.2	-2.2	-1.5	-6.5
1048	2019-01-28 10:28:00	-25.0	-25.8	-28.5	-27.0	-2.2	-1.8	-6.2
1049	2019-01-28 10:29:00	-24.8	-25.2	-28.0	-26.2	-2.2	-1.8	-6.0
1050	2019-01-28 10:30:00	-24.5	-24.8	-27.8	-25.8	-2.0	-2.0	-6.0
1051	2019-01-28 10:31:00	-24.0	-24.8	-27.8	-25.5	-2.0	-2.0	-5.8
1052	2019-01-28 10:32:00	-24.2	-25.5	-28.0	-26.0	-2.0	-2.0	-5.5
1053	2019-01-28 10:33:00	-25.0	-26.2	-28.2	-26.8	-2.0	-2.0	-5.2
1054	2019-01-28 10:34:00	-25.8	-26.8	-28.5	-27.0	-2.0	-2.2	-5.2
1055	2019-01-28 10:35:00	-26.2	-27.2	-28.8	-27.5	-2.0	-2.2	-5.0
1056	2019-01-28 10:36:00	-26.8	-27.5	-29.0	-27.8	-2.2	-2.2	-5.0
1057	2019-01-28 10:37:00	-27.2	-27.8	-29.0	-28.0	-2.2	-2.2	-5.0
1058	2019-01-28 10:38:00	-27.5	-27.0	-29.0	-28.0	-3.5	-3.2	-5.8
1059	2019-01-28 10:39:00	-27.0	-27.2	-29.0	-27.8	-5.0	-3.2	-7.0

1060 rows × 8 columns

处理重复数据

df = DataFrame(data=np.random.randint(0,100,size=(8,6)))
df.iloc[1] = [1,1,1,1,1,1]
df.iloc[3] = [1,1,1,1,1,1]
df.iloc[5] = [1,1,1,1,1,1]
df

	0	1	2	3	4	5
0	3	29	47	11	69	7
1	1	1	1	1	1	1
2	6	43	65	79	52	82
3	1	1	1	1	1	1
4	2	67	90	8	96	76
5	1	1	1	1	1	1
6	38	6	56	50	71	30
7	9	16	12	67	32	0

#检测哪些行存有重复的数据
df.duplicated(keep='first')

0    False
1    False
2    False
3     True
4    False
5     True
6    False
7    False
dtype: bool

df.loc[~df.duplicated(keep='first')]

	0	1	2	3	4	5
0	3	29	47	11	69	7
1	1	1	1	1	1	1
2	6	43	65	79	52	82
4	2	67	90	8	96	76
6	38	6	56	50	71	30
7	9	16	12	67	32	0

#异步到位删除
df.drop_duplicates(keep='first')

	0	1	2	3	4	5
0	3	29	47	11	69	7
1	1	1	1	1	1	1
2	6	43	65	79	52	82
4	2	67	90	8	96	76
6	38	6	56	50	71	30
7	9	16	12	67	32	0

处理异常数据

• 自定义一个1000行3列（A，B，C）取值范围为0-1的数据源，然后将C列中的值大于其两倍标准差的异常值进行清洗

df = DataFrame(data=np.random.random(size=(1000,3)),columns=['A','B','C'])
df.head()

	A	B	C
0	0.093731	0.358117	0.599607
1	0.415552	0.124499	0.820207
2	0.278742	0.222851	0.786832
3	0.537346	0.339687	0.276611
4	0.414794	0.179321	0.094958

#制定判定异常值的条件
twice_std = df['C'].std() * 2
twice_std

0.5664309886908782

df.loc[~(df['C'] > twice_std)]

	A	B	C
3	0.537346	0.339687	0.276611
4	0.414794	0.179321	0.094958
5	0.397169	0.610316	0.420824
7	0.740718	0.730160	0.302804
8	0.483627	0.146781	0.126308
9	0.530481	0.426309	0.126464
11	0.450225	0.399873	0.564089
12	0.149199	0.706758	0.271892
14	0.503967	0.098280	0.239464
15	0.630785	0.661415	0.139548
18	0.055859	0.917004	0.285184
19	0.153187	0.876151	0.499308
20	0.743452	0.679769	0.563917
21	0.151130	0.469532	0.382700
22	0.941487	0.967286	0.234784
24	0.459680	0.080923	0.249299
26	0.973956	0.553966	0.094996
28	0.084031	0.689861	0.067293
29	0.484953	0.188724	0.160719
31	0.512300	0.392566	0.100742
32	0.237209	0.004658	0.449384
34	0.345122	0.658354	0.391020
35	0.709282	0.236713	0.499434
39	0.055908	0.924785	0.490992
41	0.047936	0.205894	0.160804
45	0.929671	0.799439	0.335439
46	0.169237	0.037535	0.494065
50	0.808850	0.804284	0.223080
51	0.320385	0.385899	0.189706
52	0.926416	0.748975	0.183758
...	...	...	...
949	0.596174	0.224759	0.280238
951	0.382255	0.906360	0.401972
954	0.434587	0.983834	0.118732
955	0.914734	0.279118	0.078705
957	0.977942	0.194291	0.253350
958	0.139654	0.683716	0.118146
959	0.308191	0.612879	0.445845
961	0.081917	0.100586	0.116678
962	0.390555	0.762205	0.083272
964	0.712109	0.870591	0.393287
965	0.107085	0.056523	0.304899
968	0.588549	0.535405	0.248742
971	0.498968	0.489234	0.080411
973	0.165779	0.110859	0.384091
974	0.778701	0.489504	0.533272
975	0.057621	0.839546	0.275676
976	0.605409	0.293276	0.482304
977	0.555336	0.287849	0.468799
978	0.484669	0.993484	0.151512
979	0.418097	0.858759	0.220208
983	0.033246	0.539796	0.128987
984	0.973549	0.277905	0.311013
986	0.469728	0.046535	0.274008
987	0.037183	0.136681	0.279782
989	0.824013	0.938513	0.022778
992	0.683002	0.567619	0.003076
993	0.493820	0.617086	0.202174
996	0.177156	0.248502	0.096410
998	0.914179	0.470827	0.129195
999	0.276110	0.942467	0.510295

573 rows × 3 columns