Pandas

Question

。

大家好！

我有一个 dfA (Table A)，其中包含某些产品可用的天数 (days_survived)。我需要计算每天可用的产品总数 (Table B)。我的意思是，我需要计算 dfA 中的行数以发现前 5 天 (df2) 的每一天的存活率。

Table答：

+-------+--------------+
| id    | days_survived|
+-------+--------------+
| 1     |  1           |
| 2     |  3           |
| 3     |  10          | 
| 4     |  40          |
| 5     |  4           |
| 6     |  9           |
+-------+--------------+

Table B（前5天预期结果分析）：

+-------+----------------+
| day   | #count_survived|
+-------+----------------+
| 1     |  6             |
| 2     |  5             |
| 3     |  5             | 
| 4     |  4             |
| 5     |  3             |
+-------+----------------+

这个结果意味着第一天总共有6个产品可用，然后第二天和第三天只有5个，然后第四天只有4个，最后第五天只有3个。

代码：

# create df
import pandas as pd
d = {'id': [1,2,3,4,5,6], 'days_survived': [1,3,10,40,4,9]}
dfA = pd.DataFrame(data=d)

有人能帮帮我吗？ :)

Answer 1

将列表理解与展平和过滤结合使用，然后计数：

comp = [y for x in dfA['days_survived'] for y in range(1, x + 1) if y < 6]
s = pd.Series(comp).value_counts().rename_axis('day').reset_index(name='#count_survived')
print (s)
   day  #count_survived
0    1                6
1    3                5
2    2                5
3    4                4
4    5                3

Counter的另一个解决方案：

from collections import Counter

comp = [y for x in dfA['days_survived'] for y in range(1, x + 1) if y < 6]
d = Counter(comp)
df = pd.DataFrame({'day':list(d.keys()), '#count_survived':list(d.values())})

Answer 2

这是使用集合，创建一个列表，列出某个项目出现的所有天数，然后计算列表中每一天的出现次数

import pandas as pd
import numpy as np
from collections import Counter

df = pd.DataFrame(data={'id': [1,2,3,4,5,6], 'days_survived': [1,3,10,40,4,9]})
# We will create a new column having values as a list of all the days for which item was present
df['Days'] = df.apply(lambda a :  list(np.arange(1,a.days_survived+1)),axis=1)
# Applyin Counter to the flattened list of all elements in 'Days' column
cnt= Counter([item for items in list(df['Days']) for item in items])
#Converting cnt Counter object to Dataframe
df_new = pd.DataFrame(data= {'Days':list(cnt.keys()),'count':list(cnt.values())})

希望这对您有所帮助。

Pandas - 计算 df 中的行以发现每天的存活率

Pandas - Counting rows in a df to discover the survival rate each day

python

sklearn-pandas

pandas-groupby