熊猫数据框中某列的百分位数等级
让我们看看如何在熊猫数据框中找到一列的百分位数。我们将使用带有参数pct = True
的**[rank()](https://www.geeksforgeeks.org/python-pandas-dataframe-rank/)**
函数来找到百分位数。
例 1 :
# import the module
import pandas as pd
# create a DataFrame
data = {'Name': ['Mukul', 'Rohan', 'Mayank',
'Shubham', 'Aakash'],
'Location' : ['Saharanpur', 'Meerut', 'Agra',
'Saharanpur', 'Meerut'],
'Pay' : [50000, 70000, 62000, 67000, 56000]}
df = pd.DataFrame(data)
# create a new column of percentile rank
df['Percentile Rank'] = df.Pay.rank(pct = True)
# displaying the percentile rank
display(df)
输出:
例 2 :
# import the module
import pandas as pd
# create a DataFrame
ODI_runs = {'name': ['Tendulkar', 'Sangakkara', 'Ponting',
'Jayasurya', 'Jayawardene', 'Kohli',
'Haq', 'Kallis', 'Ganguly', 'Dravid'],
'runs': [18426, 14234, 13704, 13430, 12650,
11867, 11739, 11579, 11363, 10889]}
df = pd.DataFrame(ODI_runs)
# create a new column of percentile rank
df['Percentile Rank'] = df.runs.rank(pct = True)
# displaying the percentile rank
display(df)
输出: