有没有更有效的方法来聚合数据集并在 Python 或 R 中计算频率？

Question

i have a dataset [0, 1, 1, 2], I want to aggregate it. to do this, I have to compute and put the 'frequency':1/4 manually into a DataFrame. here is the code.

>>> df = pd.DataFrame({'value':[0, 1, 1, 2],
...             'frequency':1/4})
>>> df.groupby('value').sum()
       frequency
value           
0           0.25
1           0.50
2           0.25

is there a more efficient way to aggregate the dataset and calculate the frequency automatically in Python or R?

Answer 1

df['value'].value_counts(normalize=True,sort=False)

Maybe you could try this...

Reference:-

1. pandas.Series.value_counts()

Answer 2

In R

prop.table(table(dat$value))

   0    1    2 
0.25 0.50 0.25 

In python, NumPy

import numpy as np 
u,c=np.unique(df.value,return_counts=True)
pd.Series(c/c.sum(),index=u)
0    0.25
1    0.50
2    0.25
dtype: float64

Answer 3

In R you could do something like

library(data.table)
dt <- data.table(sample(0:2,100,replace=TRUE))
dt[,.N/nrow(dt),V1]

## > dt[,.N/nrow(dt),V1]

##    V1   V1
## 1:  1 0.33
## 2:  2 0.32
## 3:  0 0.35

Answer 4

without using pandas you could use Counter

from collections import Counter
z = [0,1,1,2]
Counter(z)
Counter({1: 2, 0: 1, 2: 1})

and then to a dataframe

x = Counter(z)
df = pd.DataFrame.from_dict(x, orient='index').reset_index()

and then take the values divided by 4 (your desired Freq)

Answer 5

import pandas as pd
pd.Series([0, 1, 1, 2]).value_counts(normalize=True, sort=False)