Scale columns in dataframe based on another column

Question

I have the dataframe shown below :

column1	column2	column3
1	0.01	0.001
2	0.02	0.002
3	0.03	0.003

I want to scale column 2 and column 3 based on the min and max scaler of column 1, in order to obtain the following :

column1	column2	column3
1	1	1
2	2	2
3	3	3

These numbers are just imputed to understand the question easily.

All the solutions I have seen online basically uses a min max where they scale the columns between 0-1 but I want to scale it based on the min max of the first column.

Answer 1

You can do this:

from sklearn.preprocessing import MinMaxScaler

scaler = MinMaxScaler() 
  
df[["column_i"]] = scaler.fit_transform(df[["column_i"]])

Answer 2

@Adrien's answer is very nice, but if you want to do it without external dependency:

MIN = df.min()
MAX = df.max()
(df-MIN)/(MAX-MIN)*(MAX['column1']-MIN['column1'])+MIN['column1']

Answer 3

You can fit a min-max scaler on column 1 and apply it to other columns :

from sklearn.preprocessing import MinMaxScaler

scaler = MinMaxScaler()
scaler.fit(df['column1'])

Apply it to column i :

df['column_i'] = scaler.transform(df['column_i'])