$$x_{scale}=\frac{x-x_{min}}{x_{max}-x_{min}}$$
$$x=x_{scale}({x_{max}-x_{min})+x_{min}}$$
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
X = [
[2, 3],
[4, 8],
[5, 2],
[2, 4]
]
X_scale = scaler.fit_transform(X)
print(scaler.data_min_)
print(scaler.data_max_)
X = scaler.inverse_transform(X_scale)
$$x_{scale}=\frac{x-\bar{x}}{s}$$
$$x=x_{scale} * s + \bar{x}$$
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler()
X = [
[2, 3],
[4, 8],
[5, 2],
[2, 4]
]
X_scale = scaler.fit_transform(X)
print(scaler.mean_)
print(scaler.var_)
X = scaler.inverse_transform(X_scale)
$$x_{scale}=\frac{x}{|x|_{max}}$$
$$x={x_{scale}}{|x|_{max}}$$
from sklearn.preprocessing import MaxAbsScaler
scaler = MaxAbsScaler()
X = [
[7, 2],
[4, -8],
[2, 5],
[8, 4],
[1, 5],
[-25, 4]
]
X_scale = scaler.fit_transform(X)
print(scaler.max_abs_)
scaler.inverse_transform(X_scale)