scaler module¶
Data scaler¶
The Scaler
class implements the default scaling method applying to
some parameter \(z\):
where \(\bar{z}\) is the scaled version of z. This scaling method is a linear transformation parameterized by an offset and a coefficient.
In this default scaling method, the offset is equal to 0 and the coefficient is equal to 1. Consequently, the scaling operation is the identity: \(\bar{z}=z\). This method has to be overloaded.
See also
-
class
gemseo.mlearning.transform.scaler.scaler.
Scaler
(name='Scaler', offset=0.0, coefficient=1.0)[source]¶ Bases:
gemseo.mlearning.transform.transformer.Transformer
Data scaler.
Constructor.
- Parameters
name (str) – name of the scaler.
offset (float) – offset of the linear transformation. Default: 0.
coefficient (float) – coefficient of the linear transformation. Default: 1.
-
property
coefficient
¶ Coefficient.
-
compute_jacobian
(data)[source]¶ Compute Jacobian of the scaler transform.
- Parameters
data (ndarray) – data where the Jacobian is to be computed.
- Returns
Jacobian matrix.
- Return type
ndarray
-
compute_jacobian_inverse
(data)[source]¶ Compute Jacobian of the scaler inverse_transform.
- Parameters
data (ndarray) – data where the Jacobian is to be computed.
- Returns
Jacobian matrix.
- Return type
ndarray
-
fit
(data)[source]¶ Fit scaler to data. Offset and coefficient terms are already defined in the constructor.
- Parameters
data (ndarray) – data to be fitted.
-
inverse_transform
(data)[source]¶ Unscale data using the offset and coefficient terms.
- Parameters
data (ndarray) – data to be inverse transformed.
- Returns
inverse transformed data.
- Return type
ndarray
-
property
offset
¶ Offset.