<figure><img src="../.gitbook/assets/image (322).png" alt="" width="529"><figcaption></figcaption></figure>

1. Click on **Data Prep** in the **Machine Learning** category.

<figure><img src="../.gitbook/assets/image (323).png" alt="" width="563"><figcaption></figcaption></figure>

* [**Encoding**](3.-data-prep.md#encoding)

* [**Scaling**](3.-data-prep.md#scaling)

* [**ETC**](3.-data-prep.md#etc-simpleimputer-smote-makecolumntransformer)

<figure><img src="../.gitbook/assets/image (324).png" alt="" width="563"><figcaption></figcaption></figure>

1. _**Sparse (OneHotEncoder)**_: If _**true**,_ returns the encoding result as a sparse matrix.

2. _**Handle unknown (OneHotEncoder, OrdinalEncoder)**_: Used when encoding, if there is a category that exists in the training data but not in the test data. If _**ignore** is_ selected, it will be set to 0, and if _**error**_ is selected, a ValueError will be raised.

3. _**Unknown values (OrdinalEncoder)**_: Fill with a specific value, not ignore or error.

4. _**Cols (TargetEncoder)**_: Select the columns to encode.

5. _**Handle missing (TargetEncoder)**_: Choose how to handle missing values.

6. _**Smoothing (TargetEncoder)**_: When the number of data in a particular category is small, it adds the entered values and calculates the average of the categories to prevent overfitting.

<figure><img src="../.gitbook/assets/image (325).png" alt="" width="563"><figcaption></figcaption></figure>

1. _**With mean (StandardScaler)**_: Center the mean of the data to zero.

2. _**With std (StandardScaler)**_: Scale the standard deviation of the data to 1.

3. _**With centering (RobustScaler)**_: Performs centering by Q-subtracting the median from each attribute (column)_._

4. _**With scaling (RobustScaler)**_: Scales each attribute by dividing it by its IQR.

5. _**Feature range (MinMaxScaler)**_: Sets the minimum and maximum values for the scaled result.

6. _**Norm (Normalizer)**:_

1. _**L1**_: The sum of the absolute values of each attribute will be 1.

2. _**L2**_: Scale the vectors so that their Euclidean distance is 1.&#x20;

3. _**Max Norm**_: Ensures that the scaling result does not exceed an existing maximum value.

7. _**N bins (KBins Discretizer)**_: Determines how many bins to divide the variable into.

8. _**Strategy (KBins Discretizer)**_:

1. _**uniform**_: Divide the section by a uniform width.

2. _**QUANTILE**_: Divide so that each bin has an even number of data.

9. _**Encode (KBins Discretizer)**_: Specify the encoding method.

1. _**ordinal**_: Encodes each interval as an integer.

2. _**onehot**_: Encodes each interval as a binary vector.

<figure><img src="../.gitbook/assets/image (326).png" alt="" width="563"><figcaption></figcaption></figure>

1. _**Missing values (SimpleImputer)**_: Treats the entered values as missing.

2. _**Fill value (SimpleImputer)**_: Replaces _the_ missing value with the input value.

3. _**Copy (SimpleImputer)**_: Returns the original data unchanged, as new data.

4. _**Add indicator (SimpleImputer)**_: Adds a new column with 0s and 1s, with a 1 for rows with missing values and a 0 for rows without.

5. _**K neighbors (SMOTE)**_: Specifies the number of neighbors to group together based on center point data.

6. _**Sampling strategy (SMOTE)**_:

1. _**auto**_: Automatically adjusts the ratio of minority to majority class data to balance out class imbalances.&#x20;

2. _**minority**_: Makes the size of the minority class dataset equal to the size of the majority class dataset.

3. _**float**_: You can specify the desired class ratio. For example, setting it to 0.5 makes the minority class dataset half the size of the majority class dataset.

7. _**Estimator (MakeColumnTransformer)**_: You can specify different global models to apply to each column. The model selected here will be applied to the columns selected _in Columns_ below.