generateFeature

This function automatically scans through each variable and generate features based on the type listed in the "details"

Implementing some of the pillars of an automated machine learning pipeline such as (i) Automated data preparation, (ii) Feature engineering, (iii) Model building in classification context that includes techniques such as (a) Regularised regression [1], (b) Logistic regression [2], (c) Random Forest [3], (d) Decision tree [4] and (e) Extreme Gradient Boosting (xgboost) [5], and finally, (iv) Model explanation (using lift chart and partial dependency plots). Accomplishes the above tasks by running the function instead of writing lengthy R codes. Also provides some additional features such as generating missing at random (MAR) variables and automated exploratory data analysis. Moreover, function exports the model results with the required plots in an HTML vignette report format that follows the best practices of the industry and the academia. [1] Gonzales G B and De Saeger (2018) <doi:10.1038/s41598-018-21851-7>, [2] Sperandei S (2014) <doi:10.11613/BM.2014.003>, [3] Breiman L (2001) <doi:10.1023/A:1010933404324>, [4] Kingsford C and Salzberg S (2008) <doi:10.1038/nbt0908-1011>, [5] Chen Tianqi and Guestrin Carlos (2016) <doi:10.1145/2939672.2939785>.

Dayanand Ubrangala

DriveML

Self-Drive Machine Learning Projects

Dayanand Ubrangala 

Sayan Putatunda

Kiran R

Ravi Prasad Kondapalli

generateFeature function

<dl><dt>data</dt>
<dd>[data.frame | Required] dataframe or data.table</dd>
<dt>varlist</dt>
<dd>[text | Required] variable list to generate the additional features</dd>
<dt>type</dt>
<dd>[text | Required] variable transformation with type - 'Dummy','Outlier','Frequent' or 'Interaction'</dd>
<dt>method</dt>
<dd>[text | Required] input for variabe transformation for type = 'Frequent' then the method should be 'Frequency' or 'Percent'. Please refer to the "details" section to know more</dd></dl>

Arguments

Automated column transformer — generateFeature

<dl>

<dt>data</dt>
<dd>[data.frame | Required] dataframe or data.table</dd>


<dt>varlist</dt>
<dd>[text | Required] variable list to generate the additional features</dd>


<dt>type</dt>
<dd>[text | Required] variable transformation with type - 'Dummy','Outlier','Frequent' or 'Interaction'</dd>


<dt>method</dt>
<dd>[text | Required] input for variabe transformation for type = 'Frequent' then the method should be 'Frequency' or 'Percent'. Please refer to the "details" section to know more</dd>

</dl>

generateFeature: Automated column transformer

Description

Usage

Value

Arguments

Details

Examples