Skip to end of metadata
Go to start of metadata



Regression

Author: Irene Cho 

This plugin can do data analysis using various regression models.

Primary Features

  • Simple data analysis can be conducted using 17 regression models and 3 plots which are residuals, prediction error and Cook’s distance will evaluate the fitted model.
  • ‘Compare models’ operation compares the fitted regression models by measuring the average of squared of the errors between predicted values and actual values.
  • ‘Fit Model’ gives the summary of average of squared of the errors, proportion of variance and accuracy of the fitted model with the 3 useful plots.

Prerequisites

  • Input datafile (format: csv, tsv, xlsm, xls, xlsx, json)

Data Pre-processing

  • Normalization: Process of recognizing data to reduce or even remove redundancy.
  • Onehot-Encoding: Required process for categorical variables that removes the integer encoded variable and add a new binary variable.



Need help?

Technical contact to tech@argos-labs.com


May you search all operations,



Parameters


1. Compare Models

Required parameters

Type

Note

Target variable

str

A target variable to predict



2. Fit Model

Required parameters

Type

Note

Target variable

Str

A target variable to predict

Model Type

Str

Choose one of the given models

Plot Output (optional)

File path

Absolute file path for plot output

Plot Type (optional)

Str



(info) Example of the plugin in STU – Compare Models (Output: ‘.html’)


Text from Image


(info) Example of the plugin in STU – Fit Model (Output: CSV or plot image)


Text from Image






All Plugins