Data Normalization

Version: 1.0.0

Commit Hash: 556d78defb0ebc01c5b0ec2bdf1a8e5647dc4827

Author: CauldronGO Team

Category: preprocessing

Normalize data using various scaling methods

Install with Cauldron get_app
README

Data Normalization

ID: normalization
Version: 1.0.0
Category: preprocessing
Author: CauldronGO Team

Description

Normalize data using various scaling methods

Runtime

  • Type: python
  • Script: normalization.py

Inputs

Name Label Type Required Default Visibility
input_file Input File file Yes - Always visible
columns_name Sample Columns column-selector (multiple) Yes - Always visible
scaler_type Scaler Type select (minmax, standard, robust, quantile, normalizer, power) Yes minmax Always visible
with_centering With Centering boolean No false Visible when scaler_type = standard
with_scaling With Scaling boolean No true Visible when scaler_type = standard
n_quantiles Number of Quantiles number (min: 100, max: 10000, step: 100) No 1000 Visible when scaler_type = quantile
output_distribution Output Distribution select (uniform, normal) No uniform Visible when scaler_type = quantile
norm Normalization Method select (l1, l2, max) No l2 Visible when scaler_type = normalizer
power_method Power Transform Method select (yeo-johnson, box-cox) No yeo-johnson Visible when scaler_type = power

Input Details

Input File (input_file)

Data file to normalize

Sample Columns (columns_name)

Select columns to normalize

  • Column Source: input_file

Scaler Type (scaler_type)

Normalization method to use

  • Options: minmax, standard, robust, quantile, normalizer, power

With Centering (with_centering)

Center the data before scaling

With Scaling (with_scaling)

Scale the data to unit variance

Number of Quantiles (n_quantiles)

Number of quantiles for quantile transformation

Output Distribution (output_distribution)

Distribution for quantile transformation output

  • Options: uniform, normal

Normalization Method (norm)

Norm to use for normalizer scaler

  • Options: l1, l2, max

Power Transform Method (power_method)

Method for power transformation

  • Options: yeo-johnson, box-cox

Outputs

Name File Type Format Description
normalized_data normalized.data.txt data tsv Normalized data matrix

Requirements

  • Python: >=3.11
  • Packages:
  • numpy>=1.24.0
  • pandas>=2.0.0
  • scikit-learn>=1.3.0

Example Data

This plugin includes example data for testing:

  input_file: diann/imputed.data.txt
  columns_name_source: diann/imputed.data.txt
  columns_name: [C:\Raja\DIA-NN searches\June 2022\LT-CBQCA-Test_DIA\RN-DS_220106_BCA_LT-IP_01.raw C:\Raja\DIA-NN searches\June 2022\LT-CBQCA-Test_DIA\RN-DS_220106_BCA_LT-IP_02.raw C:\Raja\DIA-NN searches\June 2022\LT-CBQCA-Test_DIA\RN-DS_220106_BCA_LT-IP_03.raw C:\Raja\DIA-NN searches\June 2022\LT-CBQCA-Test_DIA\RN-DS_220106_BCA_LT-MockIP_01.raw C:\Raja\DIA-NN searches\June 2022\LT-CBQCA-Test_DIA\RN-DS_220106_BCA_LT-MockIP_02.raw C:\Raja\DIA-NN searches\June 2022\LT-CBQCA-Test_DIA\RN-DS_220106_BCA_LT-MockIP_03.raw]
  scaler_type: minmax

Load example data by clicking the Load Example button in the UI.

Usage

Via UI

  1. Navigate to preprocessing → Data Normalization
  2. Fill in the required inputs
  3. Click Run Analysis

Via Plugin System

const jobId = await pluginService.executePlugin('normalization', {
  // Add parameters here
});