Difference between revisions of "Regression analysis advanced (analysis)"

Latest revision as of 18:15, 9 December 2020

Analysis title

Regression analysis advanced

Provider

Institute of Systems Biology

Class

RegressionAnalysisAdvanced

Plugin

biouml.plugins.machinelearning (Machine learning)

[edit] Description

Create and save regression model or load regression model for prediction of response or cross-validation of regression model.

[edit] Parameters:

• Regression mode – Select regression mode
• Regression type – Select regression type
• Path to data matrix – Path to table or file with data matrix
• Variable names – Select variable names
• Response name – Select response name
• Path to folder with saved model – Path to folder with saved model
• Parameters for OLS-regression – Please, determine parameters for Odinary least squares regression
  • Max number of rotations – Maximal number of rotations for calculation of inverse matrix or eigen vectors
  • Epsilon for rotations – Epsilon for calculation of inverse matrix or eigen vectors
• Parameters for WLS-regression – Please, determine parameters for Weighted least squares regression
  • Max number of rotations – Maximal number of rotations for calculation of inverse matrix or eigen vectors
  • Epsilon for rotations – Epsilon for calculation of inverse matrix or eigen vectors
• Parameters for PC-regression – Please, determine parameters for Principal component regression
  • Max number of rotations – Maximal number of rotations for calculation of inverse matrix or eigen vectors
  • Epsilon for rotations – Epsilon for calculation of inverse matrix or eigen vectors
  • Number of principal components – Number of principal components
  • Principal component sorting type – Sorting type of principal components
• Parameters for Tree-based regression – Please, determine parameters for Tree-based regression
  • Minimal node size – Minimal size of node
  • Minimal variance – Minimal variance
• Parameters for Ridge regression – Please, determine parameters for Ridge regression
  • Max number of rotations – Maximal number of rotations for calculation of inverse matrix or eigen vectors
  • Epsilon for rotations – Epsilon for calculation of inverse matrix or eigen vectors
  • Shrinkage parameter – Shrinkage parameter, k >= 0
• Parameters for combined regression – Please, determine parameters for combined regression
  • Number of regressions – Number of regressions
  • Regressions type – Select regressions type
  • Number of variables for regressions – Number of variables for each regression
  • Number of outlier detection steps – Number of outlier detection steps
  • Multiplier for sigma, t – Observation x is outlier if Abs(x - predicted x) > t * sigma
  • Classification type – Select classification type
  • Number of variables for classification – Number of variables for classification
  • Type of variable selection in classification – Type of variable selection in classification
• Parameters for cross-validation – Please, determine parameters for cross-validation
  • Percentage of data for training – Proportion (in %) of data for training
• Parameters for variable selection – Parameters for variable selection
  • Number of selected variables – Number of selected variables
  • Variable selection criterion – Please, determine variable selection criterion
  • Variable selection type – Please, determine variable selection type
• Parameters for outlier detection – Parameters for outlier detection
  • Multiplier for sigma, t – Observation x is outlier if Abs(x - predicted x) > t * sigma
  • Number of outlier detection steps – Number of outlier detection steps
• Path to output folder – Path to output folder