Tool Logistic Regression (OpenCV)

Integration of the OpenCV Machine Learning library for Logistic Regression based classification of gridded features.

Optimization algorithms like Batch Gradient Descent and Mini-Batch Gradient Descent are supported in Logistic Regression. It is important that we mention the number of iterations these optimization algorithms have to run. The number of iterations can be thought as number of steps taken and learning rate specifies if it is a long step or a short step. This and previous parameter define how fast we arrive at a possible solution.

In order to compensate for overfitting regularization can be performed. (L1 or L2 norm).

Logistic regression implementation provides a choice of two training methods with Batch Gradient Descent or the Mini-Batch Gradient Descent.

References

Open Source Computer Vision

Author: O.Conrad (c) 2019
Menu: Imagery|Classification|Machine Learning

Parameters

	Name	Type	Identifier	Description	Constraints
Input	Features	Grid list, input	`FEATURES`	-	-
Input	Training Areas	Shapes, input	`TRAIN_AREAS`	-	-
Output	Classification	Grid, output	`CLASSES`	-	-
Output	Look-up Table (*)	Table, output, optional	`CLASSES_LUT`	A reference list of the grid values that have been assigned to the training classes.	-
Options	Grid System	Grid system	`PARAMETERS_GRID_SYSTEM`	-	-
	Normalize	Boolean	`NORMALIZE`	-	Default: 0
	Update Colors from Features	Boolean, GUI	`RGB_COLORS`	Use the first three features in list to obtain blue, green, red components for class colour in look-up table.	Default: 0
	Load Model	File path	`MODEL_LOAD`	Use a model previously stored to file.	-
	Class Identifier	Table field	`TRAIN_CLASS`	-	-
	Buffer Size	Floating point	`TRAIN_BUFFER`	For non-polygon type training areas, creates a buffer with a diameter of specified size.	Minimum: 0.000000 Default: 1.000000
	Save Model	File path	`MODEL_SAVE`	Stores model to file to be used for subsequent classifications instead of training areas.	-
	Learning Rate	Floating point	`LOGR_LEARNING_RATE`	The learning rate determines how fast we approach the solution.	Minimum: 0.000000 Default: 1.000000
	Number of Iterations	Integer	`LOGR_ITERATIONS`	-	Minimum: 1 Default: 300
	Regularization	Choice	`LOGR_REGULARIZATION`	-	Available Choices: [0] disabled [1] L1 norm [2] L2 norm Default: 0
	Training Method	Choice	`LOGR_TRAIN_METHOD`	-	Available Choices: [0] Batch Gradient Descent [1] Mini-Batch Gradient Descent Default: 0
	Mini-Batch Size	Integer	`LOGR_MINIBATCH_SIZE`	-	Minimum: 1 Default: 1
(*) optional

Command-line

Usage: saga_cmd imagery_opencv 12 [-FEATURES <str>] [-NORMALIZE <str>] [-CLASSES <str>] [-CLASSES_LUT <str>] [-MODEL_LOAD <str>] [-TRAIN_AREAS <str>] [-TRAIN_CLASS <str>] [-TRAIN_BUFFER <double>] [-MODEL_SAVE <str>] [-LOGR_LEARNING_RATE <double>] [-LOGR_ITERATIONS <num>] [-LOGR_REGULARIZATION <str>] [-LOGR_TRAIN_METHOD <str>] [-LOGR_MINIBATCH_SIZE <num>]
  -FEATURES:<str>             	Features
	Grid list, input
  -NORMALIZE:<str>            	Normalize
	Boolean
	Default: 0
  -CLASSES:<str>              	Classification
	Grid, output
  -CLASSES_LUT:<str>          	Look-up Table
	Table, output, optional
  -MODEL_LOAD:<str>           	Load Model
	File path
  -TRAIN_AREAS:<str>          	Training Areas
	Shapes, input
  -TRAIN_CLASS:<str>          	Class Identifier
	Table field
  -TRAIN_BUFFER:<double>      	Buffer Size
	Floating point
	Minimum: 0.000000
	Default: 1.000000
  -MODEL_SAVE:<str>           	Save Model
	File path
  -LOGR_LEARNING_RATE:<double>	Learning Rate
	Floating point
	Minimum: 0.000000
	Default: 1.000000
  -LOGR_ITERATIONS:<num>      	Number of Iterations
	Integer
	Minimum: 1
	Default: 300
  -LOGR_REGULARIZATION:<str>  	Regularization
	Choice
	Available Choices:
	[0] disabled
	[1] L1 norm
	[2] L2 norm
	Default: 0
  -LOGR_TRAIN_METHOD:<str>    	Training Method
	Choice
	Available Choices:
	[0] Batch Gradient Descent
	[1] Mini-Batch Gradient Descent
	Default: 0
  -LOGR_MINIBATCH_SIZE:<num>  	Mini-Batch Size
	Integer
	Minimum: 1
	Default: 1