---
name: "pkregann"
section: "1"
platform: "linux"
package: "pktools"
version: "2.6.7.6+ds-6+b3"
release: "Debian 13 (Trixie)"
platforms: ["linux"]
url: "https://new.carta.tech/pkregann"
---

# pkregann(1)

regression with artificial neural network (multi-layer perceptron)

## NAME

pkregann - regression with artificial neural network (multi-layer perceptron)

## SYNOPSIS

FC**pkregann**F\[0\]  FC**-i**F\[0\] *input* FC**-t**F\[0\] *training* \[FC**-ic**F\[0\] *col*\] \[FC**-oc**F\[0\] *col*\] FC**-o**F\[0\] *output* \[*options*\] \[*advanced options*\]

## DESCRIPTION

**pkregann** performs a regression based on an artificial neural network. The regression is trained from the input (FC**-ic**F\[0\]) and output (FC**-oc**F\[0\]) columns in a training text file. Each row in the training file represents one sampling unit. Multi-dimensional input features can be defined with multiple input options (e.g., FC**-ic**F\[0\] *0* FC**-ic**F\[0\] *1* FC**-ic**F\[0\] *2* for three dimensional features).

## OPTIONS

- FC**-i**F\[0\] *filename*, FC**--input**F\[0\] *filename*

  input ASCII file

- FC**-t**F\[0\] *filename*, FC**--training**F\[0\] *filename*

  training ASCII file (each row represents one sampling unit. Input features should be provided as columns, followed by output)

- FC**-o**F\[0\] *filename*, FC**--output**F\[0\] *filename*

  output ASCII file for result

- FC**-ic**F\[0\] *col*, FC**--inputCols**F\[0\] *col*

  input columns (e.g., for three dimensional input data in first three columns use: FC**-ic**F\[0\] *0* FC**-ic**F\[0\] *1* FC**-ic**F\[0\] *2*

- FC**-oc**F\[0\] *col*, FC**--outputCols**F\[0\] *col*

  output columns (e.g., for two dimensional output in columns 3 and 4 (starting from *0*) use: FC**-oc**F\[0\] *3* FC**-oc**F\[0\] *4*

- FC**-from**F\[0\] *row*, FC**--from**F\[0\] *row*

  start from this row in training file (start from 0)

- FC**-to**F\[0\] *row*, FC**--to**F\[0\] *row*

  read until this row in training file (start from 0 or set leave 0 as default to read until end of file)

- FC**-cv**F\[0\] *size*, FC**--cv**F\[0\] *size*

  n-fold cross validation mode

- FC**-nn**F\[0\] *number*, FC**--nneuron**F\[0\] *number*

  number of neurons in hidden layers in neural network (multiple hidden layers are set by defining multiple number of neurons: FC**-n**F\[0\] *15* FC**-n**F\[0\] *1*, default is one hidden layer with 5 neurons)

- FC**-v**F\[0\] *level*, FC**--verbose**F\[0\] *level*

  set to: 0 (results only), 1 (confusion matrix), 2 (debug)

Advanced options

- FC**--offset**F\[0\] *value*

  offset value for each spectral band input features: FCrefl\[band\]=(DN\[band\]-offset\[band\])/scale\[band\]F\[0\]

- FC**--scale**F\[0\] *value*

  scale value for each spectral band input features: FCrefl=(DN\[band\]-offset\[band\])/scale\[band\]F\[0\] (use FC0F\[0\] if scale min and max in each band to FC-1.0F\[0\] and FC1.0F\[0\])

- FC**--connection**F\[0\] *rate*

  connection rate (default: 1.0 for a fully connected network)

- FC**-l**F\[0\] *rate*, FC**--learning**F\[0\] *rate*

  learning rate (default: 0.7)

- FC**--maxit**F\[0\] *number*

  number of maximum iterations (epoch) (default: 500)