Return to the Neural Network Software Page

THINKS and ThinksPro (TP) Neural Network Software

General NN Software Features

Intuitive, interactive and highly responsive user interface
Built-in spreadsheets for convenient editing of data
Dynamic graphing and data displays continuously illustrate the neural network training progress
Industry standard, documented, field-proven algorithms, make these products the perfect tool for students and professionals
Ideal for forecasting, prediction, signal processing, classification, function approximation, decision making, sensor interpretation, and nonlinear regression

Maximum Flexibility in 6 Neural Networks Function Categories

Preprocessing Functions
Interconnection Architectures
Processing Element Summation Functions
Processing Element Transfer Functions
Learning Algorithms
Learning Error Criteria

Preprocessing Functions

Thinks provides special input layer preprocessing to simplify working with all forms of data.

Min/Max and Mean/Std Input Preprocessing
Sum Inputs to 1 Normalization
Sum of Squares to 1 Normalization
Input Disabling (TP)
Time Series Inputs (TP)
Input Added Noise (TP)

Processing Element Summation Functions

The function can be chosen independently on each layer of the network.

Dot Product Quadratic Sum
L1 Distance L2 Distance
Radial Basis Function Sigma-Pi
GRNN Sum (TP)

Processing Element Transfer Functions

Thinks offers greater efficiency and training with 14 neural network transfer functions.

Sigmoid Bipolar Sigmoid
Arctan Bipolar Arctan
Sin Bipolar Sin
Threshold Linear Bipolar Threshold Linear
Threshold Bipolar Threshold
Linear Gaussian
Cauchy Winner Take All
Stochastic Threshold (Hopfield Net TP)
Mean Field Threshold (Hopfield Net TP)

Interconnection Architectures

Processing elements can be connected in neural networks up to five layers.

Multilayer Normal Feed Forward
Multilayer Full Feed Forward
Total Recurrent
Cascade
Prior Recurrent
Cascade Recurrent
Jordan Recurrent (TP)
Elman Recurrent (TP)

Learning Algorithms

The Thinks neural network system supports these industry standard algorithms:

Back Propagation
Quick Propagation
Jacob's Enhanced Back Propagation
Recurrent Back Propagation
Kohonen Winner Take All
Kohonen Learning Vector Quantization
Simulated Annealing
Solis & Wets Simulated Annealing (TP)
Cascade Correlation Learning
Simplex Simulated Annealing (reinforcement learning) (TP)
Powell's Method (reinforcement learning) (TP)
Conjugate Gradient Training (TP)
Probabilistic Neural Network (TP)
General Regression Neural Network (TP)
Levenberg-Marquardt Training (TP)
Temporal Differences Algorithm (TP)
Hopfield (TP)
Hopfield Mean Field Annealing (TP)

Learning Error Criteria

Control the network's sensitivity to outliers in the training data with five error function options.

Mean Squared Error
Mean Absolute Error
Mean Fourth Power Error
Hyperbolic Square Error
Bipolar Hyperbolic Square Error
Classification Error (TP)
User Defined using EvalNet DLL (TP)
Error Tolerance Training (TP)

The ThinksPro neural network system contains the following additional features

Adding Noise to inputs during training
Tolerance training
Selective deactivation of inputs
Time series window on inputs
Sensitivity Analysis of inputs
Confusion Matrix for classification problems
Reinforcement Learning with user defined DLL for net evaluation
Core network functions DLL at no additional charge
Code Generation of run time application development
TrainDos 32bit Extended Dos Appl. for Batch mode training.
ThinksPro Views will allow viewing of Inputs, Weights, States and Outputs as Time series, Hinton or Color diagrams.

Thinks and ThinksPro take the confusion out of the choice of neural network options and training by providing demonstration training files and examples. Each option is clearly described in on-line help files. The documentation includes a run-time library reference for each of the available functions.

Containing some of the most sophisticated neural network algorithms and capabilities available today, these products provide a window into how networks train and work. With THINKS or ThinksPro, you can be training your neural network in minutes.

This page has been accessed

times within the last 30 days.