ANNHUB

Simplify machine learning design and deployment

ANNHUB_ui

Key Features​

  1. Easy to use: Does not require deep machine learning knowledge to design a neural network. No coding required.
  2.  Support advanced training algorithms: Scaled Conjugate Gradient, Levenberg Marquardt, Quasi Newton, Bayesian Regularization.
  3. Support different types of activation functions and cost functions
  4. Support Confusion matrix, ROC curves, regression curves, performance metrics to evaluate trained neural network model(s).
  5. Easy to deploy trained neural network model(s) to LabVIEW, LabWindow CVI, C, C++, C#, iOS, and Android applications.

Choose your version

ANNHub is available in three versions

Student

$ 49
/ license
  • Easy to use. No coding required.
  • Support advanced training algorithms.
  • Support Bayesian Neural Network.
  • Support different types of activation functions and cost functions
  • Support confusion matrix, ROC curvers, regression curves
  • Export results to high quality images ready for publications
  • Easy to deploy trained neural network models into C, C++, C#, Androd, iOS and LabVIEW applications.
  • Support an evidence framework to find an optimal neural network.

standard

$ 299
/ license
  • Easy to use. No coding required.
  • Support advanced training algorithms.
  • Support Bayesian Neural Network.
  • Support different types of activation functions and cost functions.
  • Support confusion matrix, ROC curvers, regression curves.
  • Export results to high quality images ready for publications.
  • Easy to deploy trained neural network models into LabVIEW applications. LabVIEW API is provided.
  • Support an evidence framework to find an optimal neural network.
Popular

professional

$ 499
/ license
  • Easy to use. No coding required.
  • Support advanced training algorithms.
  • Support Bayesian Neural Network.
  • Support different types of activation functions and cost functions
  • Support confusion matrix, ROC curvers, regression curves
  • Export results to high quality images ready for publications
  • Easy to deploy trained neural network models into C, C++, C#, Androd, iOS and LabVIEW applications.
  • Support an evidence framework to find an optimal neural network.

Five steps

To design and deploy a neural network into your applications

Machine Learning Applications

Pattern recognition
More information
Time series prediction
More information

DEPLOYing neural network

Deploying a neural network into applications has never been easier. Following steps are required for neural network deployment.

  1. Step 1: Export a trained Neural Network Model into a text file with “ann” extension.
  2. Step 2: Download supported Application Programming Interfaces (APIs) from Customer Portal and install these APIs in  supported programming environment such as LabVIEW, LabWindow CVI, C, C++, C#, iOS and Android.
  3. Step 3: Use supported Application Programming Interfaces (APIs) to import a trained model in “ann” extension text file into a machine learning application developed in supported programming environment.
  4. Step 4: Compile and deploy this machine learning application.
Step 1: Export a trained model
Export a trained Neural Network Model into a text file with "ann" extension.
Step 2: Download APIs
Download supported Application Programming Interfaces (APIs) from Customer Portal and install these APIs in supported programming environment such as LabVIEW, LabWindow CVI, C, C++, C#, iOS and Android.
Step 3: Use API to import a trained model
Use supported Application Programming Interfaces (APIs) to import a trained model in "ann" extension text file into a machine learning application developed in supported programming environment such as LabVIEW, LabWindow CVI, C, C++, C#, iOS and Android.
Step 4: Deployment
Compile and deploy this machine learning application.

More information

Please subscribe to receive free tutorials on machine learning and deep learning.

© ANSCenter 2018. All rights reserved