Skip to main content

ANNHUB Model Training

Load dataset​

Once you complete the data preparation, browse to the training data file (.csv) and import to ANNHUB

info

ANNHUB will display a small data sample to visualize how it looks

  • After the dataset is loaded, click Configure Neural Network to continue.

Configure neural network​

  • In this step, ANNHUB will auto-suggest the neural network type and parameters based on the imported dataset, or you can adjust the setting to achieve the best result.
  • Click Create Neural Network to save your setting
  • Click Train Neural Network to continue

Train neural network​

  • In this step, ANNHUB will also auto-suggest the training parameter, or you can adjust the setting to achieve the best result
  • Click Train Neural Network to begin the training process
  • After training is completed, you can re-adjust the training parameter and press Retrain Neural Network if needed
  • Click Evaluate Neural Network to continue

Evaluate neural network​

ANNHUB provides many evaluation techniques that help users evaluate the trained model.

Pattern recognition application​

ANNHUB supports Confusion Matrix and ROC curve techniques to evaluate the trained model.

The Confusion Matrix for training, validation, and test datasets is presented in great detail.

Valuable information such as the total number of samples in a specific class is correctly identified and how many samples are miss-classified.

This information is summarized in the Accuracy, Sensitivity, and Specificity columns.

Function approximation/fitting application​

ANNHUB supports the Regression curve that presents how well the predicted outputs approximate their targets.

The Regression curves for training, validation, and test dataset are all available.

info

If the Bayesian Regularization training algorithm is selected, the evaluation information for the validation set will not be displayed as that dataset is not required.

Test model with unknown data (optional)​

After evaluation, the neural network can still be tested to ensure it can cope with an entirely new dataset.

Depending on your Neural Network application, appropriate evaluation techniques will be used.

info

You can import the .cvs file or manually test the model

Test pattern recognition model​

Test function fitting/ prediction model​

Export neural network for deployment​

Click on Export Neural Network to export the model as .zip file and can be loaded directly into C++, C#, Python for deployment