How to test ANN model

I have a trained ANN model , I want to check the output of the ANN model having certain set of input and compare it with experiimental data. help me

Answer 1

[Krishna]

Hello Sunita,

Based on the details provided, it seems you possess a trained artificial neural network (ANN) model and are seeking guidance on applying this model to predict outcomes for test data. Additionally, you wish to evaluate the model's performance by assessing its accuracy or calculating the mean squared error in comparison to your actual data.

Given that the specific release version wasn't specified, it's presumed that you're utilizing the most recent version of MATLAB and that your model has been trained using dlnetworks. For regression tasks with an ANN, you can employ the predict function to estimate the output of your test data. Conversely, if your ANN is intended for classification tasks, the classify function would be appropriate for making predictions with dlnetworks. Please have a look at the documentation for predict and classify,

https://www.mathworks.com/help/stats/linearmodel.predict.html

https://www.mathworks.com/help/deeplearning/ref/seriesnetwork.classify.html

Please go through this example to lean model how to use networks using dlnetworks,

https://www.mathworks.com/help/deeplearning/ug/create-simple-deep-learning-network-for-classification.html

To create ANN using dlnetworks you can use ‘fullyconnectedlayer’, please refer the documentation for more information,

https://www.mathworks.com/help/deeplearning/ref/nnet.cnn.layer.fullyconnectedlayer.html

Hope this helps.