Number of hidden and output neurons for different dataset

[Abrar2652]

I have got a dataset of shape (2820520, 15), which has 14 independent features and 1 binary target feature. I'm performing SMOTE instead of undersampling on this QNN model using that particular dataset. What should be the number of neurons in this case? Is there anything that I need to modify for that particular dataset?

[Note: The features are not PCAs like credit card data]

[CatalinaAlbornoz]

Hi @Abrar2652, 14 neurons should do the job. Just to be sure you can test with a smaller subset of your dataset since you have over 2 million datapoints.

We can probably continue this conversation on our discussion forum.