How to find optimized x values (input features) after training in deep learning?

Question

I did deep learning training by Keras.

I have done the training part by model.fit If I do model.predict, it only gives me y value. But I want to know x (input features) that gives the best y value, which means the optimized x values after training.

Is there any way to fetch the optimized x values (input features) in Keras or else?

I really appreciate your help.

Thank you.

ggaugler · Answer

If I understand your question correctly, you want to find the best features to optimize your model's performance?

I think what you're looking for is feature selection, which is an entire subject within machine learning/statistics. There is a whole tag on here about feature selection, you'd be better off asking your question with that. I would poke around a little online to learn more about feature selection, and if after that you still want help here you should provide more information about the problem you're trying to solve, the data you're working with, and more details about your model.

Henrik · Answer

The model doesn't know which X values would yield the "best" result.

You could run the model a couple of times (tf.predict()) with dummy data, record the results (inputs and predictions), build a Genetic ML model that optimizes X combinations for high (if that's what you consider "good") Y values.

How to find optimized x values (input features) after training in deep learning?

2 Answers

Add your own answers!

Ask a Question