Suppose I want to predict the probability of a person to buy something. I want to analyze the person image and I can use a convolutional neural network, but I also want to input in my predictive model how many times he bought something, where he goes more frequently and so on. What is a correct way to input in a deep learning model informations from different deep learning models?


Actually you can merge two different DL model into a big one.

I'm using Keras+Tensorflow for deep learning, and there's an example about models with multiple inputs(outputs). In this example, tweets are fed into a RNN, while the extra data are fed into a fully-connected layer along with the output of this RNN.

  • $\begingroup$ In general I can choose a model I want to merge informations about different inputs or is there some rules? $\endgroup$ – GGA Feb 5 '17 at 8:09
  • $\begingroup$ Personally speaking, we might merge different related data into one model, but it's hard to put the earth into a network and ask for predicting the future. More data leads to a good model, along with the difficulty on model training and tuning. $\endgroup$ – Icyblade Feb 5 '17 at 8:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.