I would like to know if it is possible to analyze with a DL4J Convolutional neural network model data that includes both static data and time-series data for several different parameters that evolve over time within the same time frame.
…maybe? This description is fairly general. Can you be more specific about your use case?
Normally I see CNNs referenced in relation to image processing, but you don’t mention that above. So maybe LSTMs (or RNNs generally) might be a better choice? It’s hard to say.
Also, have you considered using KNIME’s Keras integration instead of DL4J, or is DL4J a requirement for your task?
Thanks. The data consists on medical patients data both static and dynamic as I described above. The dynamic data are similar to vital signs. Is there any model of the DL4J integration like the Convolutional neural networks that could be used for that?
Vital signs sounds more like time series data, and I’m not sure why you would want to use a CNN for that if there’s not image data involved. Is there a particular reason you want to use CNNs here, and the DL4J version specifically?
If you have sample/dummy data available, and a description of what you are trying to do (forecasting? classification? clustering?) maybe someone can offer a more specific recommendation.