0
$\begingroup$

I am just starting off with TensorFlow and trying to implement an RNN for a dataset which consists of 11 Features (all numeric). These features will be used to predict the output of another column.

I am currently lost on where to start and tho I am able to understand how a RNN functions all the tutorials I could find were mainly related to image and text datasets.

$\endgroup$
3
  • $\begingroup$ Answering the titular question: tensorflow.org/programmers_guide/datasets $\endgroup$
    – Emre
    Nov 15 '17 at 19:50
  • $\begingroup$ Does each row of your dataset represent a point in time? If not, RNNs are not what you want. If yes, please clarify. $\endgroup$
    – tom
    Nov 15 '17 at 21:40
  • $\begingroup$ @tom yes, it does. These are readings taken of a patient every second. $\endgroup$ Nov 16 '17 at 1:32
2
$\begingroup$

There is no fundamental difference between using RNNs for text vs. for numeric values. In fact, text is more difficult because you have to preprocess the text to convert into numeric values first. In any case you might have a look at this tutorial:

https://machinelearningmastery.com/time-series-prediction-lstm-recurrent-neural-networks-python-keras/

I find this site to be pretty useful in general for examples of working with Keras.

$\endgroup$

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.