How to predict values of a timeseries using tensorflow.js?

See more:

Javascript

tensorflow

How do i have to adapt the predict function in @arthurratz example TensorFlow.js: Predicting Time Series Using Recurrent Neural Networks (RNN) With Long Short-Term Memory (LSTM) Cells[^] that i will predict the n+1 window?

So far i understand that the algorithm starts predicting, depending on the size of the .slice(), and determines sma values.

How can we get "future" values, beyond the last calculated sma?

What I have tried:

So far i have analized the parameters on the console.log. Te function is the following

function Predict(inputs, size, model) 
{
    var inps = inputs.slice(Math.floor(size / 100 * inputs.length), inputs.length);

    const outps = model.predict(tf.tensor2d(inps, [inps.length, 
	inps[0].length]).div(tf.scalar(10))).mul(10);

    return Array.from(outps.dataSync());
}