3threeFX

@matthieu-vergne thanks for your quick response ... mean while we wait for @ajnebro , I want to explain what I did: I believe 3 main methods can Implemented for constrained...

first of all @ajnebro thanks for your response ... I checked that issue and realized that leaders are picked up from non-dominated feasible solutions as you said before and What...

In the latter case of my previous comment about negative values of constrictionCoefficient(c1, c2) I think something like this might happen respect to [updateVelocity()](https://github.com/jMetal/jMetal/blob/master/jmetal-algorithm/src/main/java/org/uma/jmetal/algorithm/multiobjective/smpso/SMPSO.java#L263) :  some negative return values...

> > > @3threeFX , sorry for not responding to your messages. I have not forgotten that, but this week is complicated to me and I have time to take...

> Hi, I'm also very interested in knowing the answer for this question. me too...

> LSTM is not supported in Keras yet But I found another implementation [https://github.com/SciSharp/TensorFlow.NET/blob/master/src/TensorFlowNET.Core/Operations/NnOps/BasicLSTMCell.cs#L14](url) I don’t know whether it works. Is there a timeline for when LSTM will likely be...