Patrick Emami
Patrick Emami
Hi, I've come across this problem as well. Were you able to figure out what was going on?
Hey, thanks for the comments. I am currently in the process of getting results for experiments on the Travelling Salesman Problem, and in the process I've been improving the code...
@ricgama I implemented the LSTM cell in the decoder myself to add the "glimpse" and "pointer" attention components manually. I took inspiration from https://github.com/MaximumEntropy/Seq2Seq-PyTorch/blob/master/model.py
I'm not 100% sure, but I don't think it is the same. The attending and pointing mechanisms, as indicated in the paper, take as input both the encoder LSTM's hidden...
I see, you're calling `forward` repeatedly and passing the last decoder hidden state as input. Yeah that should be equivalent!
@ricgama How is it going for you? I have yet to get results on TSP20. The hyperparameters mentioned in the paper don't seem to be working for me :(
@ricgama Is there a particular difficulty with using Github issues for you? I find communicating via Issues quite nice actually. Yeah for n=10, I tried halving the size of the...
@ricgama are you setting the logits of the already selected nodes during decoding to -infty to ensure only valid tours are used?
@ricgama Also, did you do a distributed implementation? Or single-threaded on a GPU?
Ok, let me know if you find any bugs. Maybe I have a similar one then. I've tried to double and triple check how I implemented the equations, and it...