Manually Loading Weights

CodeReclaimers / neat-python

Python implementation of the NEAT neuroevolution algorithm

BSD 3-Clause "New" or "Revised" License

1.42k stars 493 forks source link

Manually Loading Weights #140

Open blaeberry opened 6 years ago

blaeberry commented 6 years ago

I have pretrained my network in Keras, and was looking to somehow take the saved h5 model and load that into NEAT-Python to do further training. Is there any way to do that? I was thinking about getting the actual connection weights, and then somehow loading those into a genome or something. Any ideas would be greatly appreciated.

tunstek commented 5 years ago

I'm looking to do the same. Did you find any solution?