My main problem is to save several NN into a octree structure. I have created a octree, with a very simple definition :
class Tree(object): def __init__(self): self.childs =  self.model = None self.depth = None self.minRightValue = torch.ones(3) self.maxLeftValue = torch.ones(3) def createChildren(self,amount): for i in range(0,amount): self.childs.append(Tree())
Where tree.model is a trained Neural network.
This definition works quite well. However, what I want to do is to save the octree into one file where each child have is a trained NN.
The only idea I have for now is to save each NN with a complex naming convention to read each of them and create the octree in another computer. But I think there should be easier way to do it.
Any ideas? This is my first time using octrees, so if there is any recommendation I would love one.