tcstewar

Support for changing a learning rate

8

There currently doesn't seem to be a way to change the learning rate for a learning rule after it's been created. Suppose I have this model: ``` import nengo model...

Initialize state of a built model based on the state of a previous model

2

I was just thinking that there are a few situations where it'd be handy to be able to do this: ```python model = nengo.Network() with model: ... # define the...

Improve gui experience for deep learning

1

Now that nengo is starting to support interesting deep learning (and reservoir computing) components, what can we do to improve the user interface side of these things? This issue is...

Improve gui experience for large models (like Spaun)

1

Right now, it's pretty awkward to work with large models. This issue is just meant to brainstorm some possibilities of what we can do to improve things. Some initial ideas:...

Add Python3 to TravisCI testing

2

To make sure we keep Python3 support, we should also run our tests with it (like nengo does).

Style guidelines

17

Once 0.2 is done, that'd be a great time to do a pass through and fix formatting. This would require: - [ ] make sure all the python code passes...

If you try setting the range to (2,3), it will give an error.

Display of Connection types

6

It'd be nice to have a few different Connection types displayed in some way. For example: - normal NEF connections - inhibitory connections - excitatory connections (?) - learning connections...

Setting values in Semantic Pointer clouds needs a better controller

3

When you do a "Set value..." on a Semantic pointer cloud, it doesn't quite reach the value you've asked it to go to. For example, try this minimal network: ```...