Cannot supply array of eval_points to EnsembleArray

[arvoelke]

...
  File "c:\users\aaron\ctn\nengo\nengo\networks\ensemblearray.py", line 55, in __init__
    self.config[nengo.Ensemble].update(ens_kwargs)
  File "c:\users\aaron\ctn\nengo\nengo\config.py", line 105, in update
    setattr(self, key, d[key])
  File "c:\users\aaron\ctn\nengo\nengo\config.py", line 56, in __setattr__
    param.validate(self, value)
  File "c:\users\aaron\ctn\nengo\nengo\dists.py", line 336, in validate
    dist = super(DistOrArrayParam, self).validate(instance, dist)
  File "c:\users\aaron\ctn\nengo\nengo\params.py", line 212, in validate
    "distribution instead." % attr)
ValueError: dimensions not yet initialized; cannot determine if shape is correct. Consider using a distribution instead.

Possible solutions?

• Force people to only use distributions with EnsembleArrays. This seems like an unreasonable limitation to me.

• Include the number of dimensions before the ensembles are configured, i.e.

  self.config[nengo.Ensemble].dimensions = ens_dimensions
  self.config[nengo.Ensemble].update(ens_kwargs)
  

  But this is hacky because this solution has to be repeated for all similar situations with other networks. Also, it's important to note that you can't just do ens_kwargs['dimensions'] = ens_dimensions otherwise you get undefined behaviour, because it relies on the order of dictionary iteration to update dimensions before eval_points, which may or may not happen depending on the alignment of the sun with Neptune.

• Create a new StaticDistribution that always iterates through a fixed list, and then cast eval_points arrays to this distribution when the param is set (thus removing DistOrArrayParam)? I think this is a better long-term solution, assuming I'm not overlooking anything.

Note: Add a unit test for this.

[jgosmann]

Has this been resolved with the merge of #1233?

[jgosmann]

ping @arvoelke

[arvoelke]

Yep!

import nengo
import numpy as np
with nengo.networks.EnsembleArray(3, 2, eval_points=np.zeros((3, 1))) as model:
    for ens in model.ensembles:
        print(ens.eval_points)

Samples(samples=array([[ 0.],
       [ 0.],
       [ 0.]]))
Samples(samples=array([[ 0.],
       [ 0.],
       [ 0.]]))

Thanks.