Artificial Neural Networks recommendation

[Gereon Kaiping]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I'd like to use ANNs in python, especially Simple Recurrent Networks.

Ideally I'd like to find a quick, pythonic module that is able to
simulate different styles of network architectures including some types
of recurrent networks (must: SRN/Elman, may: other recurrent networks
like ESN or LSTM to experiment with them).

So far I found the following Python ANN modules:
- - PyNN (just a builder, requires external simulator) –
http://neuralensemble.org/trac/PyNN/
- - Con-x (part of pyro) – http://pyrorobotics.org/?page=Conx
- - PyNeurGen (includes genetic algorithms) –
http://pyneurgen.sourceforge.net/
- - ffnet (only fast forward) – http://ffnet.sourceforge.net/
- - brian (spiking networks) – http://www.briansimulator.org/
- - PyBrain (machine learning) – http://pybrain.org/

Can you give me a recommendation or a review on some of these? Are there
other modules for simulating ANNs?

Greetings,
Gereon

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkuFIHIACgkQFWJnZLsO/Wvp1ACfa2dhOd0b0SIVkOzZN0ebRJdd
WFQAoIttIDNbIuqViDbPeVyS+wtGj6tI
=N1oM
-----END PGP SIGNATURE-----