def testCopyable(self):
net = pybrainbridge._RecurrentNetwork()
self.lstm_network(net)
success = False
try:
copied = net.copy()
success = True
except TypeError, e:
success = False
self.assert_(success, e)
def testParametersDerivatives(self):
rnet = pybrainbridge._RecurrentNetwork()
self.lstm_network(rnet)
rnet.sortModules()
self.assert_(getattr(rnet, '_derivs', None) is not None)
fnet = pybrainbridge._FeedForwardNetwork()
self.two_layer_network(fnet)
fnet.sortModules()
self.assert_(getattr(fnet, '_derivs', None) is not None)
def testCopyable(self):
net = pybrainbridge._RecurrentNetwork()
self.lstm_network(net)
success = False
try:
copied = net.copy()
success = True
except TypeError, e:
success = False
self.assert_(success, e)
def testParametersDerivatives(self):
rnet = pybrainbridge._RecurrentNetwork()
self.lstm_network(rnet)
rnet.sortModules()
self.assert_(getattr(rnet, '_derivs', None) is not None)
fnet = pybrainbridge._FeedForwardNetwork()
self.two_layer_network(fnet)
fnet.sortModules()
self.assert_(getattr(fnet, '_derivs', None) is not None)
def testTimesteps(self):
_net = pybrainbridge._RecurrentNetwork()
self.rec_two_layer_network(_net)
_net.sortModules()
netproxy = _net.proxies[_net]
inproxy = _net.proxies[_net['in']]
outproxy = _net.proxies[_net['out']]
conproxy = _net.proxies[_net.connections[_net['in']][0]]
rconproxy = _net.proxies[_net.recurrentConns[0]]
proxies = netproxy, inproxy, outproxy, conproxy, rconproxy
for proxy in proxies:
self.assertEqual(proxy.get_mode(), 2)
self.assertEqual(_net.offset, 0)
for proxy in proxies:
self.assertEqual(proxy.timestep(), 0,
"%s has wrong timestep." % proxy)
_net.activate((0., 0.))
for proxy in proxies:
self.assertEqual(proxy.timestep(), 1)
_net.activate((0., 0.))
for proxy in proxies:
self.assertEqual(proxy.timestep(), 2)
_net.activate((0., 0.))
for proxy in proxies:
self.assertEqual(proxy.timestep(), 3)
_net.backActivate((0., 0.))
self.assertEqual(_net.offset, 2)
for proxy in proxies:
self.assertEqual(proxy.timestep(), 2)
_net.backActivate((0., 0.))
self.assertEqual(_net.offset, 1)
for proxy in proxies:
self.assertEqual(proxy.timestep(), 1)
_net.backActivate((0., 0.))
self.assertEqual(_net.offset, 0)
for proxy in proxies:
self.assertEqual(proxy.timestep(), 0)
def testTimesteps(self):
_net = pybrainbridge._RecurrentNetwork()
self.rec_two_layer_network(_net)
_net.sortModules()
netproxy = _net.proxies[_net]
inproxy = _net.proxies[_net['in']]
outproxy = _net.proxies[_net['out']]
conproxy = _net.proxies[_net.connections[_net['in']][0]]
rconproxy = _net.proxies[_net.recurrentConns[0]]
proxies = netproxy, inproxy, outproxy, conproxy, rconproxy
for proxy in proxies:
self.assertEqual(proxy.get_mode(), 2)
self.assertEqual(_net.offset, 0)
for proxy in proxies:
self.assertEqual(proxy.timestep(), 0,
"%s has wrong timestep." % proxy)
_net.activate((0., 0.))
for proxy in proxies:
self.assertEqual(proxy.timestep(), 1)
_net.activate((0., 0.))
for proxy in proxies:
self.assertEqual(proxy.timestep(), 2)
_net.activate((0., 0.))
for proxy in proxies:
self.assertEqual(proxy.timestep(), 3)
_net.backActivate((0., 0.))
self.assertEqual(_net.offset, 2)
for proxy in proxies:
self.assertEqual(proxy.timestep(), 2)
_net.backActivate((0., 0.))
self.assertEqual(_net.offset, 1)
for proxy in proxies:
self.assertEqual(proxy.timestep(), 1)
_net.backActivate((0., 0.))
self.assertEqual(_net.offset, 0)
for proxy in proxies:
self.assertEqual(proxy.timestep(), 0)
def convertToFastNetwork(self):
""" Attempt to transform the network into a fast network. If fast networks are not available,
or the network cannot be converted, it returns None. """
from pybrain.structure.networks import FeedForwardNetwork, RecurrentNetwork
try:
from arac.pybrainbridge import _RecurrentNetwork, _FeedForwardNetwork # @UnresolvedImport
except:
print "No fast networks available."
return None
net = self.copy()
if isinstance(net, FeedForwardNetwork):
cnet = _FeedForwardNetwork()
elif isinstance(net, RecurrentNetwork):
cnet = _RecurrentNetwork()
for m in net.inmodules:
cnet.addInputModule(m)
for m in net.outmodules:
cnet.addOutputModule(m)
for m in net.modules:
cnet.addModule(m)
for clist in net.connections.values():
for c in clist:
cnet.addConnection(c)
if isinstance(net, RecurrentNetwork):
for c in net.recurrentConns:
cnet.addRecurrentConnection(c)
try:
cnet.sortModules()
except ValueError:
print "Network cannot be converted."
return None
cnet.owner = cnet
return cnet
def convertToFastNetwork(self):
""" Attempt to transform the network into a fast network. If fast networks are not available,
or the network cannot be converted, it returns None. """
from pybrain.structure.networks import FeedForwardNetwork, RecurrentNetwork
try:
from arac.pybrainbridge import _RecurrentNetwork, _FeedForwardNetwork #@UnresolvedImport
except:
print("No fast networks available.")
return None
net = self.copy()
if isinstance(net, FeedForwardNetwork):
cnet = _FeedForwardNetwork()
elif isinstance(net, RecurrentNetwork):
cnet = _RecurrentNetwork()
for m in net.inmodules:
cnet.addInputModule(m)
for m in net.outmodules:
cnet.addOutputModule(m)
for m in net.modules:
cnet.addModule(m)
for clist in list(net.connections.values()):
for c in clist:
cnet.addConnection(c)
if isinstance(net, RecurrentNetwork):
for c in net.recurrentConns:
cnet.addRecurrentConnection(c)
try:
cnet.sortModules()
except ValueError:
print("Network cannot be converted.")
return None
cnet.owner = cnet
return cnet
def equivalence_recurrent(self, builder):
_net = pybrainbridge._RecurrentNetwork()
builder(_net)
net = RecurrentNetwork()
builder(net)
super(TestNetworkEquivalence, self).equivalence_recurrent(net, _net)
def equivalence_recurrent(self, builder):
_net = pybrainbridge._RecurrentNetwork()
builder(_net)
net = RecurrentNetwork()
builder(net)
super(TestNetworkEquivalence, self).equivalence_recurrent(net, _net)
