print" - With the default gcc libm, exp in float32 is slower then in float64! Try Theano flags floatX=float64 or install amdlibm and set the theano flags lib.amdlibm=True"
print" - With the default gcc libm, exp in float32 is slower then in float64! Try Theano flag floatX=float64, or install amdlibm and set the theano flags lib.amdlibm=True"
print" - You have a dot operation that was not optimized to dot22 that is faster. Make sure the inputs are float32 or 64 and are the same for both input. Currently they are:",[i.typeforiinnode.inputs]
print" - You have a dot operation that was not optimized to dot22 (which is faster). Make sure the inputs are float32 or 64, and are the same for both inputs. Currently they are:",[i.typeforiinnode.inputs]
printed_tip=True
printed_tip=True
#tip 5
#tip 5
...
@@ -544,9 +544,9 @@ Test them first as they are not guaranteed to always provide a speedup."""
...
@@ -544,9 +544,9 @@ Test them first as they are not guaranteed to always provide a speedup."""
node=a[1]
node=a[1]
ifisinstance(node.op,RandomFunction):
ifisinstance(node.op,RandomFunction):
printed_tip=True
printed_tip=True
print" - Replace the default random number generator by 'from theano.sandbox.rng_mrg import MRG_RandomStreams as RandomStreams' as this is is faster. It is still experimental, but seam to work correctly."
print" - Replace the default random number generator by 'from theano.sandbox.rng_mrg import MRG_RandomStreams as RandomStreams', as this is is faster. It is still experimental, but seems to work correctly."
ifconfig.device.startswith("gpu"):
ifconfig.device.startswith("gpu"):
print" - MRG_RandomStreams is the only random number supported on the GPU."
print" - MRG_RandomStreams is the only random number generator supported on the GPU."