Mercurial > pylearn

changeset 924:be420f1836bb

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
make tests faster in debug mode
author Frederic Bastien <nouiz@nouiz.org>
date Tue, 30 Mar 2010 13:46:08 -0400
parents 06cc2b0c06ea
children 6be337c30928
files pylearn/algorithms/tests/test_exponential_mean.py pylearn/algorithms/tests/test_sgd.py
diffstat 2 files changed, 13 insertions(+), 7 deletions(-) [+]
line wrap: on
line diff
--- a/pylearn/algorithms/tests/test_exponential_mean.py	Tue Mar 30 12:09:11 2010 -0400
+++ b/pylearn/algorithms/tests/test_exponential_mean.py	Tue Mar 30 13:46:08 2010 -0400
@@ -1,4 +1,5 @@
 import theano, numpy
+from theano.compile.debugmode import DebugMode
 from pylearn.algorithms import exponential_mean
 
 def test_mean():
@@ -50,9 +51,9 @@
     assert i > rows_to_test
 
 def test_dynamic_normalizer():
-    m = theano.compile.mode.get_defautl_mode()
-    if isinstance(m,DebugMode):
-        m='FAST_RUN'
+    mode = theano.compile.mode.get_default_mode()
+    if isinstance(mode,DebugMode):
+        mode = 'FAST_RUN'
     x = theano.tensor.dvector()
 
     rows_to_test = 100
@@ -79,7 +80,7 @@
 
     M.f = theano.Method([x], [D.output, M.dn_mean.curval, M.dn_var.curval, M.x_mean.curval] , updates)
 
-    m = M.make(mode=m)
+    m = M.make(mode=mode)
     m.dn.initialize()
     m.dn_mean.initialize()
     m.dn_var.initialize()
--- a/pylearn/algorithms/tests/test_sgd.py	Tue Mar 30 12:09:11 2010 -0400
+++ b/pylearn/algorithms/tests/test_sgd.py	Tue Mar 30 13:46:08 2010 -0400
@@ -1,6 +1,11 @@
 import theano
+from theano.compile.debugmode import DebugMode
 from pylearn.algorithms import sgd
 
+mode = theano.compile.mode.get_default_mode()
+if isinstance(mode,DebugMode):
+    mode = 'FAST_RUN'
+
 def test_sgd0():
 
     x = theano.tensor.dscalar('x')
@@ -8,7 +13,7 @@
 
     M = sgd.StochasticGradientDescent([x], (1.0 - x * y)**2, [y], stepsize=0.01)
     M.y = y
-    m = M.make()
+    m = M.make(mode=mode)
     m.y = 5.0
     for i in xrange(100):
         c = m.step_cost(3.0)
@@ -26,7 +31,7 @@
     M = sgd.StochasticGradientDescent([x], (1.0 - x * y)**2, [y], stepsize=lr)
     M.y = y
     M.lr = lr
-    m = M.make()
+    m = M.make(mode=mode)
     m.y = 5.0
     m.lr = 0.01
     for i in xrange(100):
@@ -54,7 +59,7 @@
 
     M = sgd.StochasticGradientDescent([x], (1.0 - x * y)**2, [y])
     M.y = y
-    m = M.make()
+    m = M.make(mode=mode)
     m.y = 5.0
     #there should be a learning rate here by default
     assert m.stepsize is None