Merge e5a87f9 into 3a06089

aigamedev · Nov 20, 2015 · b56bd4a · b56bd4a
2 parents 3a06089 + e5a87f9
commit b56bd4a
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 6 deletions.
diff --git a/sknn/backend/lasagne/mlp.py b/sknn/backend/lasagne/mlp.py
@@ -276,6 +276,8 @@ def _array_to_mlp(self, array, nn):
         for layer, data in zip(nn, array):
             if data is None: continue
             weights, biases = data
+            weights = weights.astype(theano.config.floatX)
+            biases = biases.astype(theano.config.floatX)
 
             while not hasattr(layer, 'W') and not hasattr(layer, 'b'):
                 layer = layer.input_layer

diff --git a/sknn/tests/test_data.py b/sknn/tests/test_data.py
@@ -4,6 +4,7 @@
 import logging
 
 import numpy
+import theano
 from sknn.mlp import Regressor as MLPR
 from sknn.mlp import Layer as L, Convolution as C
 
@@ -60,8 +61,8 @@ def test_SetLayerParamsList(self):
         nn.set_parameters([(weights, biases)])
 
         p = nn.get_parameters()
-        assert_true((p[0].weights == weights).all())
-        assert_true((p[0].biases == biases).all())
+        assert_true((p[0].weights == weights.astype(theano.config.floatX)).all())
+        assert_true((p[0].biases == biases.astype(theano.config.floatX)).all())
 
     def test_LayerParamsSkipOneWithNone(self):
         nn = MLPR(layers=[L("Sigmoid", units=32), L("Linear", name='abcd')])
@@ -73,8 +74,8 @@ def test_LayerParamsSkipOneWithNone(self):
         nn.set_parameters([None, (weights, biases)])
 
         p = nn.get_parameters()
-        assert_true((p[1].weights == weights).all())
-        assert_true((p[1].biases == biases).all())
+        assert_true((p[1].weights == weights.astype(theano.config.floatX)).all())
+        assert_true((p[1].biases == biases.astype(theano.config.floatX)).all())
 
     def test_SetLayerParamsDict(self):
         nn = MLPR(layers=[L("Sigmoid", units=32), L("Linear", name='abcd')])
@@ -86,5 +87,5 @@ def test_SetLayerParamsDict(self):
         nn.set_parameters({'abcd': (weights, biases)})
 
         p = nn.get_parameters()
-        assert_true((p[1].weights == weights).all())
-        assert_true((p[1].biases == biases).all())
+        assert_true((p[1].weights == weights.astype(theano.config.floatX)).all())
+        assert_true((p[1].biases == biases.astype(theano.config.floatX)).all())