partial rewrite of the ConvOp grad test. It was not testing in float32…

theano/sandbox/test_conv.py

@@ -420,65 +420,52 @@ class TestConvOp(unittest.TestCase):
 
 
     def test_ConvOpGrad(self):
-        nkern = 3
-        bsize = 2
-        imgs  = T.dmatrix('imgs')
-        kerns = T.dmatrix('kerns')
-        kshps = [(3,3), (5,5)]
-
-        for mode in 'valid', 'full':
-
-            for imshp in (5,5),(2,3,3),(3,6,6): # (12,10), (3,12,11):
-                # 'full' mode should support kernels bigger than the input
-                if mode == 'valid' and (kshps[0] > imshp[1]):
-                    continue
-
-                visdim = 1 if len(imshp)!=3 else imshp[0]
-                for kshp in kshps:
-                    imgvals = N.random.random(N.hstack((bsize,imshp)))
-                    print 'imgvals.shape = ', imgvals.shape
-                    imgvals = imgvals.reshape(bsize,-1)
-
-                    if visdim == 1: 
-                        kernvals = N.random.rand(nkern,kshp[0],kshp[1])
-                    else:
-                        kernvals = N.random.rand(nkern,visdim,kshp[0],kshp[1])
-                    kernvals = kernvals.reshape(nkern,-1)
-
-                    def testf(imgs, kerns):
-                        out, outshp = convolve2(kerns, kshp, nkern, imgs, 
-                                                   imshp, bsize, mode=mode)
-                        return out
-
-                    try:
-                        utt.verify_grad(testf, [imgvals, kernvals])
-                    except NotImplementedError, e:
-                        print e
-
-    def test_ConvOpGrad32(self):
+        """
+        test the gradient in float and double
+        """
         nkern = 4
         bsize = 3
-        imgs  = T.fmatrix('imgs')
-        kerns = T.fmatrix('kerns')
-     
-        def testf(imgs, kerns):
-            out, outshp = convolve2(kerns, kshp, nkern, imgs, 
-                                       imshp, bsize, mode='valid')
-            return out
-            
-        for mode in 'valid', 'full':
-            for imshp in (1,5,5),(2,10,10): # (12,10), (3,12,11):
-                visdim = 1 if len(imshp)!=3 else imshp[0]
-                for kshp in (3,3),:# (6,7):
-                    imgvals = N.random.random(N.hstack((bsize,imshp)))
-                    print 'imgvals.shape = ', imgvals.shape
-                    imgvals = imgvals.reshape(bsize,-1)
-
-                    kernvals = N.random.rand(nkern,visdim,kshp[0],kshp[1])
-                    print 'kernvals.shape = ', kernvals.shape
-                    kernvals = kernvals.reshape(nkern,-1)
-
-                    utt.verify_grad(testf, [imgvals, kernvals])
+        types = ["float32", "float32"]#flot32 is BUGGED
+        kshps = [(3,3),(5,5)]# TODO: (6,7)show a bug that is not fixed!!
+        imshps = [(1,5,5),(2,10,10),(2,12,12)] #TODO bugged (2,12,10), (3,12,11):
+        modes = ['valid', 'full']
+        unroll_batch=[0,1,3]
+        unroll_kern=[0,1,2,4]
+        
+        for t in types:
+            imgs  = T.TensorType(t, (False, False, False, False),'imgs')
+            kerns = T.TensorType(t, (False, False, False, False),'kerns')
+            for mode in modes:
+                for imshp in imshps:
+                    visdim = 1 if len(imshp)!=3 else imshp[0]
+                    for kshp in kshps:
+                        t=numpy.array([imshp[1]-kshp[0],imshp[2]-kshp[1]])
+                        # 'full' mode should support kernels bigger than the input
+                        if mode == 'valid' and (t<0).any():
+                            continue
+                        for un_b in unroll_batch:
+                            for un_k in unroll_kern:
+                                imgvals = N.array(N.random.random(N.hstack((bsize,imshp))),dtype=imgs.dtype)
+#                                print 'imgvals.shape = ', imgvals.shape, imgvals.dtype
+#                                imgvals = imgvals.reshape(bsize,-1)
+
+                                kernvals = N.array(N.random.rand(nkern,visdim,kshp[0],
+                                                         kshp[1]),dtype=kerns.dtype)
+
+#                                print 'kernvals.shape = ', kernvals.shape, kernvals.dtype
+#                                kernvals = kernvals.reshape(nkern,-1)
+
+                                def testf(imgs, kerns):
+                                    out, outshp = convolve2(kerns, kshp, nkern, 
+                                                            imgs, imshp, bsize, 
+                                                            mode=mode,
+                                                            unroll_batch=un_b,
+                                                            unroll_kern=un_k)
+                                    print "2",imgs.dtype,kerns.dtype
+                                    out = ConvOp(imshp,kshp,nkern,bsize,1,1,mode)(imgs,kerns)
+                                    return out
+
+                                utt.verify_grad(testf, [imgvals, kernvals])
 
 if __name__ == '__main__':
 #    t = TestConvOp("test_convolution")