add numgroups to conv3d grad functions

theano/tensor/nnet/abstract_conv.py

@@ -912,7 +912,8 @@ def conv3d_grad_wrt_inputs(output_grad,
                            border_mode='valid',
                            subsample=(1, 1, 1),
                            filter_flip=True,
-                           filter_dilation=(1, 1, 1)):
+                           filter_dilation=(1, 1, 1),
+                           num_groups=1):
     """Compute conv output gradient w.r.t its inputs
 
     This function builds the symbolic graph for getting the
@@ -1035,7 +1036,8 @@ def conv3d_grad_wrt_inputs(output_grad,
                                               border_mode=border_mode,
                                               subsample=subsample,
                                               filter_flip=filter_flip,
-                                              filter_dilation=filter_dilation)
+                                              filter_dilation=filter_dilation,
+                                              num_groups=num_groups)
 
     return grad_input_op(filters, output_grad, input_shape[-3:])
 
@@ -1179,7 +1181,8 @@ def conv3d_grad_wrt_weights(input,
                             border_mode='valid',
                             subsample=(1, 1, 1),
                             filter_flip=True,
-                            filter_dilation=(1, 1, 1)):
+                            filter_dilation=(1, 1, 1),
+                            num_groups=1):
     """Compute conv output gradient w.r.t its weights
 
     This function will build the symbolic graph for getting the
@@ -1293,7 +1296,8 @@ def conv3d_grad_wrt_weights(input,
                                                border_mode=border_mode,
                                                subsample=subsample,
                                                filter_flip=filter_flip,
-                                               filter_dilation=filter_dilation)
+                                               filter_dilation=filter_dilation,
+                                               num_groups=num_groups)
 
     return gradWeight_op(input, output_grad, filter_shape[-3:])