Check groups attribute existence in convolution op ONNX export.

Source/CNTKv2LibraryDll/proto/onnx/CNTKToONNX.cpp

@@ -3198,7 +3198,7 @@ void CNTKToONNXHelper::CopyAttributes(const FunctionPtr& src, ONNXIR::Node* node
             auto autoPadding = AsVector<bool>(src->Attributes()[L"autoPadding"].Value<std::vector<DictionaryValue>>());
             auto dilations = (NDShape)src->Attributes()[L"dilation"].Value<NDShape>();
             auto transpose = (bool)src->Attributes()[L"transpose"].Value<bool>();
-            auto groups = (size_t)src->Attributes()[L"groups"].Value<size_t>();
+            size_t groups = (src->Attributes().Contains(L"groups")) ? (size_t)src->Attributes()[L"groups"].Value<size_t>() : 1u;
 
             //
             // Remove the channel part for ONNX. This is because ONNX, unlike CNTK, does