Update on "[Lite Interpreter] Enable __setstate__ "

Martin Yuan · Martin Yuan · commit 7d2d4988ff0c · 2020-03-05T11:30:36.000-08:00
1. Serialize bytecode of __setstate__ and run it when loading the model. 2. One use case is quantization. To test this use case a few operators are registered temporarily for lite interpreter. The "_" prefix registration will be removed when the operators are all migrated to mobile. Differential Revision: [D20162898](https://our.internmc.facebook.com/intern/diff/D20162898)
diff --git a/torch/csrc/jit/serialization/export_module.cpp b/torch/csrc/jit/serialization/export_module.cpp
@@ -41,7 +41,7 @@ static IValue Table(const std::vector<std::pair<std::string, IValue>>& entries)
 c10::IValue getFunctionTuple(const Function& func) {
   auto graph = func.graph()->copy();
   Inline(*graph);
-  torch::jit::Code code(graph);
+  torch::jit::Code code(graph, func.name());
 
   // operator names
   std::vector<c10::OperatorName> opnames;
