Use concrete types on call sites for Dict/List (pytorch#22004)

Summary: Pull Request resolved: pytorch#22004 In future, we want all dicts/lists to store information about the types they contain. This is only possible if the creation API doesn't allow creating lists/dicts without type information. This diff removes some call sites that don't specify type information and have it specify type information. Reviewed By: dzhulgakov Differential Revision: D15906387 fbshipit-source-id: 64766a2534b52c221e8a5501a85eaad13812e7bd
IvanKobzarev · Jul 2, 2019 · 6d58713 · 6d58713
1 parent 693871d
commit 6d58713
Show file tree

Hide file tree

Showing 5 changed files with 4 additions and 7 deletions.
diff --git a/aten/src/ATen/core/ivalue.h b/aten/src/ATen/core/ivalue.h
@@ -237,7 +237,6 @@ struct CAFFE2_API IValue final {
   c10::ArrayRef<at::Tensor> toTensorListRef() const;
 
   //GenericList
-  IValue(std::vector<IValue> v);
   IValue(c10::List<IValue> v);
   bool isGenericList() const { return Tag::GenericList == tag; }
   c10::List<IValue> toGenericList() &&;

diff --git a/aten/src/ATen/core/ivalue_inl.h b/aten/src/ATen/core/ivalue_inl.h
@@ -593,8 +593,6 @@ inline IValue::IValue(c10::impl::GenericList v)
 : tag(Tag::GenericList), is_intrusive_ptr(true) {
   payload.as_intrusive_ptr = v.impl_.release();
 }
-inline IValue::IValue(std::vector<IValue> v)
-: IValue(c10::impl::toList(std::move(v))) {}
 
 template<class T> inline IValue::IValue(c10::List<T> v)
 : IValue(impl::toGenericList<T>(std::move(v))) {
@@ -619,7 +617,7 @@ inline IValue::IValue(c10::Dict<Key, Value> v)
 : IValue(impl::toGenericDict(std::move(v))) {}
 
 template<class Key, class Value> inline IValue::IValue(std::unordered_map<Key, Value> v)
-: IValue(impl::GenericDict()) {
+: IValue(Dict<Key, Value>()) {
   auto dict = to<c10::Dict<Key, Value>>();
   dict.reserve(v.size());
   for (auto& e : v) {

diff --git a/test/cpp/api/jit.cpp b/test/cpp/api/jit.cpp
@@ -93,7 +93,7 @@ TEST(TorchScriptTest, TestDictArgMatching) {
       def dict_op(a: Dict[str, Tensor], b: str):
         return a[b]
     )JIT");
-  c10::impl::GenericDict dict;
+  c10::Dict<std::string, at::Tensor> dict;
   dict.insert("hello", torch::ones({2}));
   auto output = module->run_method("dict_op", dict, std::string("hello"));
   ASSERT_EQ(1, output.toTensor()[0].item<int64_t>());

diff --git a/torch/csrc/jit/pybind_utils.h b/torch/csrc/jit/pybind_utils.h
@@ -305,7 +305,7 @@ inline TypedStack toTypedStack(const py::tuple& inputs) {
 }
 
 inline IValue createGenericList(py::handle obj, const TypePtr& elem_type) {
-  c10::List<IValue> elems;
+  c10::impl::GenericList elems;
   for (auto elem : obj) {
     elems.push_back(toIValue(elem, elem_type));
   }

diff --git a/torch/csrc/jit/register_prim_ops.cpp b/torch/csrc/jit/register_prim_ops.cpp
@@ -966,7 +966,7 @@ RegisterOperators reg(
            } else {
              return [=](Stack& stack) {
                const size_t stack_size = stack.size();
-               c10::List<IValue> vals;
+               c10::impl::GenericList vals;
                vals.reserve(num_inputs);
                for (size_t i = stack_size - num_inputs; i < stack_size; ++i) {
                  vals.emplace_back(std::move(stack[i]));