Add ScalarType argument to Type::options() (#19270)

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/19270
ghimport-source-id: a5ade6131f3260066c5750ea1fa9ed5c998bb791

Differential Revision: D14938707

Pulled By: li-roy

fbshipit-source-id: 018fb3f01706531a06515d6d861e5683a455a705

18 files changed, 76 insertions, 73 deletions

diff --git a/aten/src/ATen/Context.h b/aten/src/ATen/Context.h
index 925b327304..b77728e322 100644
--- a/aten/src/ATen/Context.h
+++ b/aten/src/ATen/Context.h
@@ -176,6 +176,11 @@ CAFFE2_API TypeExtendedInterface& getType(const Tensor&);
 
 CAFFE2_API Allocator* getCPUAllocator();
 
+static inline DeprecatedTypeProperties& getNonVariableDeprecatedTypeProperties(Backend p, ScalarType s) {
+  return globalDeprecatedTypePropertiesRegistry().getDeprecatedTypeProperties(
+      p, s, /*is_variable*/false);
+}
+
 static inline DeprecatedTypeProperties& CPU(ScalarType s) {
   return globalDeprecatedTypePropertiesRegistry().getDeprecatedTypeProperties(
       Backend::CPU, s, /*is_variable*/false);
diff --git a/aten/src/ATen/core/Type.h b/aten/src/ATen/core/Type.h
index 2d1c627817..ad7eba20ee 100644
--- a/aten/src/ATen/core/Type.h
+++ b/aten/src/ATen/core/Type.h
@@ -176,9 +176,8 @@ struct CAFFE2_API Type {
     return this != &other;
   }
 
-  /// Constructs the `TensorOptions` from a type and a `device_index`.
-  TensorOptions options(int16_t device_index = -1) const {
-    return TensorOptions().dtype(typeMeta())
+  TensorOptions options(ScalarType s, int16_t device_index = -1) const {
+    return TensorOptions().dtype(s)
                           .device(device_type(), device_index)
                           .layout(layout())
                           .is_variable(is_variable());
@@ -186,20 +185,16 @@ struct CAFFE2_API Type {
 
   /// Constructs the `TensorOptions` from a type and a Device.  Asserts that
   /// the device type matches the device type of the type.
-  TensorOptions options(c10::optional<Device> device_opt) const {
+  TensorOptions options(ScalarType s, c10::optional<Device> device_opt) const {
     if (!device_opt.has_value()) {
-      return options(-1);
+      return options(s, -1);
     } else {
       Device device = device_opt.value();
       AT_ASSERT(device.type() == device_type());
-      return options(device.index());
+      return options(s, device.index());
     }
   }
 
-  operator TensorOptions() const {
-    return options();
-  }
-
   // example
   // virtual Tensor * add(Tensor & a, Tensor & b) = 0;
   virtual Tensor abs(const Tensor & self) const = 0;
diff --git a/aten/src/ATen/function_wrapper.py b/aten/src/ATen/function_wrapper.py
index 8143be29e1..81d40813da 100644
--- a/aten/src/ATen/function_wrapper.py
+++ b/aten/src/ATen/function_wrapper.py
@@ -1604,7 +1604,8 @@ def create_derived(backend_type_env, declarations):
                     # e.g. x.sum(0) and x.sum() return the same type. We explicitly cast to the
                     # ScalarType before constructing the scalar_tensor to avoid overflow checking.
                     elif ret['type'] == 'accreal' or ret['type'] == 'real':
-                        return_scalar = 'return at::scalar_tensor(convert<${ScalarType}>(${call}), options());'
+                        return_scalar = ('return at::scalar_tensor(convert<${ScalarType}>(${call}), '
+                                         'options(ScalarType::${ScalarName}));')
                         case_body.append(CodeTemplate(return_scalar).substitute(case_env, call=call))
                     else:
                         # we using int64_t for long in the API, so correct it here...
diff --git a/aten/src/ATen/templates/Type.h b/aten/src/ATen/templates/Type.h
index 14900acc7c..0e1051ed7e 100644
--- a/aten/src/ATen/templates/Type.h
+++ b/aten/src/ATen/templates/Type.h
@@ -119,9 +119,8 @@ struct CAFFE2_API Type {
     return this != &other;
   }
 
-  /// Constructs the `TensorOptions` from a type and a `device_index`.
-  TensorOptions options(int16_t device_index = -1) const {
-    return TensorOptions().dtype(typeMeta())
+  TensorOptions options(ScalarType s, int16_t device_index = -1) const {
+    return TensorOptions().dtype(s)
                           .device(device_type(), device_index)
                           .layout(layout())
                           .is_variable(is_variable());
@@ -129,20 +128,16 @@ struct CAFFE2_API Type {
 
   /// Constructs the `TensorOptions` from a type and a Device.  Asserts that
   /// the device type matches the device type of the type.
-  TensorOptions options(c10::optional<Device> device_opt) const {
+  TensorOptions options(ScalarType s, c10::optional<Device> device_opt) const {
     if (!device_opt.has_value()) {
-      return options(-1);
+      return options(s, -1);
     } else {
       Device device = device_opt.value();
       AT_ASSERT(device.type() == device_type());
-      return options(device.index());
+      return options(s, device.index());
     }
   }
 
-  operator TensorOptions() const {
-    return options();
-  }
-
   // example
   // virtual Tensor * add(Tensor & a, Tensor & b) = 0;
   ${pure_virtual_type_method_declarations}
diff --git a/test/cpp/api/tensor_options.cpp b/test/cpp/api/tensor_options.cpp
index cebb374203..6e345b306e 100644
--- a/test/cpp/api/tensor_options.cpp
+++ b/test/cpp/api/tensor_options.cpp
@@ -66,10 +66,10 @@ TEST(TensorOptionsTest, ConstructsWellFromCPUTypes) {
   options = TensorOptions(kInt);
   REQUIRE_OPTIONS(kCPU, -1, kInt, kStrided);
 
-  options = TensorOptions(getNonVariableType(Backend::SparseCPU, kFloat));
+  options = TensorOptions(getNonVariableDeprecatedTypeProperties(Backend::SparseCPU, kFloat));
   REQUIRE_OPTIONS(kCPU, -1, kFloat, kSparse);
 
-  options = TensorOptions(getNonVariableType(Backend::SparseCPU, kByte));
+  options = TensorOptions(getNonVariableDeprecatedTypeProperties(Backend::SparseCPU, kByte));
   REQUIRE_OPTIONS(kCPU, -1, kByte, kSparse);
 }
 
@@ -77,7 +77,7 @@ TEST(TensorOptionsTest, ConstructsWellFromCPUTensors) {
   auto options = empty(5, kDouble).options();
   REQUIRE_OPTIONS(kCPU, -1, kDouble, kStrided);
 
-  options = empty(5, getNonVariableType(Backend::SparseCPU, kByte)).options();
+  options = empty(5, getNonVariableDeprecatedTypeProperties(Backend::SparseCPU, kByte)).options();
   REQUIRE_OPTIONS(kCPU, -1, kByte, kSparse);
 }
 
diff --git a/test/cpp/api/tensor_options_cuda.cpp b/test/cpp/api/tensor_options_cuda.cpp
index a048ce5ac6..04229f58c2 100644
--- a/test/cpp/api/tensor_options_cuda.cpp
+++ b/test/cpp/api/tensor_options_cuda.cpp
@@ -42,17 +42,17 @@ TEST(TensorOptionsTest, ConstructsWellFromCUDATypes_CUDA) {
   options = CUDA(kInt).options();
   REQUIRE_OPTIONS(kCUDA, -1, kInt, kStrided);
 
-  options = getNonVariableType(Backend::SparseCUDA, kFloat).options();
+  options = getNonVariableDeprecatedTypeProperties(Backend::SparseCUDA, kFloat).options();
   REQUIRE_OPTIONS(kCUDA, -1, kFloat, kSparse);
 
-  options = getNonVariableType(Backend::SparseCUDA, kByte).options();
+  options = getNonVariableDeprecatedTypeProperties(Backend::SparseCUDA, kByte).options();
   REQUIRE_OPTIONS(kCUDA, -1, kByte, kSparse);
 
   options = CUDA(kFloat).options(/*device=*/5);
   REQUIRE_OPTIONS(kCUDA, 5, kFloat, kStrided);
 
   options =
-      getNonVariableType(Backend::SparseCUDA, kFloat).options(/*device=*/5);
+      getNonVariableDeprecatedTypeProperties(Backend::SparseCUDA, kFloat).options(/*device=*/5);
   REQUIRE_OPTIONS(kCUDA, 5, kFloat, kSparse);
 }
 
@@ -60,7 +60,7 @@ TEST(TensorOptionsTest, ConstructsWellFromCUDATensors_MultiCUDA) {
   auto options = empty(5, device(kCUDA).dtype(kDouble)).options();
   REQUIRE_OPTIONS(kCUDA, 0, kDouble, kStrided);
 
-  options = empty(5, getNonVariableType(Backend::SparseCUDA, kByte)).options();
+  options = empty(5, getNonVariableDeprecatedTypeProperties(Backend::SparseCUDA, kByte)).options();
   REQUIRE_OPTIONS(kCUDA, 0, kByte, kSparse);
 
   if (torch::cuda::device_count() > 1) {
diff --git a/tools/autograd/templates/Functions.h b/tools/autograd/templates/Functions.h
index 99b6ab3a2c..bd81dc13ed 100644
--- a/tools/autograd/templates/Functions.h
+++ b/tools/autograd/templates/Functions.h
@@ -35,13 +35,13 @@ struct TypeAndSize {
   /* implicit */
   TypeAndSize(const Tensor & t)
     : sizes(t.sizes().vec())
-    , type(&t.dispatch_type()) {}
+    , type(&t.type()) {}
 
   Tensor zeros() { return at::zeros(sizes, *type); }
 
 private:
   std::vector<int64_t> sizes;
-  Type* type;
+  at::DeprecatedTypeProperties* type;
 };
 
 ${autograd_function_declarations}
diff --git a/torch/csrc/autograd/engine.cpp b/torch/csrc/autograd/engine.cpp
index b4716ddb3a..db9b0c4f0b 100644
--- a/torch/csrc/autograd/engine.cpp
+++ b/torch/csrc/autograd/engine.cpp
@@ -334,7 +334,7 @@ static variable_list call_post_hooks(Function& fn, variable_list outputs, const
   return outputs;
 }
 
-static bool is_compatible_type(const at::Type& expected, const at::Type& actual) {
+static bool is_compatible_type(const at::DeprecatedTypeProperties& expected, const at::DeprecatedTypeProperties& actual) {
   // Types are compatible if they exactly match or if the gradient is a sparse
   // version of the expected type.
   return expected == actual || (actual.is_sparse() &&
@@ -372,7 +372,7 @@ static void validate_outputs(const edge_list& edges, variable_list& grads, const
       }
       grads[i] = at::sum_to(std::move(grads[i]), metadata.shape());
     }
-    if (!is_compatible_type(metadata.type(), grads[i].dispatch_type())) {
+    if (!is_compatible_type(metadata.type(), grads[i].type())) {
       std::stringstream ss;
       ss << "invalid gradient at index " << i << " - expected type ";
       ss << metadata.type() << " but got " << grads[i].type();
diff --git a/torch/csrc/autograd/function.h b/torch/csrc/autograd/function.h
index 380dfd9cc7..512cfb60cb 100644
--- a/torch/csrc/autograd/function.h
+++ b/torch/csrc/autograd/function.h
@@ -130,7 +130,7 @@ struct TORCH_API Function : std::enable_shared_from_this<Function> {
   /// Adds the type and shape metadata for a new input. Returns the index of
   /// of the new input.
   uint32_t add_input_metadata(
-    const at::Type& type
+    const at::DeprecatedTypeProperties& type
   , at::IntArrayRef shape
   , at::Device device) noexcept {
     uint32_t input_nr = input_metadata_.size();
diff --git a/torch/csrc/autograd/input_metadata.h b/torch/csrc/autograd/input_metadata.h
index df767de639..94655989d2 100644
--- a/torch/csrc/autograd/input_metadata.h
+++ b/torch/csrc/autograd/input_metadata.h
@@ -12,17 +12,17 @@ namespace torch { namespace autograd {
 struct InputMetadata {
   InputMetadata() = default;
 
-  InputMetadata(const at::Type& type, at::IntArrayRef shape, at::Device device)
+  InputMetadata(const at::DeprecatedTypeProperties& type, at::IntArrayRef shape, at::Device device)
   : type_{&type} , shape_{shape}, device_{device} { }
 
   InputMetadata(const at::Tensor& t)
-  : InputMetadata(t.dispatch_type(), t.sizes(), t.device()) { }
+  : InputMetadata(t.type(), t.sizes(), t.device()) { }
 
   bool is_valid() const {
     return type_ != nullptr;
   }
 
-  const at::Type& type() const {
+  const at::DeprecatedTypeProperties& type() const {
     AT_ASSERT(type_);
     return *type_;
   }
@@ -40,7 +40,7 @@ struct InputMetadata {
   }
 
 private:
-  const at::Type* type_ = nullptr;
+  const at::DeprecatedTypeProperties* type_ = nullptr;
   at::DimVector shape_;
   at::Device device_ = at::kCPU;
 };
diff --git a/torch/csrc/autograd/python_function.cpp b/torch/csrc/autograd/python_function.cpp
index 50fc16094e..cc4e23fd5a 100644
--- a/torch/csrc/autograd/python_function.cpp
+++ b/torch/csrc/autograd/python_function.cpp
@@ -46,6 +46,7 @@ namespace torch { namespace autograd {
 VariableInfo::VariableInfo(const Variable& var)
   : type(&var.dispatch_type())
   , device(var.device())
+  , scalar_type(var.scalar_type())
   , size(var.sizes().vec())
   , requires_grad(var.requires_grad()) {
 }
@@ -53,7 +54,7 @@ VariableInfo::VariableInfo(const Variable& var)
 Variable VariableInfo::zeros(at::OptionalDeviceGuard& device_guard) const {
   // NB: This will NOT work if we ever get mixed device gradients
   device_guard.reset_device(device);
-  return at::zeros(size, type->options());
+  return at::zeros(size, type->options(scalar_type));
 }
 
 auto PyFunction::legacy_apply(const variable_list& inputs) -> variable_list {
diff --git a/torch/csrc/autograd/python_function.h b/torch/csrc/autograd/python_function.h
index 52ead11ba6..a2d1c6c87e 100644
--- a/torch/csrc/autograd/python_function.h
+++ b/torch/csrc/autograd/python_function.h
@@ -25,6 +25,7 @@ struct VariableInfo {
 
   at::Type* type;
   at::Device device = at::kCPU;
+  at::ScalarType scalar_type = at::kFloat;
   std::vector<int64_t> size;
   bool requires_grad;
 };
diff --git a/torch/csrc/autograd/python_legacy_variable.cpp b/torch/csrc/autograd/python_legacy_variable.cpp
index 932e4b241c..b61e6fbdb1 100644
--- a/torch/csrc/autograd/python_legacy_variable.cpp
+++ b/torch/csrc/autograd/python_legacy_variable.cpp
@@ -46,7 +46,8 @@ static PyObject *THPVariable_pynew(PyTypeObject* type, PyObject *args, PyObject
   if (!data || data == Py_None) {
     // For legacy serialization code, create an empty tensor. This is also used
     // by nn.Parameter() with no arguments.
-    auto var = at::empty({0}, torch::tensors::get_default_tensor_type().options());
+    auto scalar_type = torch::tensors::get_default_scalar_type();
+    auto var = at::empty({0}, torch::tensors::get_default_tensor_type().options(scalar_type));
     tensor = static_cast<Variable&>(var).data();
   } else if (THPVariable_Check(data)) {
     tensor = ((THPVariable*)data)->cdata.data();
diff --git a/torch/csrc/autograd/python_variable_indexing.cpp b/torch/csrc/autograd/python_variable_indexing.cpp
index eff82a92eb..1415a48544 100644
--- a/torch/csrc/autograd/python_variable_indexing.cpp
+++ b/torch/csrc/autograd/python_variable_indexing.cpp
@@ -110,15 +110,15 @@ static Variable sequenceToVariable(const at::Type& type, PyObject* seq) {
   return torch::utils::indexing_tensor_from_data(idx_type, kLong, c10::nullopt, seq);
 }
 
-static Variable valueToTensor(const at::Type & type, PyObject* value) {
+static Variable valueToTensor(const at::Type & type, const ScalarType scalar_type, PyObject* value) {
   if (THPVariable_Check(value)) {
     return reinterpret_cast<THPVariable*>(value)->cdata;
   }
   if (THPUtils_checkLong(value) || PyBool_Check(value)) {
-    return at::scalar_tensor(Scalar(THPUtils_unpackLong(value)), type.options());
+    return at::scalar_tensor(Scalar(THPUtils_unpackLong(value)), type.options(scalar_type));
   }
   if (PyFloat_Check(value)) {
-    return at::scalar_tensor(Scalar(THPUtils_unpackDouble(value)), type.options());
+    return at::scalar_tensor(Scalar(THPUtils_unpackDouble(value)), type.options(scalar_type));
   }
   throw TypeError("can't assign a %s to a %s", Py_TYPE(value)->tp_name, type.toString());
 }
@@ -334,7 +334,7 @@ int THPVariable_setitem(PyObject* self, PyObject* index, PyObject* py_value) {
 
   auto& self_ = reinterpret_cast<THPVariable*>(self)->cdata;
   OptionalDeviceGuard device_guard(device_of(self_));
-  auto value = valueToTensor(self_.dispatch_type(), py_value);
+  auto value = valueToTensor(self_.dispatch_type(), self_.scalar_type(), py_value);
 
   // handle simple types: integers, slices, ellipsis, bool
   if (index == Py_False) { // NOLINT(cppcoreguidelines-pro-type-cstyle-cast)
diff --git a/torch/csrc/autograd/variable.cpp b/torch/csrc/autograd/variable.cpp
index a330230f01..9fa2033c15 100644
--- a/torch/csrc/autograd/variable.cpp
+++ b/torch/csrc/autograd/variable.cpp
@@ -214,7 +214,7 @@ const std::shared_ptr<Function>& Variable::grad_fn() const {
       fn->storage_offset = data().storage_offset();
       fn->set_next_edges(collect_next_edges(diff_view_meta->base_));
       fn->add_input_metadata(
-        diff_view_meta->base_.dispatch_type()
+        diff_view_meta->base_.type()
       , sizes() // Note: sizes(), not base_.sizes(), is intentional
       , diff_view_meta->base_.device());
       diff_view_meta->grad_fn_ = std::move(fn);
diff --git a/torch/csrc/cuda/comm.cpp b/torch/csrc/cuda/comm.cpp
index 53faa6baa5..a1743355bf 100644
--- a/torch/csrc/cuda/comm.cpp
+++ b/torch/csrc/cuda/comm.cpp
@@ -59,7 +59,7 @@ std::vector<Tensor> broadcast(const Tensor& tensor, IntArrayRef devices) {
     tensors.push_back(tensor);
     for (auto device : devices.slice(1)) {
       _device_guard.set_index(device);
-      tensors.push_back(at::empty(tensor.sizes(), type.options()));
+      tensors.push_back(at::empty(tensor.sizes(), type.options(tensor.scalar_type())));
     }
     nccl::broadcast(tensors);
   } else {
diff --git a/torch/csrc/jit/passes/shape_analysis.cpp b/torch/csrc/jit/passes/shape_analysis.cpp
index c221dd9f01..39b87b6a25 100644
--- a/torch/csrc/jit/passes/shape_analysis.cpp
+++ b/torch/csrc/jit/passes/shape_analysis.cpp
@@ -157,10 +157,9 @@ class ShapePropagator {
       return *iv;
     }
     if (CompleteTensorTypePtr type = type_->cast<CompleteTensorType>()) {
-      auto backend =
-          type->device().is_cpu() ? at::Backend::CPU : at::Backend::CUDA;
+      auto attype = type->device().is_cpu() ?
+          at::CPU(type->scalarType()) : at::CUDA(type->scalarType());
       at::DeviceGuard device_guard(type->device());
-      auto& attype = at::getNonVariableType(backend, type->scalarType());
       auto t =
           at::empty_strided(type->sizes(), type->strides(), attype.options())
               .zero_();
diff --git a/torch/csrc/utils/tensor_new.cpp b/torch/csrc/utils/tensor_new.cpp
index 908a27fd4a..6c23d0a13c 100644
--- a/torch/csrc/utils/tensor_new.cpp
+++ b/torch/csrc/utils/tensor_new.cpp
@@ -53,32 +53,32 @@ void maybe_initialize_cuda(const Device device) {
   }
 }
 
-Tensor dispatch_zeros(const Type& type, optional<Device> device, IntArrayRef sizes) {
+Tensor dispatch_zeros(const Type& type, const ScalarType scalar_type, optional<Device> device, IntArrayRef sizes) {
   maybe_initialize_cuda(type);
   AutoNoGIL no_gil;
-  return torch::zeros(sizes, type.options(std::move(device)));
+  return torch::zeros(sizes, type.options(scalar_type, std::move(device)));
 }
 
-Tensor dispatch_ones(const Type& type, optional<Device> device, IntArrayRef sizes) {
+Tensor dispatch_ones(const Type& type, const ScalarType scalar_type, optional<Device> device, IntArrayRef sizes) {
   maybe_initialize_cuda(type);
   AutoNoGIL no_gil;
-  return torch::ones(sizes, type.options(std::move(device)));
+  return torch::ones(sizes, type.options(scalar_type, std::move(device)));
 }
 
-Tensor dispatch_full(const Type& type, Scalar fill_value, optional<Device> device, IntArrayRef sizes) {
+Tensor dispatch_full(const Type& type, const ScalarType scalar_type, Scalar fill_value, optional<Device> device, IntArrayRef sizes) {
   maybe_initialize_cuda(type);
   AutoNoGIL no_gil;
-  return torch::full(sizes, fill_value, type.options(std::move(device)));
+  return torch::full(sizes, fill_value, type.options(scalar_type, std::move(device)));
 }
 
-Tensor new_with_sizes(const Type& type, optional<Device> device, IntArrayRef sizes) {
+Tensor new_with_sizes(const Type& type, const ScalarType scalar_type, optional<Device> device, IntArrayRef sizes) {
   maybe_initialize_cuda(type);
   AutoNoGIL no_gil;
-  return torch::empty(sizes, type.options(std::move(device)));
+  return torch::empty(sizes, type.options(scalar_type, std::move(device)));
 }
 
-Tensor new_with_storage(const Type& type, Storage storage) {
-  auto tensor = at::empty({}, type.options());
+Tensor new_with_storage(const Type& type, const ScalarType scalar_type, Storage storage) {
+  auto tensor = at::empty({}, type.options(scalar_type));
   tensor.set_(std::move(storage));
   return tensor;
 }
@@ -281,7 +281,7 @@ Tensor legacy_sparse_tensor_ctor(const Type& type, ScalarType scalar_type, PyObj
   if (r.idx == 0) {
     auto deviceOptional = r.deviceOptional(0);
     check_legacy_ctor_device(type, deviceOptional);
-    return at::empty({0}, type.options(r.deviceOptional(0)));
+    return at::empty({0}, type.options(scalar_type, r.deviceOptional(0)));
   } else if (r.idx == 1) {
     auto cdata = reinterpret_cast<void*>(r.toInt64(0));
     return type.unsafeTensorFromTH(cdata, true);
@@ -304,7 +304,7 @@ Tensor legacy_sparse_tensor_ctor(const Type& type, ScalarType scalar_type, PyObj
       // unless the sequences is a torch.Size
       return legacy_new_from_sequence(type, scalar_type, deviceOptional, r.pyobject(0));
     }
-    return new_with_sizes(type, r.deviceOptional(1), r.intlist(0));
+    return new_with_sizes(type, scalar_type, r.deviceOptional(1), r.intlist(0));
   }
   throw std::runtime_error("new(): invalid arguments");
 }
@@ -323,7 +323,7 @@ Tensor legacy_sparse_tensor_new(const Type& type, ScalarType scalar_type, PyObje
     auto deviceOptional = r.deviceOptional(0);
     check_legacy_ctor_device(type, deviceOptional);
     at::OptionalDeviceGuard device_guard(deviceOptional);
-    return at::empty({0}, type.options());
+    return at::empty({0}, type.options(scalar_type));
   } else if (r.idx == 1) {
     auto cdata = reinterpret_cast<void*>(r.toInt64(0));
     return type.unsafeTensorFromTH(cdata, true);
@@ -350,7 +350,7 @@ Tensor legacy_sparse_tensor_new(const Type& type, ScalarType scalar_type, PyObje
       // unless the sequences is a torch.Size
       return legacy_new_from_sequence(type, scalar_type, deviceOptional, r.pyobject(0));
     }
-    return new_with_sizes(type, r.deviceOptional(1), r.intlist(0));
+    return new_with_sizes(type, scalar_type, r.deviceOptional(1), r.intlist(0));
   }
   throw std::runtime_error("new(): invalid arguments");
 }
@@ -384,9 +384,9 @@ Tensor legacy_tensor_ctor(const Type& type, ScalarType scalar_type, PyObject* ar
     auto deviceOptional = r.deviceOptional(0);
     check_legacy_ctor_device(type, deviceOptional);
     at::OptionalDeviceGuard device_guard(deviceOptional);
-    return at::empty({0}, type.options());
+    return at::empty({0}, type.options(scalar_type));
   } else if (r.idx == 1) {
-    return new_with_storage(type, r.storage(0));
+    return new_with_storage(type, scalar_type, r.storage(0));
   } else if (r.idx == 2) {
     auto cdata = reinterpret_cast<void*>(r.toInt64(0));
     return type.unsafeTensorFromTH(cdata, true);
@@ -401,7 +401,7 @@ Tensor legacy_tensor_ctor(const Type& type, ScalarType scalar_type, PyObject* ar
       // unless the sequences is a torch.Size
       return legacy_new_from_sequence(type, scalar_type, deviceOptional, r.pyobject(0));
     }
-    return new_with_sizes(type, r.deviceOptional(1), r.intlist(0));
+    return new_with_sizes(type, scalar_type, r.deviceOptional(1), r.intlist(0));
   } else if (r.idx == 5) {
     auto deviceOptional = r.deviceOptional(1);
     check_legacy_ctor_device(type, deviceOptional);
@@ -430,9 +430,9 @@ Tensor legacy_tensor_new(const Type& type, ScalarType scalar_type, PyObject* arg
     auto deviceOptional = r.deviceOptional(0);
     check_legacy_ctor_device(type, deviceOptional);
     at::OptionalDeviceGuard device_guard(deviceOptional);
-    return at::empty({0}, type.options());
+    return at::empty({0}, type.options(scalar_type));
   } else if (r.idx == 1) {
-    return new_with_storage(type, r.storage(0));
+    return new_with_storage(type, scalar_type, r.storage(0));
   } else if (r.idx == 2) {
     auto cdata = reinterpret_cast<void*>(r.toInt64(0));
     return type.unsafeTensorFromTH(cdata, true);
@@ -447,7 +447,7 @@ Tensor legacy_tensor_new(const Type& type, ScalarType scalar_type, PyObject* arg
       // unless the sequences is a torch.Size
       return legacy_new_from_sequence(type, scalar_type, deviceOptional, r.pyobject(0));
     }
-    return new_with_sizes(type, r.deviceOptional(1), r.intlist(0));
+    return new_with_sizes(type, scalar_type, r.deviceOptional(1), r.intlist(0));
   } else if (r.idx == 5) {
     auto deviceOptional = r.deviceOptional(1);
     check_legacy_ctor_device(type, deviceOptional);
@@ -504,8 +504,9 @@ Tensor sparse_coo_tensor_ctor(const Type& default_type, ScalarType scalar_type,
     return at::sparse_coo_tensor(indices, values, r.intlist(2), values.options().layout(at::kSparse)).set_requires_grad(r.toBool(5));
   } else if (r.idx == 2) {
     const auto& type = typeWithDefault(r, 1, 2, default_type, scalar_type);
+    const auto actual_scalar_type = r.scalartypeWithDefault(1, scalar_type);
     at::OptionalDeviceGuard device_guard(r.deviceOptional(2));
-    return at::sparse_coo_tensor(r.intlist(0), type.options().layout(at::kSparse)).set_requires_grad(r.toBool(3));
+    return at::sparse_coo_tensor(r.intlist(0), type.options(actual_scalar_type).layout(at::kSparse)).set_requires_grad(r.toBool(3));
   }
   throw std::runtime_error("sparse_coo_tensor(): invalid arguments");
 }
@@ -603,7 +604,8 @@ Tensor new_empty(const Type& type, ScalarType scalar_type, PyObject* args, PyObj
   auto r = parser.parse(args, kwargs, parsed_args);
   if (r.idx == 0) {
     const auto& actual_type = typeWithDefault(r, 1, 2, type, scalar_type);
-    return new_with_sizes(actual_type, r.deviceOptional(2), r.intlist(0)).set_requires_grad(r.toBool(4));
+    const auto actual_scalar_type = r.scalartypeWithDefault(1, scalar_type);
+    return new_with_sizes(actual_type, actual_scalar_type, r.deviceOptional(2), r.intlist(0)).set_requires_grad(r.toBool(4));
   }
   throw std::runtime_error("new_empty(): invalid arguments");
 }
@@ -617,7 +619,8 @@ Tensor new_full(const Type& type, ScalarType scalar_type, PyObject* args, PyObje
   auto r = parser.parse(args, kwargs, parsed_args);
   if (r.idx == 0) {
     const auto& actual_type = typeWithDefault(r, 2, 3, type, scalar_type);
-    return dispatch_full(actual_type, r.scalar(1), r.deviceOptional(3), r.intlist(0)).set_requires_grad(r.toBool(4));
+    const auto actual_scalar_type = r.scalartypeWithDefault(2, scalar_type);
+    return dispatch_full(actual_type, actual_scalar_type, r.scalar(1), r.deviceOptional(3), r.intlist(0)).set_requires_grad(r.toBool(4));
   }
   throw std::runtime_error("new_full(): invalid arguments");
 }
@@ -631,7 +634,8 @@ Tensor new_ones(const Type& type, ScalarType scalar_type, PyObject* args, PyObje
   auto r = parser.parse(args, kwargs, parsed_args);
   if (r.idx == 0) {
     const auto& actual_type = typeWithDefault(r, 1, 2, type, scalar_type);
-    return dispatch_ones(actual_type, r.deviceOptional(2), r.intlist(0)).set_requires_grad(r.toBool(3));
+    const auto actual_scalar_type = r.scalartypeWithDefault(1, scalar_type);
+    return dispatch_ones(actual_type, actual_scalar_type, r.deviceOptional(2), r.intlist(0)).set_requires_grad(r.toBool(3));
   }
   throw std::runtime_error("new_ones(): invalid arguments");
 }
@@ -645,7 +649,8 @@ Tensor new_zeros(const Type& type, ScalarType scalar_type, PyObject* args, PyObj
   auto r = parser.parse(args, kwargs, parsed_args);
   if (r.idx == 0) {
     const auto& actual_type = typeWithDefault(r, 1, 2, type, scalar_type);
-    return dispatch_zeros(actual_type, r.deviceOptional(2), r.intlist(0)).set_requires_grad(r.toBool(3));
+    const auto actual_scalar_type = r.scalartypeWithDefault(1, scalar_type);
+    return dispatch_zeros(actual_type, actual_scalar_type, r.deviceOptional(2), r.intlist(0)).set_requires_grad(r.toBool(3));
   }
   throw std::runtime_error("new_zeros(): invalid arguments");
 }