Add files using upload-large-folder tool

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cast_Float.h

@@ -0,0 +1,30 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/_cast_Float_ops.h>
+
+namespace at {
+
+
+// aten::_cast_Float(Tensor self, bool non_blocking=False) -> Tensor
+inline at::Tensor _cast_Float(const at::Tensor & self, bool non_blocking=false) {
+    return at::_ops::_cast_Float::call(self, non_blocking);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_coalesce.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/_coalesce_ops.h>
+
+namespace at {
+
+
+// aten::_coalesce(Tensor self) -> Tensor
+inline at::Tensor _coalesce(const at::Tensor & self) {
+    return at::_ops::_coalesce::call(self);
+}
+
+// aten::_coalesce.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _coalesce_out(at::Tensor & out, const at::Tensor & self) {
+    return at::_ops::_coalesce_out::call(self, out);
+}
+// aten::_coalesce.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _coalesce_outf(const at::Tensor & self, at::Tensor & out) {
+    return at::_ops::_coalesce_out::call(self, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cudnn_rnn_backward_ops.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API _cudnn_rnn_backward {
+  using schema = ::std::tuple<at::Tensor,at::Tensor,at::Tensor,::std::vector<at::Tensor>> (const at::Tensor &, at::TensorList, int64_t, const at::Tensor &, const at::Tensor &, const c10::optional<at::Tensor> &, const at::Tensor &, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, int64_t, c10::SymInt, c10::SymInt, int64_t, bool, double, bool, bool, c10::SymIntArrayRef, const c10::optional<at::Tensor> &, const at::Tensor &, ::std::array<bool,4>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_cudnn_rnn_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_cudnn_rnn_backward(Tensor input, Tensor[] weight, int weight_stride0, Tensor weight_buf, Tensor hx, Tensor? cx, Tensor output, Tensor? grad_output, Tensor? grad_hy, Tensor? grad_cy, int mode, SymInt hidden_size, SymInt proj_size, int num_layers, bool batch_first, float dropout, bool train, bool bidirectional, SymInt[] batch_sizes, Tensor? dropout_state, Tensor reserve, bool[4] output_mask) -> (Tensor, Tensor, Tensor, Tensor[])")
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor,::std::vector<at::Tensor>> call(const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const c10::optional<at::Tensor> & cx, const at::Tensor & output, const c10::optional<at::Tensor> & grad_output, const c10::optional<at::Tensor> & grad_hy, const c10::optional<at::Tensor> & grad_cy, int64_t mode, c10::SymInt hidden_size, c10::SymInt proj_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, c10::SymIntArrayRef batch_sizes, const c10::optional<at::Tensor> & dropout_state, const at::Tensor & reserve, ::std::array<bool,4> output_mask);
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor,::std::vector<at::Tensor>> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const c10::optional<at::Tensor> & cx, const at::Tensor & output, const c10::optional<at::Tensor> & grad_output, const c10::optional<at::Tensor> & grad_hy, const c10::optional<at::Tensor> & grad_cy, int64_t mode, c10::SymInt hidden_size, c10::SymInt proj_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, c10::SymIntArrayRef batch_sizes, const c10::optional<at::Tensor> & dropout_state, const at::Tensor & reserve, ::std::array<bool,4> output_mask);
+};
+
+struct TORCH_API _cudnn_rnn_backward_out {
+  using schema = void (const at::Tensor &, at::TensorList, int64_t, const at::Tensor &, const at::Tensor &, const c10::optional<at::Tensor> &, const at::Tensor &, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, int64_t, c10::SymInt, c10::SymInt, int64_t, bool, double, bool, bool, c10::SymIntArrayRef, const c10::optional<at::Tensor> &, const at::Tensor &, ::std::array<bool,4>, at::Tensor &, at::Tensor &, at::Tensor &, at::TensorList);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_cudnn_rnn_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_cudnn_rnn_backward.out(Tensor input, Tensor[] weight, int weight_stride0, Tensor weight_buf, Tensor hx, Tensor? cx, Tensor output, Tensor? grad_output, Tensor? grad_hy, Tensor? grad_cy, int mode, SymInt hidden_size, SymInt proj_size, int num_layers, bool batch_first, float dropout, bool train, bool bidirectional, SymInt[] batch_sizes, Tensor? dropout_state, Tensor reserve, bool[4] output_mask, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2, Tensor(d!)[] out3) -> ()")
+  static void call(const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const c10::optional<at::Tensor> & cx, const at::Tensor & output, const c10::optional<at::Tensor> & grad_output, const c10::optional<at::Tensor> & grad_hy, const c10::optional<at::Tensor> & grad_cy, int64_t mode, c10::SymInt hidden_size, c10::SymInt proj_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, c10::SymIntArrayRef batch_sizes, const c10::optional<at::Tensor> & dropout_state, const at::Tensor & reserve, ::std::array<bool,4> output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::TensorList out3);
+  static void redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const c10::optional<at::Tensor> & cx, const at::Tensor & output, const c10::optional<at::Tensor> & grad_output, const c10::optional<at::Tensor> & grad_hy, const c10::optional<at::Tensor> & grad_cy, int64_t mode, c10::SymInt hidden_size, c10::SymInt proj_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, c10::SymIntArrayRef batch_sizes, const c10::optional<at::Tensor> & dropout_state, const at::Tensor & reserve, ::std::array<bool,4> output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::TensorList out3);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fill_mem_eff_dropout_mask_native.h

@@ -0,0 +1,21 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor & _fill_mem_eff_dropout_mask_(at::Tensor & self, double dropout_p, int64_t seed, int64_t offset);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_addcmul.h

@@ -0,0 +1,82 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/_foreach_addcmul_ops.h>
+
+namespace at {
+
+
+// aten::_foreach_addcmul.Scalar(Tensor[] self, Tensor[] tensor1, Tensor[] tensor2, Scalar value=1) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_addcmul(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Scalar & value=1) {
+    return at::_ops::_foreach_addcmul_Scalar::call(self, tensor1, tensor2, value);
+}
+
+// aten::_foreach_addcmul.ScalarList(Tensor[] self, Tensor[] tensor1, Tensor[] tensor2, Scalar[] scalars) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_addcmul(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, at::ArrayRef<at::Scalar> scalars) {
+    return at::_ops::_foreach_addcmul_ScalarList::call(self, tensor1, tensor2, scalars);
+}
+
+// aten::_foreach_addcmul.Tensor(Tensor[] self, Tensor[] tensor1, Tensor[] tensor2, Tensor scalars) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_addcmul(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Tensor & scalars) {
+    return at::_ops::_foreach_addcmul_Tensor::call(self, tensor1, tensor2, scalars);
+}
+
+// aten::_foreach_addcmul_.Scalar(Tensor(a!)[] self, Tensor[] tensor1, Tensor[] tensor2, Scalar value=1) -> ()
+inline void _foreach_addcmul_(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Scalar & value=1) {
+    return at::_ops::_foreach_addcmul__Scalar::call(self, tensor1, tensor2, value);
+}
+
+// aten::_foreach_addcmul_.ScalarList(Tensor(a!)[] self, Tensor[] tensor1, Tensor[] tensor2, Scalar[] scalars) -> ()
+inline void _foreach_addcmul_(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, at::ArrayRef<at::Scalar> scalars) {
+    return at::_ops::_foreach_addcmul__ScalarList::call(self, tensor1, tensor2, scalars);
+}
+
+// aten::_foreach_addcmul_.Tensor(Tensor(a!)[] self, Tensor[] tensor1, Tensor[] tensor2, Tensor scalars) -> ()
+inline void _foreach_addcmul_(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Tensor & scalars) {
+    return at::_ops::_foreach_addcmul__Tensor::call(self, tensor1, tensor2, scalars);
+}
+
+// aten::_foreach_addcmul.Scalar_out(Tensor[] self, Tensor[] tensor1, Tensor[] tensor2, Scalar value=1, *, Tensor(a!)[] out) -> ()
+inline void _foreach_addcmul_out(at::TensorList out, at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Scalar & value=1) {
+    return at::_ops::_foreach_addcmul_Scalar_out::call(self, tensor1, tensor2, value, out);
+}
+// aten::_foreach_addcmul.Scalar_out(Tensor[] self, Tensor[] tensor1, Tensor[] tensor2, Scalar value=1, *, Tensor(a!)[] out) -> ()
+inline void _foreach_addcmul_outf(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Scalar & value, at::TensorList out) {
+    return at::_ops::_foreach_addcmul_Scalar_out::call(self, tensor1, tensor2, value, out);
+}
+
+// aten::_foreach_addcmul.ScalarList_out(Tensor[] self, Tensor[] tensor1, Tensor[] tensor2, Scalar[] scalars, *, Tensor(a!)[] out) -> ()
+inline void _foreach_addcmul_out(at::TensorList out, at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, at::ArrayRef<at::Scalar> scalars) {
+    return at::_ops::_foreach_addcmul_ScalarList_out::call(self, tensor1, tensor2, scalars, out);
+}
+// aten::_foreach_addcmul.ScalarList_out(Tensor[] self, Tensor[] tensor1, Tensor[] tensor2, Scalar[] scalars, *, Tensor(a!)[] out) -> ()
+inline void _foreach_addcmul_outf(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, at::ArrayRef<at::Scalar> scalars, at::TensorList out) {
+    return at::_ops::_foreach_addcmul_ScalarList_out::call(self, tensor1, tensor2, scalars, out);
+}
+
+// aten::_foreach_addcmul.Tensor_out(Tensor[] self, Tensor[] tensor1, Tensor[] tensor2, Tensor scalars, *, Tensor(a!)[] out) -> ()
+inline void _foreach_addcmul_out(at::TensorList out, at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Tensor & scalars) {
+    return at::_ops::_foreach_addcmul_Tensor_out::call(self, tensor1, tensor2, scalars, out);
+}
+// aten::_foreach_addcmul.Tensor_out(Tensor[] self, Tensor[] tensor1, Tensor[] tensor2, Tensor scalars, *, Tensor(a!)[] out) -> ()
+inline void _foreach_addcmul_outf(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Tensor & scalars, at::TensorList out) {
+    return at::_ops::_foreach_addcmul_Tensor_out::call(self, tensor1, tensor2, scalars, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log_cpu_dispatch.h

@@ -0,0 +1,24 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API ::std::vector<at::Tensor> _foreach_log(at::TensorList self);
+TORCH_API void _foreach_log_(at::TensorList self);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_indices_copy_ops.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API _indices_copy {
+  using schema = at::Tensor (const at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_indices_copy")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_indices_copy(Tensor self) -> Tensor")
+  static at::Tensor call(const at::Tensor & self);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self);
+};
+
+struct TORCH_API _indices_copy_out {
+  using schema = at::Tensor & (const at::Tensor &, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_indices_copy")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_indices_copy.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_make_per_tensor_quantized_tensor_ops.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API _make_per_tensor_quantized_tensor {
+  using schema = at::Tensor (const at::Tensor &, double, int64_t);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_make_per_tensor_quantized_tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_make_per_tensor_quantized_tensor(Tensor self, float scale, int zero_point) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, double scale, int64_t zero_point);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, double scale, int64_t zero_point);
+};
+
+struct TORCH_API _make_per_tensor_quantized_tensor_out {
+  using schema = at::Tensor & (const at::Tensor &, double, int64_t, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_make_per_tensor_quantized_tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_make_per_tensor_quantized_tensor.out(Tensor self, float scale, int zero_point, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, double scale, int64_t zero_point, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, double scale, int64_t zero_point, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_select_backward.h

@@ -0,0 +1,47 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/_nested_select_backward_ops.h>
+
+namespace at {
+
+
+// aten::_nested_select_backward(Tensor grad_output, Tensor self, int dim, SymInt index) -> Tensor
+inline at::Tensor _nested_select_backward(const at::Tensor & grad_output, const at::Tensor & self, int64_t dim, int64_t index) {
+    return at::_ops::_nested_select_backward::call(grad_output, self, dim, index);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor _nested_select_backward(const at::Tensor & grad_output, const at::Tensor & self, int64_t dim, int64_t index) {
+    return at::_ops::_nested_select_backward::call(grad_output, self, dim, index);
+  }
+}
+
+// aten::_nested_select_backward(Tensor grad_output, Tensor self, int dim, SymInt index) -> Tensor
+inline at::Tensor _nested_select_backward_symint(const at::Tensor & grad_output, const at::Tensor & self, int64_t dim, c10::SymInt index) {
+    return at::_ops::_nested_select_backward::call(grad_output, self, dim, index);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor _nested_select_backward(const at::Tensor & grad_output, const at::Tensor & self, int64_t dim, c10::SymInt index) {
+    return at::_ops::_nested_select_backward::call(grad_output, self, dim, index);
+  }
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data_compositeexplicitautogradnonfunctional_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeexplicitautogradnonfunctional {
+
+TORCH_API at::Tensor _softmax_backward_data(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_stack_cpu_dispatch.h

@@ -0,0 +1,25 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API at::Tensor _stack(at::TensorList tensors, int64_t dim=0);
+TORCH_API at::Tensor & _stack_out(at::Tensor & out, at::TensorList tensors, int64_t dim=0);
+TORCH_API at::Tensor & _stack_outf(at::TensorList tensors, int64_t dim, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_fused_lstm_cell_backward_ops.h

@@ -0,0 +1,28 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API _thnn_fused_lstm_cell_backward {
+  using schema = ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor,at::Tensor> (const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, const at::Tensor &, const at::Tensor &, const at::Tensor &, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_thnn_fused_lstm_cell_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_thnn_fused_lstm_cell_backward(Tensor? grad_hy, Tensor? grad_cy, Tensor cx, Tensor cy, Tensor workspace, bool has_bias) -> (Tensor, Tensor, Tensor, Tensor, Tensor)")
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor,at::Tensor> call(const c10::optional<at::Tensor> & grad_hy, const c10::optional<at::Tensor> & grad_cy, const at::Tensor & cx, const at::Tensor & cy, const at::Tensor & workspace, bool has_bias);
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor,at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, const c10::optional<at::Tensor> & grad_hy, const c10::optional<at::Tensor> & grad_cy, const at::Tensor & cx, const at::Tensor & cy, const at::Tensor & workspace, bool has_bias);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_fused_lstm_cell_cuda_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor> _thnn_fused_lstm_cell(const at::Tensor & input_gates, const at::Tensor & hidden_gates, const at::Tensor & cx, const c10::optional<at::Tensor> & input_bias={}, const c10::optional<at::Tensor> & hidden_bias={});
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_use_cudnn_ctc_loss_ops.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API _use_cudnn_ctc_loss {
+  using schema = bool (const at::Tensor &, const at::Tensor &, at::IntArrayRef, at::IntArrayRef, int64_t);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_use_cudnn_ctc_loss")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_use_cudnn_ctc_loss(Tensor log_probs, Tensor targets, int[] input_lengths, int[] target_lengths, int blank) -> bool")
+  static bool call(const at::Tensor & log_probs, const at::Tensor & targets, at::IntArrayRef input_lengths, at::IntArrayRef target_lengths, int64_t blank);
+  static bool redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & log_probs, const at::Tensor & targets, at::IntArrayRef input_lengths, at::IntArrayRef target_lengths, int64_t blank);
+};
+
+struct TORCH_API _use_cudnn_ctc_loss_Tensor {
+  using schema = bool (const at::Tensor &, const at::Tensor &, const at::Tensor &, const at::Tensor &, int64_t);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_use_cudnn_ctc_loss")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_use_cudnn_ctc_loss.Tensor(Tensor log_probs, Tensor targets, Tensor input_lengths, Tensor target_lengths, int blank) -> bool")
+  static bool call(const at::Tensor & log_probs, const at::Tensor & targets, const at::Tensor & input_lengths, const at::Tensor & target_lengths, int64_t blank);
+  static bool redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & log_probs, const at::Tensor & targets, const at::Tensor & input_lengths, const at::Tensor & target_lengths, int64_t blank);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d.h

@@ -0,0 +1,91 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/adaptive_avg_pool2d_ops.h>
+
+namespace at {
+
+
+// aten::adaptive_avg_pool2d.out(Tensor self, SymInt[2] output_size, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & adaptive_avg_pool2d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size) {
+    return at::_ops::adaptive_avg_pool2d_out::call(self, c10::fromIntArrayRefSlow(output_size), out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & adaptive_avg_pool2d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size) {
+    return at::_ops::adaptive_avg_pool2d_out::call(self, c10::fromIntArrayRefSlow(output_size), out);
+  }
+}
+
+// aten::adaptive_avg_pool2d.out(Tensor self, SymInt[2] output_size, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & adaptive_avg_pool2d_outf(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out) {
+    return at::_ops::adaptive_avg_pool2d_out::call(self, c10::fromIntArrayRefSlow(output_size), out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & adaptive_avg_pool2d_outf(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out) {
+    return at::_ops::adaptive_avg_pool2d_out::call(self, c10::fromIntArrayRefSlow(output_size), out);
+  }
+}
+
+// aten::adaptive_avg_pool2d.out(Tensor self, SymInt[2] output_size, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & adaptive_avg_pool2d_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size) {
+    return at::_ops::adaptive_avg_pool2d_out::call(self, output_size, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & adaptive_avg_pool2d_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size) {
+    return at::_ops::adaptive_avg_pool2d_out::call(self, output_size, out);
+  }
+}
+
+// aten::adaptive_avg_pool2d.out(Tensor self, SymInt[2] output_size, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & adaptive_avg_pool2d_symint_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, at::Tensor & out) {
+    return at::_ops::adaptive_avg_pool2d_out::call(self, output_size, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & adaptive_avg_pool2d_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, at::Tensor & out) {
+    return at::_ops::adaptive_avg_pool2d_out::call(self, output_size, out);
+  }
+}
+
+// aten::adaptive_avg_pool2d(Tensor self, SymInt[2] output_size) -> Tensor
+inline at::Tensor adaptive_avg_pool2d(const at::Tensor & self, at::IntArrayRef output_size) {
+    return at::_ops::adaptive_avg_pool2d::call(self, c10::fromIntArrayRefSlow(output_size));
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor adaptive_avg_pool2d(const at::Tensor & self, at::IntArrayRef output_size) {
+    return at::_ops::adaptive_avg_pool2d::call(self, c10::fromIntArrayRefSlow(output_size));
+  }
+}
+
+// aten::adaptive_avg_pool2d(Tensor self, SymInt[2] output_size) -> Tensor
+inline at::Tensor adaptive_avg_pool2d_symint(const at::Tensor & self, c10::SymIntArrayRef output_size) {
+    return at::_ops::adaptive_avg_pool2d::call(self, output_size);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor adaptive_avg_pool2d(const at::Tensor & self, c10::SymIntArrayRef output_size) {
+    return at::_ops::adaptive_avg_pool2d::call(self, output_size);
+  }
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool3d_native.h

@@ -0,0 +1,24 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor adaptive_avg_pool3d_symint(const at::Tensor & self, c10::SymIntArrayRef output_size);
+TORCH_API at::Tensor & adaptive_avg_pool3d_out_cpu(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out);
+TORCH_API at::Tensor & adaptive_avg_pool3d_out_cuda(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out);
+TORCH_API at::Tensor & adaptive_avg_pool3d_out_quantized_cpu(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/allclose_compositeexplicitautograd_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeexplicitautograd {
+
+TORCH_API bool allclose(const at::Tensor & self, const at::Tensor & other, double rtol=1e-05, double atol=1e-08, bool equal_nan=false);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/and_ops.h

@@ -0,0 +1,61 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API __and___Scalar {
+  using schema = at::Tensor (const at::Tensor &, const at::Scalar &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::__and__")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "__and__.Scalar(Tensor self, Scalar other) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const at::Scalar & other);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & other);
+};
+
+struct TORCH_API __and___Tensor {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::__and__")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "__and__.Tensor(Tensor self, Tensor other) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const at::Tensor & other);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other);
+};
+
+struct TORCH_API __iand___Scalar {
+  using schema = at::Tensor & (at::Tensor &, const at::Scalar &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::__iand__")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "__iand__.Scalar(Tensor(a!) self, Scalar other) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, const at::Scalar & other);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, const at::Scalar & other);
+};
+
+struct TORCH_API __iand___Tensor {
+  using schema = at::Tensor & (at::Tensor &, const at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::__iand__")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "__iand__.Tensor(Tensor(a!) self, Tensor other) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, const at::Tensor & other);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, const at::Tensor & other);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/angle.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/angle_ops.h>
+
+namespace at {
+
+
+// aten::angle(Tensor self) -> Tensor
+inline at::Tensor angle(const at::Tensor & self) {
+    return at::_ops::angle::call(self);
+}
+
+// aten::angle.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & angle_out(at::Tensor & out, const at::Tensor & self) {
+    return at::_ops::angle_out::call(self, out);
+}
+// aten::angle.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & angle_outf(const at::Tensor & self, at::Tensor & out) {
+    return at::_ops::angle_out::call(self, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/as_strided_scatter_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor & as_strided_scatter_out_symint(const at::Tensor & self, const at::Tensor & src, c10::SymIntArrayRef size, c10::SymIntArrayRef stride, c10::optional<c10::SymInt> storage_offset, at::Tensor & out);
+TORCH_API at::Tensor as_strided_scatter_symint(const at::Tensor & self, const at::Tensor & src, c10::SymIntArrayRef size, c10::SymIntArrayRef stride, c10::optional<c10::SymInt> storage_offset=c10::nullopt);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bartlett_window_native.h

@@ -0,0 +1,24 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor bartlett_window(int64_t window_length, c10::optional<at::ScalarType> dtype={}, c10::optional<at::Layout> layout={}, c10::optional<at::Device> device={}, c10::optional<bool> pin_memory={});
+TORCH_API at::Tensor & bartlett_window_out(int64_t window_length, at::Tensor & out);
+TORCH_API at::Tensor bartlett_window(int64_t window_length, bool periodic, c10::optional<at::ScalarType> dtype={}, c10::optional<at::Layout> layout={}, c10::optional<at::Device> device={}, c10::optional<bool> pin_memory={});
+TORCH_API at::Tensor & bartlett_window_periodic_out(int64_t window_length, bool periodic, at::Tensor & out);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_right_shift_meta_dispatch.h

@@ -0,0 +1,26 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace meta {
+
+TORCH_API at::Tensor bitwise_right_shift(const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & bitwise_right_shift_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & bitwise_right_shift_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out);
+TORCH_API at::Tensor & bitwise_right_shift_(at::Tensor & self, const at::Tensor & other);
+
+} // namespace meta
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/concatenate_compositeimplicitautograd_dispatch.h

@@ -0,0 +1,28 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeimplicitautograd {
+
+TORCH_API at::Tensor concatenate(at::TensorList tensors, int64_t dim=0);
+TORCH_API at::Tensor & concatenate_out(at::Tensor & out, at::TensorList tensors, int64_t dim=0);
+TORCH_API at::Tensor & concatenate_outf(at::TensorList tensors, int64_t dim, at::Tensor & out);
+TORCH_API at::Tensor concatenate(at::TensorList tensors, at::Dimname dim);
+TORCH_API at::Tensor & concatenate_out(at::Tensor & out, at::TensorList tensors, at::Dimname dim);
+TORCH_API at::Tensor & concatenate_outf(at::TensorList tensors, at::Dimname dim, at::Tensor & out);
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/data_native.h

@@ -0,0 +1,21 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor data(const at::Tensor & self);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagonal_backward_ops.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API diagonal_backward {
+  using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef, int64_t, int64_t, int64_t);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::diagonal_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "diagonal_backward(Tensor grad_output, SymInt[] input_sizes, int offset, int dim1, int dim2) -> Tensor")
+  static at::Tensor call(const at::Tensor & grad_output, c10::SymIntArrayRef input_sizes, int64_t offset, int64_t dim1, int64_t dim2);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, c10::SymIntArrayRef input_sizes, int64_t offset, int64_t dim1, int64_t dim2);
+};
+
+struct TORCH_API diagonal_backward_out {
+  using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, int64_t, int64_t, int64_t, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::diagonal_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "diagonal_backward.out(Tensor grad_output, SymInt[] input_sizes, int offset, int dim1, int dim2, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & grad_output, c10::SymIntArrayRef input_sizes, int64_t offset, int64_t dim1, int64_t dim2, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, c10::SymIntArrayRef input_sizes, int64_t offset, int64_t dim1, int64_t dim2, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagonal_compositeimplicitautograd_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeimplicitautograd {
+
+TORCH_API at::Tensor diagonal(const at::Tensor & self, at::Dimname outdim, at::Dimname dim1, at::Dimname dim2, int64_t offset=0);
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/erfc_meta.h

@@ -0,0 +1,27 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeMetaFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/TensorIterator.h>
+#include <ATen/TensorMeta.h>
+#include <tuple>
+#include <vector>
+
+namespace at {
+namespace meta {
+
+struct TORCH_API structured_erfc : public TensorIteratorBase {
+    
+    
+    void meta(const at::Tensor & self);
+};
+
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/full_ops.h

@@ -0,0 +1,61 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API full_names {
+  using schema = at::Tensor (at::IntArrayRef, const at::Scalar &, c10::optional<at::DimnameList>, c10::optional<at::ScalarType>, c10::optional<at::Layout>, c10::optional<at::Device>, c10::optional<bool>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::full")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "names")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "full.names(int[] size, Scalar fill_value, *, Dimname[]? names, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor")
+  static at::Tensor call(at::IntArrayRef size, const at::Scalar & fill_value, c10::optional<at::DimnameList> names, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, at::IntArrayRef size, const at::Scalar & fill_value, c10::optional<at::DimnameList> names, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+};
+
+struct TORCH_API full {
+  using schema = at::Tensor (c10::SymIntArrayRef, const at::Scalar &, c10::optional<at::ScalarType>, c10::optional<at::Layout>, c10::optional<at::Device>, c10::optional<bool>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::full")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "full(SymInt[] size, Scalar fill_value, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor")
+  static at::Tensor call(c10::SymIntArrayRef size, const at::Scalar & fill_value, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymIntArrayRef size, const at::Scalar & fill_value, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+};
+
+struct TORCH_API full_out {
+  using schema = at::Tensor & (c10::SymIntArrayRef, const at::Scalar &, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::full")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "full.out(SymInt[] size, Scalar fill_value, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(c10::SymIntArrayRef size, const at::Scalar & fill_value, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymIntArrayRef size, const at::Scalar & fill_value, at::Tensor & out);
+};
+
+struct TORCH_API full_names_out {
+  using schema = at::Tensor & (at::IntArrayRef, const at::Scalar &, c10::optional<at::DimnameList>, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::full")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "names_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "full.names_out(int[] size, Scalar fill_value, *, Dimname[]? names, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(at::IntArrayRef size, const at::Scalar & fill_value, c10::optional<at::DimnameList> names, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::IntArrayRef size, const at::Scalar & fill_value, c10::optional<at::DimnameList> names, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler.h

@@ -0,0 +1,30 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/grid_sampler_ops.h>
+
+namespace at {
+
+
+// aten::grid_sampler(Tensor input, Tensor grid, int interpolation_mode, int padding_mode, bool align_corners) -> Tensor
+inline at::Tensor grid_sampler(const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners) {
+    return at::_ops::grid_sampler::call(input, grid, interpolation_mode, padding_mode, align_corners);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_2d_backward_native.h

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> grid_sampler_2d_backward_out(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, ::std::array<bool,2> output_mask, at::Tensor & out0, at::Tensor & out1);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> grid_sampler_2d_backward_cpu(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, ::std::array<bool,2> output_mask);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> grid_sampler_2d_backward_cuda(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, ::std::array<bool,2> output_mask);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_3d_backward_cuda_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> grid_sampler_3d_backward(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, ::std::array<bool,2> output_mask);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/histogram_native.h

@@ -0,0 +1,24 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> histogram(const at::Tensor & self, const at::Tensor & bins, const c10::optional<at::Tensor> & weight={}, bool density=false);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> histogram_out(const at::Tensor & self, const at::Tensor & bins, const c10::optional<at::Tensor> & weight, bool density, at::Tensor & hist, at::Tensor & bin_edges);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> histogram(const at::Tensor & self, int64_t bins=100, c10::optional<at::ArrayRef<double>> range=c10::nullopt, const c10::optional<at::Tensor> & weight={}, bool density=false);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> histogram_out(const at::Tensor & self, int64_t bins, c10::optional<at::ArrayRef<double>> range, const c10::optional<at::Tensor> & weight, bool density, at::Tensor & hist, at::Tensor & bin_edges);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/instance_norm_ops.h

@@ -0,0 +1,28 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API instance_norm {
+  using schema = at::Tensor (const at::Tensor &, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, bool, double, double, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::instance_norm")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "instance_norm(Tensor input, Tensor? weight, Tensor? bias, Tensor? running_mean, Tensor? running_var, bool use_input_stats, float momentum, float eps, bool cudnn_enabled) -> Tensor")
+  static at::Tensor call(const at::Tensor & input, const c10::optional<at::Tensor> & weight, const c10::optional<at::Tensor> & bias, const c10::optional<at::Tensor> & running_mean, const c10::optional<at::Tensor> & running_var, bool use_input_stats, double momentum, double eps, bool cudnn_enabled);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, const c10::optional<at::Tensor> & weight, const c10::optional<at::Tensor> & bias, const c10::optional<at::Tensor> & running_mean, const c10::optional<at::Tensor> & running_var, bool use_input_stats, double momentum, double eps, bool cudnn_enabled);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_nonzero_native.h

@@ -0,0 +1,21 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API bool is_nonzero(const at::Tensor & self);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_vulkan_available_ops.h

@@ -0,0 +1,28 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API is_vulkan_available {
+  using schema = bool ();
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::is_vulkan_available")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "is_vulkan_available() -> bool")
+  static bool call();
+  static bool redispatch(c10::DispatchKeySet dispatchKeySet);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/layer_norm_ops.h

@@ -0,0 +1,28 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API layer_norm {
+  using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, double, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::layer_norm")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "layer_norm(Tensor input, SymInt[] normalized_shape, Tensor? weight=None, Tensor? bias=None, float eps=1e-05, bool cudnn_enable=True) -> Tensor")
+  static at::Tensor call(const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const c10::optional<at::Tensor> & weight, const c10::optional<at::Tensor> & bias, double eps, bool cudnn_enable);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const c10::optional<at::Tensor> & weight, const c10::optional<at::Tensor> & bias, double eps, bool cudnn_enable);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lerp_cuda_dispatch.h

@@ -0,0 +1,30 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API at::Tensor lerp(const at::Tensor & self, const at::Tensor & end, const at::Scalar & weight);
+TORCH_API at::Tensor & lerp_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & end, const at::Scalar & weight);
+TORCH_API at::Tensor & lerp_outf(const at::Tensor & self, const at::Tensor & end, const at::Scalar & weight, at::Tensor & out);
+TORCH_API at::Tensor & lerp_(at::Tensor & self, const at::Tensor & end, const at::Scalar & weight);
+TORCH_API at::Tensor lerp(const at::Tensor & self, const at::Tensor & end, const at::Tensor & weight);
+TORCH_API at::Tensor & lerp_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & end, const at::Tensor & weight);
+TORCH_API at::Tensor & lerp_outf(const at::Tensor & self, const at::Tensor & end, const at::Tensor & weight, at::Tensor & out);
+TORCH_API at::Tensor & lerp_(at::Tensor & self, const at::Tensor & end, const at::Tensor & weight);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_cholesky_ex_cpu_dispatch.h

@@ -0,0 +1,25 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> linalg_cholesky_ex(const at::Tensor & self, bool upper=false, bool check_errors=false);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> linalg_cholesky_ex_out(at::Tensor & L, at::Tensor & info, const at::Tensor & self, bool upper=false, bool check_errors=false);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> linalg_cholesky_ex_outf(const at::Tensor & self, bool upper, bool check_errors, at::Tensor & L, at::Tensor & info);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_eigh_compositeimplicitautograd_dispatch.h

@@ -0,0 +1,25 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeimplicitautograd {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> linalg_eigh(const at::Tensor & self, c10::string_view UPLO="L");
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> linalg_eigh_out(at::Tensor & eigvals, at::Tensor & eigvecs, const at::Tensor & self, c10::string_view UPLO="L");
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> linalg_eigh_outf(const at::Tensor & self, c10::string_view UPLO, at::Tensor & eigvals, at::Tensor & eigvecs);
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_ldl_solve_native.h

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+#include <ATen/ops/linalg_ldl_solve_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_linalg_ldl_solve_out : public at::meta::structured_linalg_ldl_solve {
+void impl(const at::Tensor & LD, const at::Tensor & pivots, const at::Tensor & B, bool hermitian, const at::Tensor & out);
+};
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_factor_ex_cuda_dispatch.h

@@ -0,0 +1,25 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor> linalg_lu_factor_ex(const at::Tensor & A, bool pivot=true, bool check_errors=false);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> linalg_lu_factor_ex_out(at::Tensor & LU, at::Tensor & pivots, at::Tensor & info, const at::Tensor & A, bool pivot=true, bool check_errors=false);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> linalg_lu_factor_ex_outf(const at::Tensor & A, bool pivot, bool check_errors, at::Tensor & LU, at::Tensor & pivots, at::Tensor & info);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_factor_ops.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API linalg_lu_factor {
+  using schema = ::std::tuple<at::Tensor,at::Tensor> (const at::Tensor &, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_lu_factor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_lu_factor(Tensor A, *, bool pivot=True) -> (Tensor LU, Tensor pivots)")
+  static ::std::tuple<at::Tensor,at::Tensor> call(const at::Tensor & A, bool pivot);
+  static ::std::tuple<at::Tensor,at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & A, bool pivot);
+};
+
+struct TORCH_API linalg_lu_factor_out {
+  using schema = ::std::tuple<at::Tensor &,at::Tensor &> (const at::Tensor &, bool, at::Tensor &, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_lu_factor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_lu_factor.out(Tensor A, *, bool pivot=True, Tensor(a!) LU, Tensor(b!) pivots) -> (Tensor(a!) LU, Tensor(b!) pivots)")
+  static ::std::tuple<at::Tensor &,at::Tensor &> call(const at::Tensor & A, bool pivot, at::Tensor & LU, at::Tensor & pivots);
+  static ::std::tuple<at::Tensor &,at::Tensor &> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & A, bool pivot, at::Tensor & LU, at::Tensor & pivots);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_vector_norm_compositeexplicitautogradnonfunctional_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeexplicitautogradnonfunctional {
+
+TORCH_API at::Tensor linalg_vector_norm(const at::Tensor & self, const at::Scalar & ord=2, at::OptionalIntArrayRef dim=c10::nullopt, bool keepdim=false, c10::optional<at::ScalarType> dtype=c10::nullopt);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logspace_ops.h

@@ -0,0 +1,105 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API logspace {
+  using schema = at::Tensor (const at::Scalar &, const at::Scalar &, int64_t, double, c10::optional<at::ScalarType>, c10::optional<at::Layout>, c10::optional<at::Device>, c10::optional<bool>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::logspace")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "logspace(Scalar start, Scalar end, int steps, float base=10.0, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor")
+  static at::Tensor call(const at::Scalar & start, const at::Scalar & end, int64_t steps, double base, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & start, const at::Scalar & end, int64_t steps, double base, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+};
+
+struct TORCH_API logspace_Tensor_Tensor {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, int64_t, double, c10::optional<at::ScalarType>, c10::optional<at::Layout>, c10::optional<at::Device>, c10::optional<bool>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::logspace")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "logspace.Tensor_Tensor(Tensor start, Tensor end, int steps, float base=10.0, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & start, const at::Tensor & end, int64_t steps, double base, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & start, const at::Tensor & end, int64_t steps, double base, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+};
+
+struct TORCH_API logspace_Tensor_Scalar {
+  using schema = at::Tensor (const at::Tensor &, const at::Scalar &, int64_t, double, c10::optional<at::ScalarType>, c10::optional<at::Layout>, c10::optional<at::Device>, c10::optional<bool>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::logspace")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Scalar")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "logspace.Tensor_Scalar(Tensor start, Scalar end, int steps, float base=10.0, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & start, const at::Scalar & end, int64_t steps, double base, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & start, const at::Scalar & end, int64_t steps, double base, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+};
+
+struct TORCH_API logspace_Scalar_Tensor {
+  using schema = at::Tensor (const at::Scalar &, const at::Tensor &, int64_t, double, c10::optional<at::ScalarType>, c10::optional<at::Layout>, c10::optional<at::Device>, c10::optional<bool>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::logspace")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "logspace.Scalar_Tensor(Scalar start, Tensor end, int steps, float base=10.0, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor")
+  static at::Tensor call(const at::Scalar & start, const at::Tensor & end, int64_t steps, double base, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & start, const at::Tensor & end, int64_t steps, double base, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory);
+};
+
+struct TORCH_API logspace_out {
+  using schema = at::Tensor & (const at::Scalar &, const at::Scalar &, int64_t, double, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::logspace")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "logspace.out(Scalar start, Scalar end, int steps, float base=10.0, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Scalar & start, const at::Scalar & end, int64_t steps, double base, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & start, const at::Scalar & end, int64_t steps, double base, at::Tensor & out);
+};
+
+struct TORCH_API logspace_Tensor_Tensor_out {
+  using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, int64_t, double, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::logspace")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Tensor_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "logspace.Tensor_Tensor_out(Tensor start, Tensor end, int steps, float base=10.0, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & start, const at::Tensor & end, int64_t steps, double base, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & start, const at::Tensor & end, int64_t steps, double base, at::Tensor & out);
+};
+
+struct TORCH_API logspace_Tensor_Scalar_out {
+  using schema = at::Tensor & (const at::Tensor &, const at::Scalar &, int64_t, double, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::logspace")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Scalar_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "logspace.Tensor_Scalar_out(Tensor start, Scalar end, int steps, float base=10.0, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & start, const at::Scalar & end, int64_t steps, double base, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & start, const at::Scalar & end, int64_t steps, double base, at::Tensor & out);
+};
+
+struct TORCH_API logspace_Scalar_Tensor_out {
+  using schema = at::Tensor & (const at::Scalar &, const at::Tensor &, int64_t, double, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::logspace")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_Tensor_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "logspace.Scalar_Tensor_out(Scalar start, Tensor end, int steps, float base=10.0, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Scalar & start, const at::Tensor & end, int64_t steps, double base, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & start, const at::Tensor & end, int64_t steps, double base, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lshift.h

@@ -0,0 +1,53 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/lshift_ops.h>
+
+namespace at {
+
+
+// aten::__lshift__.Scalar(Tensor self, Scalar other) -> Tensor
+inline at::Tensor __lshift__(const at::Tensor & self, const at::Scalar & other) {
+    return at::_ops::__lshift___Scalar::call(self, other);
+}
+
+// aten::__lshift__.Tensor(Tensor self, Tensor other) -> Tensor
+inline at::Tensor __lshift__(const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::__lshift___Tensor::call(self, other);
+}
+
+// aten::__lshift__.Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & __lshift___out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other) {
+    return at::_ops::__lshift___Scalar_out::call(self, other, out);
+}
+// aten::__lshift__.Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & __lshift___outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out) {
+    return at::_ops::__lshift___Scalar_out::call(self, other, out);
+}
+
+// aten::__lshift__.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & __lshift___out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::__lshift___Tensor_out::call(self, other, out);
+}
+// aten::__lshift__.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & __lshift___outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) {
+    return at::_ops::__lshift___Tensor_out::call(self, other, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lstm_compositeimplicitautograd_dispatch.h

@@ -0,0 +1,24 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeimplicitautograd {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor> lstm(const at::Tensor & input, at::TensorList hx, at::TensorList params, bool has_biases, int64_t num_layers, double dropout, bool train, bool bidirectional, bool batch_first);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor> lstm(const at::Tensor & data, const at::Tensor & batch_sizes, at::TensorList hx, at::TensorList params, bool has_biases, int64_t num_layers, double dropout, bool train, bool bidirectional);
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_meta.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeMetaFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/TensorIterator.h>
+#include <ATen/TensorMeta.h>
+#include <tuple>
+#include <vector>
+
+namespace at {
+namespace meta {
+
+struct TORCH_API structured_max_dim : public at::impl::MetaBase {
+    
+                template <bool DIM = false>
+                struct TORCH_API precompute_out {
+                    
+                    precompute_out<true> set_dim(int64_t value) {
+                        static_assert(DIM == false, "dim already set");
+                        precompute_out<true> ret;
+ret.dim = value;
+return ret;
+                    }
+                
+                    int64_t dim;
+            };
+    using meta_return_ty = precompute_out <true>;
+    meta_return_ty meta(const at::Tensor & self, int64_t dim, bool keepdim);
+};
+
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_unpool3d_cuda_dispatch.h

@@ -0,0 +1,28 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API at::Tensor max_unpool3d(const at::Tensor & self, const at::Tensor & indices, at::IntArrayRef output_size, at::IntArrayRef stride, at::IntArrayRef padding);
+TORCH_API at::Tensor max_unpool3d_symint(const at::Tensor & self, const at::Tensor & indices, c10::SymIntArrayRef output_size, at::IntArrayRef stride, at::IntArrayRef padding);
+TORCH_API at::Tensor & max_unpool3d_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & indices, at::IntArrayRef output_size, at::IntArrayRef stride, at::IntArrayRef padding);
+TORCH_API at::Tensor & max_unpool3d_outf(const at::Tensor & self, const at::Tensor & indices, at::IntArrayRef output_size, at::IntArrayRef stride, at::IntArrayRef padding, at::Tensor & out);
+TORCH_API at::Tensor & max_unpool3d_symint_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & indices, c10::SymIntArrayRef output_size, at::IntArrayRef stride, at::IntArrayRef padding);
+TORCH_API at::Tensor & max_unpool3d_symint_outf(const at::Tensor & self, const at::Tensor & indices, c10::SymIntArrayRef output_size, at::IntArrayRef stride, at::IntArrayRef padding, at::Tensor & out);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/minimum_ops.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API minimum {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::minimum")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "minimum(Tensor self, Tensor other) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const at::Tensor & other);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other);
+};
+
+struct TORCH_API minimum_out {
+  using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::minimum")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "minimum.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_convolution_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor & miopen_convolution_out_symint(const at::Tensor & self, const at::Tensor & weight, const c10::optional<at::Tensor> & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, at::Tensor & out);
+TORCH_API at::Tensor miopen_convolution(const at::Tensor & self, const at::Tensor & weight, const c10::optional<at::Tensor> & bias, at::IntArrayRef padding, at::IntArrayRef stride, at::IntArrayRef dilation, int64_t groups, bool benchmark, bool deterministic);
+} // namespace native
+} // namespace at