Add files using upload-large-folder tool

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_add_batch_dim_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 _add_batch_dim {
+  using schema = at::Tensor (const at::Tensor &, 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::_add_batch_dim")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_add_batch_dim(Tensor self, int batch_dim, int level) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, int64_t batch_dim, int64_t level);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t batch_dim, int64_t level);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cast_Char.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_Char_ops.h>
+
+namespace at {
+
+
+// aten::_cast_Char(Tensor self, bool non_blocking=False) -> Tensor
+inline at::Tensor _cast_Char(const at::Tensor & self, bool non_blocking=false) {
+    return at::_ops::_cast_Char::call(self, non_blocking);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cast_Int_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 _cast_Int(const at::Tensor & self, bool non_blocking=false);
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

@@ -0,0 +1,50 @@
+#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 _coalesced_ {
+  using schema = at::Tensor & (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::_coalesced_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_coalesced_(Tensor(a!) self, bool coalesced) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, bool coalesced);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, bool coalesced);
+};
+
+struct TORCH_API _coalesced_out {
+  using schema = at::Tensor & (const at::Tensor &, bool, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_coalesced")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_coalesced.out(Tensor self, bool coalesced, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, bool coalesced, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, bool coalesced, at::Tensor & out);
+};
+
+struct TORCH_API _coalesced {
+  using schema = 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::_coalesced")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_coalesced(Tensor self, bool coalesced) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, bool coalesced);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, bool coalesced);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_conj_copy.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/_conj_copy_ops.h>
+
+namespace at {
+
+
+// aten::_conj_copy(Tensor self) -> Tensor
+inline at::Tensor _conj_copy(const at::Tensor & self) {
+    return at::_ops::_conj_copy::call(self);
+}
+
+// aten::_conj_copy.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _conj_copy_out(at::Tensor & out, const at::Tensor & self) {
+    return at::_ops::_conj_copy_out::call(self, out);
+}
+// aten::_conj_copy.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _conj_copy_outf(const at::Tensor & self, at::Tensor & out) {
+    return at::_ops::_conj_copy_out::call(self, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_copy_from_and_resize_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 _copy_from_and_resize {
+  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::_copy_from_and_resize")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_copy_from_and_resize(Tensor self, Tensor dst) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const at::Tensor & dst);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & dst);
+};
+
+struct TORCH_API _copy_from_and_resize_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::_copy_from_and_resize")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_copy_from_and_resize.out(Tensor self, Tensor dst, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, const at::Tensor & dst, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & dst, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cudnn_rnn_backward_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 void _cudnn_rnn_backward_out_symint(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);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor,::std::vector<at::Tensor>> _cudnn_rnn_backward(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, int64_t hidden_size, int64_t proj_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, at::IntArrayRef batch_sizes, const c10::optional<at::Tensor> & dropout_state, const at::Tensor & reserve, ::std::array<bool,4> output_mask);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,44 @@
+#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_abs_ops.h>
+
+namespace at {
+
+
+// aten::_foreach_abs(Tensor[] self) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_abs(at::TensorList self) {
+    return at::_ops::_foreach_abs::call(self);
+}
+
+// aten::_foreach_abs_(Tensor(a!)[] self) -> ()
+inline void _foreach_abs_(at::TensorList self) {
+    return at::_ops::_foreach_abs_::call(self);
+}
+
+// aten::_foreach_abs.out(Tensor[] self, *, Tensor(a!)[] out) -> ()
+inline void _foreach_abs_out(at::TensorList out, at::TensorList self) {
+    return at::_ops::_foreach_abs_out::call(self, out);
+}
+// aten::_foreach_abs.out(Tensor[] self, *, Tensor(a!)[] out) -> ()
+inline void _foreach_abs_outf(at::TensorList self, at::TensorList out) {
+    return at::_ops::_foreach_abs_out::call(self, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sign_cuda_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 cuda {
+
+TORCH_API ::std::vector<at::Tensor> _foreach_sign(at::TensorList self);
+TORCH_API void _foreach_sign_(at::TensorList self);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_dropout_compositeexplicitautograd_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 compositeexplicitautograd {
+
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> _fused_dropout_out(at::Tensor & out0, at::Tensor & out1, const at::Tensor & self, double p, c10::optional<at::Generator> generator=c10::nullopt);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> _fused_dropout_outf(const at::Tensor & self, double p, c10::optional<at::Generator> generator, at::Tensor & out0, at::Tensor & out1);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl_cpu_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 cpu {
+
+TORCH_API at::Tensor & _index_put_impl_(at::Tensor & self, const c10::List<c10::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_neg_view_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 _neg_view(const at::Tensor & self);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_reshape_alias_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 at::Tensor _reshape_alias(const at::Tensor & self, at::IntArrayRef size, at::IntArrayRef stride);
+TORCH_API at::Tensor _reshape_alias_symint(const at::Tensor & self, c10::SymIntArrayRef size, c10::SymIntArrayRef stride);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_slow_conv2d_backward_ops.h

@@ -0,0 +1,50 @@
+#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 _slow_conv2d_backward_grad_input {
+  using schema = ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> (const at::Tensor &, const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymIntArrayRef, at::Tensor &, 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::_slow_conv2d_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_slow_conv2d_backward.grad_input(Tensor grad_output, Tensor self, Tensor weight, SymInt[2] kernel_size, SymInt[2] stride, SymInt[2] padding, *, Tensor(a!) grad_input, Tensor(b!) grad_weight, Tensor(c!) grad_bias) -> (Tensor(a!), Tensor(b!), Tensor(c!))")
+  static ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> call(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, at::Tensor & grad_input, at::Tensor & grad_weight, at::Tensor & grad_bias);
+  static ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, at::Tensor & grad_input, at::Tensor & grad_weight, at::Tensor & grad_bias);
+};
+
+struct TORCH_API _slow_conv2d_backward_output_mask {
+  using schema = ::std::tuple<at::Tensor,at::Tensor,at::Tensor> (const at::Tensor &, const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymIntArrayRef, ::std::array<bool,3>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_slow_conv2d_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "output_mask")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_slow_conv2d_backward.output_mask(Tensor grad_output, Tensor self, Tensor weight, SymInt[2] kernel_size, SymInt[2] stride, SymInt[2] padding, bool[3] output_mask) -> (Tensor grad_input, Tensor grad_weight, Tensor grad_bias)")
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor> call(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, ::std::array<bool,3> output_mask);
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, ::std::array<bool,3> output_mask);
+};
+
+struct TORCH_API _slow_conv2d_backward_output_mask_out {
+  using schema = ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> (const at::Tensor &, const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymIntArrayRef, ::std::array<bool,3>, at::Tensor &, 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::_slow_conv2d_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "output_mask_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_slow_conv2d_backward.output_mask_out(Tensor grad_output, Tensor self, Tensor weight, SymInt[2] kernel_size, SymInt[2] stride, SymInt[2] padding, bool[3] output_mask, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!))")
+  static ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> call(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, ::std::array<bool,3> output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2);
+  static ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, ::std::array<bool,3> output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2);
+};
+
+}} // namespace at::_ops

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

@@ -0,0 +1,34 @@
+#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/_sparse_bsr_tensor_unsafe_ops.h>
+
+namespace at {
+
+
+// aten::_sparse_bsr_tensor_unsafe(Tensor crow_indices, Tensor col_indices, Tensor values, int[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor _sparse_bsr_tensor_unsafe(const at::Tensor & crow_indices, const at::Tensor & col_indices, const at::Tensor & values, at::IntArrayRef size, at::TensorOptions options={}) {
+    return at::_ops::_sparse_bsr_tensor_unsafe::call(crow_indices, col_indices, values, size, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+}
+// aten::_sparse_bsr_tensor_unsafe(Tensor crow_indices, Tensor col_indices, Tensor values, int[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor _sparse_bsr_tensor_unsafe(const at::Tensor & crow_indices, const at::Tensor & col_indices, const at::Tensor & values, at::IntArrayRef size, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory) {
+    return at::_ops::_sparse_bsr_tensor_unsafe::call(crow_indices, col_indices, values, size, dtype, layout, device, pin_memory);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_mm_reduce_impl_backward_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 ::std::tuple<at::Tensor,at::Tensor> _sparse_mm_reduce_impl_backward_sparse_csr_cpu(const at::Tensor & self, const at::Tensor & grad_out, const at::Tensor & weight, c10::string_view reduce, const at::Tensor & arg_out, ::std::array<bool,2> output_mask);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_semi_structured_linear.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/_sparse_semi_structured_linear_ops.h>
+
+namespace at {
+
+
+// aten::_sparse_semi_structured_linear(Tensor input, Tensor weight, Tensor meta, *, Tensor? bias=None, str? activation=None, ScalarType? out_dtype=None) -> Tensor
+inline at::Tensor _sparse_semi_structured_linear(const at::Tensor & input, const at::Tensor & weight, const at::Tensor & meta, const c10::optional<at::Tensor> & bias={}, c10::optional<c10::string_view> activation=c10::nullopt, c10::optional<at::ScalarType> out_dtype=c10::nullopt) {
+    return at::_ops::_sparse_semi_structured_linear::call(input, weight, meta, bias, activation, out_dtype);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bicubic2d_aa_ops.h

@@ -0,0 +1,50 @@
+#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 _upsample_bicubic2d_aa_vec {
+  using schema = at::Tensor (const at::Tensor &, at::OptionalSymIntArrayRef, bool, c10::optional<at::ArrayRef<double>>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_upsample_bicubic2d_aa")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "vec")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_upsample_bicubic2d_aa.vec(Tensor input, SymInt[]? output_size, bool align_corners, float[]? scale_factors) -> Tensor")
+  static at::Tensor call(const at::Tensor & input, at::OptionalSymIntArrayRef output_size, bool align_corners, c10::optional<at::ArrayRef<double>> scale_factors);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, at::OptionalSymIntArrayRef output_size, bool align_corners, c10::optional<at::ArrayRef<double>> scale_factors);
+};
+
+struct TORCH_API _upsample_bicubic2d_aa_out {
+  using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, bool, c10::optional<double>, c10::optional<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::_upsample_bicubic2d_aa")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_upsample_bicubic2d_aa.out(Tensor self, SymInt[2] output_size, bool align_corners, float? scales_h=None, float? scales_w=None, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w, at::Tensor & out);
+};
+
+struct TORCH_API _upsample_bicubic2d_aa {
+  using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef, bool, c10::optional<double>, c10::optional<double>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_upsample_bicubic2d_aa")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_upsample_bicubic2d_aa(Tensor self, SymInt[2] output_size, bool align_corners, float? scales_h=None, float? scales_w=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact2d_backward_cpu_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 cpu {
+
+TORCH_API at::Tensor _upsample_nearest_exact2d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional<double> scales_h=c10::nullopt, c10::optional<double> scales_w=c10::nullopt);
+TORCH_API at::Tensor _upsample_nearest_exact2d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional<double> scales_h=c10::nullopt, c10::optional<double> scales_w=c10::nullopt);
+TORCH_API at::Tensor & _upsample_nearest_exact2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional<double> scales_h=c10::nullopt, c10::optional<double> scales_w=c10::nullopt);
+TORCH_API at::Tensor & _upsample_nearest_exact2d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional<double> scales_h, c10::optional<double> scales_w, at::Tensor & grad_input);
+TORCH_API at::Tensor & _upsample_nearest_exact2d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional<double> scales_h=c10::nullopt, c10::optional<double> scales_w=c10::nullopt);
+TORCH_API at::Tensor & _upsample_nearest_exact2d_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional<double> scales_h, c10::optional<double> scales_w, at::Tensor & grad_input);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_use_cudnn_rnn_flatten_weight_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 bool _use_cudnn_rnn_flatten_weight();
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool3d_backward_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 & adaptive_avg_pool3d_backward_out_cpu(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & grad_input);
+TORCH_API at::Tensor & adaptive_avg_pool3d_backward_out_cuda(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & grad_input);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addcdiv.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/addcdiv_ops.h>
+
+namespace at {
+
+
+// aten::addcdiv.out(Tensor self, Tensor tensor1, Tensor tensor2, *, Scalar value=1, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & addcdiv_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & tensor1, const at::Tensor & tensor2, const at::Scalar & value=1) {
+    return at::_ops::addcdiv_out::call(self, tensor1, tensor2, value, out);
+}
+// aten::addcdiv.out(Tensor self, Tensor tensor1, Tensor tensor2, *, Scalar value=1, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & addcdiv_outf(const at::Tensor & self, const at::Tensor & tensor1, const at::Tensor & tensor2, const at::Scalar & value, at::Tensor & out) {
+    return at::_ops::addcdiv_out::call(self, tensor1, tensor2, value, out);
+}
+
+// aten::addcdiv(Tensor self, Tensor tensor1, Tensor tensor2, *, Scalar value=1) -> Tensor
+inline at::Tensor addcdiv(const at::Tensor & self, const at::Tensor & tensor1, const at::Tensor & tensor2, const at::Scalar & value=1) {
+    return at::_ops::addcdiv::call(self, tensor1, tensor2, value);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addmm_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_addmm : public at::impl::MetaBase {
+    
+    
+    void meta(const at::Tensor & self, const at::Tensor & mat1, const at::Tensor & mat2, const at::Scalar & beta, const at::Scalar & alpha);
+};
+
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,44 @@
+#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/arccos_ops.h>
+
+namespace at {
+
+
+// aten::arccos(Tensor self) -> Tensor
+inline at::Tensor arccos(const at::Tensor & self) {
+    return at::_ops::arccos::call(self);
+}
+
+// aten::arccos_(Tensor(a!) self) -> Tensor(a!)
+inline at::Tensor & arccos_(at::Tensor & self) {
+    return at::_ops::arccos_::call(self);
+}
+
+// aten::arccos.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & arccos_out(at::Tensor & out, const at::Tensor & self) {
+    return at::_ops::arccos_out::call(self, out);
+}
+// aten::arccos.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & arccos_outf(const at::Tensor & self, at::Tensor & out) {
+    return at::_ops::arccos_out::call(self, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/argsort_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 at::Tensor argsort(const at::Tensor & self, bool stable, int64_t dim=-1, bool descending=false);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/as_strided_scatter_compositeexplicitautogradnonfunctional_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 compositeexplicitautogradnonfunctional {
+
+TORCH_API at::Tensor as_strided_scatter(const at::Tensor & self, const at::Tensor & src, at::IntArrayRef size, at::IntArrayRef stride, c10::optional<int64_t> storage_offset=c10::nullopt);
+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 compositeexplicitautogradnonfunctional
+} // namespace at

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

@@ -0,0 +1,67 @@
+#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/bitwise_and_ops.h>
+
+namespace at {
+
+
+// aten::bitwise_and.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_and_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::bitwise_and_Tensor_out::call(self, other, out);
+}
+// aten::bitwise_and.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_and_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) {
+    return at::_ops::bitwise_and_Tensor_out::call(self, other, out);
+}
+
+// aten::bitwise_and.Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_and_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other) {
+    return at::_ops::bitwise_and_Scalar_out::call(self, other, out);
+}
+// aten::bitwise_and.Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_and_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out) {
+    return at::_ops::bitwise_and_Scalar_out::call(self, other, out);
+}
+
+// aten::bitwise_and.Scalar(Tensor self, Scalar other) -> Tensor
+inline at::Tensor bitwise_and(const at::Tensor & self, const at::Scalar & other) {
+    return at::_ops::bitwise_and_Scalar::call(self, other);
+}
+
+// aten::bitwise_and.Scalar_Tensor(Scalar self, Tensor other) -> Tensor
+inline at::Tensor bitwise_and(const at::Scalar & self, const at::Tensor & other) {
+    return at::_ops::bitwise_and_Scalar_Tensor::call(self, other);
+}
+
+// aten::bitwise_and.Tensor(Tensor self, Tensor other) -> Tensor
+inline at::Tensor bitwise_and(const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::bitwise_and_Tensor::call(self, other);
+}
+
+// aten::bitwise_and.Scalar_Tensor_out(Scalar self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_and_out(at::Tensor & out, const at::Scalar & self, const at::Tensor & other) {
+    return at::_ops::bitwise_and_Scalar_Tensor_out::call(self, other, out);
+}
+// aten::bitwise_and.Scalar_Tensor_out(Scalar self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_and_outf(const at::Scalar & self, const at::Tensor & other, at::Tensor & out) {
+    return at::_ops::bitwise_and_Scalar_Tensor_out::call(self, other, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cartesian_prod_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 cartesian_prod(at::TensorList tensors);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/chain_matmul_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 chain_matmul(at::TensorList matrices);
+TORCH_API at::Tensor & chain_matmul_out(at::TensorList matrices, at::Tensor & out);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,50 @@
+#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 conj_physical {
+  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::conj_physical")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "conj_physical(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 conj_physical_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::conj_physical")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "conj_physical.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);
+};
+
+struct TORCH_API conj_physical_ {
+  using schema = 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::conj_physical_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "conj_physical_(Tensor(a!) self) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/copy_compositeexplicitautograd_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 compositeexplicitautograd {
+
+TORCH_API at::Tensor & copy_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & src, bool non_blocking=false);
+TORCH_API at::Tensor & copy_outf(const at::Tensor & self, const at::Tensor & src, bool non_blocking, at::Tensor & out);
+TORCH_API at::Tensor & copy_(at::Tensor & self, const at::Tensor & src, bool non_blocking=false);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_convolution_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_convolution {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymInt, bool, bool, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cudnn_convolution")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cudnn_convolution(Tensor self, Tensor weight, SymInt[] padding, SymInt[] stride, SymInt[] dilation, SymInt groups, bool benchmark, bool deterministic, bool allow_tf32) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, bool allow_tf32);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, bool allow_tf32);
+};
+
+struct TORCH_API cudnn_convolution_out {
+  using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymInt, bool, bool, bool, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cudnn_convolution")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cudnn_convolution.out(Tensor self, Tensor weight, SymInt[] padding, SymInt[] stride, SymInt[] dilation, SymInt groups, bool benchmark, bool deterministic, bool allow_tf32, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, bool allow_tf32, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, bool allow_tf32, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cummax_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> cummax(const at::Tensor & self, at::Dimname dim);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> cummax_out(at::Tensor & values, at::Tensor & indices, const at::Tensor & self, at::Dimname dim);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> cummax_outf(const at::Tensor & self, at::Dimname dim, at::Tensor & values, at::Tensor & indices);
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cummaxmin_backward_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 cummaxmin_backward(const at::Tensor & grad, const at::Tensor & input, const at::Tensor & indices, int64_t dim);
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach_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 detach(const at::Tensor & self);
+TORCH_API at::Tensor & detach_(at::Tensor & self);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,131 @@
+#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/empty_ops.h>
+
+namespace at {
+
+
+// aten::empty.names(int[] size, *, Dimname[]? names, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor
+inline at::Tensor empty(at::IntArrayRef size, c10::optional<at::DimnameList> names, at::TensorOptions options={}, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_names::call(size, names, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format));
+}
+// aten::empty.names(int[] size, *, Dimname[]? names, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor
+inline at::Tensor empty(at::IntArrayRef size, 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, c10::optional<at::MemoryFormat> memory_format) {
+    return at::_ops::empty_names::call(size, names, dtype, layout, device, pin_memory, memory_format);
+}
+
+// aten::empty.memory_format(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor
+inline at::Tensor empty(at::IntArrayRef size, at::TensorOptions options={}, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_memory_format::call(c10::fromIntArrayRefSlow(size), c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format));
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor empty(at::IntArrayRef size, at::TensorOptions options={}, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_memory_format::call(c10::fromIntArrayRefSlow(size), c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format));
+  }
+}
+
+// aten::empty.memory_format(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor
+inline at::Tensor empty(at::IntArrayRef size, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory, c10::optional<at::MemoryFormat> memory_format) {
+    return at::_ops::empty_memory_format::call(c10::fromIntArrayRefSlow(size), dtype, layout, device, pin_memory, memory_format);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor empty(at::IntArrayRef size, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory, c10::optional<at::MemoryFormat> memory_format) {
+    return at::_ops::empty_memory_format::call(c10::fromIntArrayRefSlow(size), dtype, layout, device, pin_memory, memory_format);
+  }
+}
+
+// aten::empty.memory_format(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor
+inline at::Tensor empty_symint(c10::SymIntArrayRef size, at::TensorOptions options={}, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_memory_format::call(size, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format));
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor empty(c10::SymIntArrayRef size, at::TensorOptions options={}, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_memory_format::call(size, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt(), c10::impl::check_tensor_options_and_extract_memory_format(options, memory_format));
+  }
+}
+
+// aten::empty.memory_format(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor
+inline at::Tensor empty_symint(c10::SymIntArrayRef size, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory, c10::optional<at::MemoryFormat> memory_format) {
+    return at::_ops::empty_memory_format::call(size, dtype, layout, device, pin_memory, memory_format);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor empty(c10::SymIntArrayRef size, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory, c10::optional<at::MemoryFormat> memory_format) {
+    return at::_ops::empty_memory_format::call(size, dtype, layout, device, pin_memory, memory_format);
+  }
+}
+
+// aten::empty.out(SymInt[] size, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_out(at::Tensor & out, at::IntArrayRef size, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_out::call(c10::fromIntArrayRefSlow(size), memory_format, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & empty_out(at::Tensor & out, at::IntArrayRef size, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_out::call(c10::fromIntArrayRefSlow(size), memory_format, out);
+  }
+}
+
+// aten::empty.out(SymInt[] size, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_outf(at::IntArrayRef size, c10::optional<at::MemoryFormat> memory_format, at::Tensor & out) {
+    return at::_ops::empty_out::call(c10::fromIntArrayRefSlow(size), memory_format, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & empty_outf(at::IntArrayRef size, c10::optional<at::MemoryFormat> memory_format, at::Tensor & out) {
+    return at::_ops::empty_out::call(c10::fromIntArrayRefSlow(size), memory_format, out);
+  }
+}
+
+// aten::empty.out(SymInt[] size, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_symint_out(at::Tensor & out, c10::SymIntArrayRef size, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_out::call(size, memory_format, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & empty_out(at::Tensor & out, c10::SymIntArrayRef size, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_out::call(size, memory_format, out);
+  }
+}
+
+// aten::empty.out(SymInt[] size, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_symint_outf(c10::SymIntArrayRef size, c10::optional<at::MemoryFormat> memory_format, at::Tensor & out) {
+    return at::_ops::empty_out::call(size, memory_format, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & empty_outf(c10::SymIntArrayRef size, c10::optional<at::MemoryFormat> memory_format, at::Tensor & out) {
+    return at::_ops::empty_out::call(size, memory_format, out);
+  }
+}
+
+// aten::empty.names_out(int[] size, *, Dimname[]? names, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_out(at::Tensor & out, at::IntArrayRef size, c10::optional<at::DimnameList> names, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_names_out::call(size, names, memory_format, out);
+}
+// aten::empty.names_out(int[] size, *, Dimname[]? names, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_outf(at::IntArrayRef size, c10::optional<at::DimnameList> names, c10::optional<at::MemoryFormat> memory_format, at::Tensor & out) {
+    return at::_ops::empty_names_out::call(size, names, memory_format, out);
+}
+
+}

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

@@ -0,0 +1,113 @@
+#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/empty_permuted_ops.h>
+
+namespace at {
+
+
+// aten::empty_permuted(SymInt[] size, int[] physical_layout, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor empty_permuted(at::IntArrayRef size, at::IntArrayRef physical_layout, at::TensorOptions options={}) {
+    return at::_ops::empty_permuted::call(c10::fromIntArrayRefSlow(size), physical_layout, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor empty_permuted(at::IntArrayRef size, at::IntArrayRef physical_layout, at::TensorOptions options={}) {
+    return at::_ops::empty_permuted::call(c10::fromIntArrayRefSlow(size), physical_layout, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+  }
+}
+
+// aten::empty_permuted(SymInt[] size, int[] physical_layout, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor empty_permuted(at::IntArrayRef size, at::IntArrayRef physical_layout, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory) {
+    return at::_ops::empty_permuted::call(c10::fromIntArrayRefSlow(size), physical_layout, dtype, layout, device, pin_memory);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor empty_permuted(at::IntArrayRef size, at::IntArrayRef physical_layout, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory) {
+    return at::_ops::empty_permuted::call(c10::fromIntArrayRefSlow(size), physical_layout, dtype, layout, device, pin_memory);
+  }
+}
+
+// aten::empty_permuted(SymInt[] size, int[] physical_layout, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor empty_permuted_symint(c10::SymIntArrayRef size, at::IntArrayRef physical_layout, at::TensorOptions options={}) {
+    return at::_ops::empty_permuted::call(size, physical_layout, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor empty_permuted(c10::SymIntArrayRef size, at::IntArrayRef physical_layout, at::TensorOptions options={}) {
+    return at::_ops::empty_permuted::call(size, physical_layout, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+  }
+}
+
+// aten::empty_permuted(SymInt[] size, int[] physical_layout, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor empty_permuted_symint(c10::SymIntArrayRef size, at::IntArrayRef physical_layout, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory) {
+    return at::_ops::empty_permuted::call(size, physical_layout, dtype, layout, device, pin_memory);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor empty_permuted(c10::SymIntArrayRef size, at::IntArrayRef physical_layout, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory) {
+    return at::_ops::empty_permuted::call(size, physical_layout, dtype, layout, device, pin_memory);
+  }
+}
+
+// aten::empty_permuted.out(SymInt[] size, int[] physical_layout, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_permuted_out(at::Tensor & out, at::IntArrayRef size, at::IntArrayRef physical_layout) {
+    return at::_ops::empty_permuted_out::call(c10::fromIntArrayRefSlow(size), physical_layout, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & empty_permuted_out(at::Tensor & out, at::IntArrayRef size, at::IntArrayRef physical_layout) {
+    return at::_ops::empty_permuted_out::call(c10::fromIntArrayRefSlow(size), physical_layout, out);
+  }
+}
+
+// aten::empty_permuted.out(SymInt[] size, int[] physical_layout, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_permuted_outf(at::IntArrayRef size, at::IntArrayRef physical_layout, at::Tensor & out) {
+    return at::_ops::empty_permuted_out::call(c10::fromIntArrayRefSlow(size), physical_layout, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & empty_permuted_outf(at::IntArrayRef size, at::IntArrayRef physical_layout, at::Tensor & out) {
+    return at::_ops::empty_permuted_out::call(c10::fromIntArrayRefSlow(size), physical_layout, out);
+  }
+}
+
+// aten::empty_permuted.out(SymInt[] size, int[] physical_layout, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_permuted_symint_out(at::Tensor & out, c10::SymIntArrayRef size, at::IntArrayRef physical_layout) {
+    return at::_ops::empty_permuted_out::call(size, physical_layout, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & empty_permuted_out(at::Tensor & out, c10::SymIntArrayRef size, at::IntArrayRef physical_layout) {
+    return at::_ops::empty_permuted_out::call(size, physical_layout, out);
+  }
+}
+
+// aten::empty_permuted.out(SymInt[] size, int[] physical_layout, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_permuted_symint_outf(c10::SymIntArrayRef size, at::IntArrayRef physical_layout, at::Tensor & out) {
+    return at::_ops::empty_permuted_out::call(size, physical_layout, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & empty_permuted_outf(c10::SymIntArrayRef size, at::IntArrayRef physical_layout, at::Tensor & out) {
+    return at::_ops::empty_permuted_out::call(size, physical_layout, out);
+  }
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_irfft2_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 fft_irfft2(const at::Tensor & self, at::OptionalIntArrayRef s=c10::nullopt, at::IntArrayRef dim={-2,-1}, c10::optional<c10::string_view> norm=c10::nullopt);
+TORCH_API at::Tensor fft_irfft2_symint(const at::Tensor & self, at::OptionalSymIntArrayRef s=c10::nullopt, at::IntArrayRef dim={-2,-1}, c10::optional<c10::string_view> norm=c10::nullopt);
+TORCH_API at::Tensor & fft_irfft2_out(at::Tensor & out, const at::Tensor & self, at::OptionalIntArrayRef s=c10::nullopt, at::IntArrayRef dim={-2,-1}, c10::optional<c10::string_view> norm=c10::nullopt);
+TORCH_API at::Tensor & fft_irfft2_outf(const at::Tensor & self, at::OptionalIntArrayRef s, at::IntArrayRef dim, c10::optional<c10::string_view> norm, at::Tensor & out);
+TORCH_API at::Tensor & fft_irfft2_symint_out(at::Tensor & out, const at::Tensor & self, at::OptionalSymIntArrayRef s=c10::nullopt, at::IntArrayRef dim={-2,-1}, c10::optional<c10::string_view> norm=c10::nullopt);
+TORCH_API at::Tensor & fft_irfft2_symint_outf(const at::Tensor & self, at::OptionalSymIntArrayRef s, at::IntArrayRef dim, c10::optional<c10::string_view> norm, at::Tensor & out);
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool3d.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/fractional_max_pool3d_ops.h>
+
+namespace at {
+
+
+// aten::fractional_max_pool3d.output(Tensor self, int[3] kernel_size, int[3] output_size, Tensor random_samples, *, Tensor(a!) output, Tensor(b!) indices) -> (Tensor(a!), Tensor(b!))
+inline ::std::tuple<at::Tensor &,at::Tensor &> fractional_max_pool3d_out(at::Tensor & output, at::Tensor & indices, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & random_samples) {
+    return at::_ops::fractional_max_pool3d_output::call(self, kernel_size, output_size, random_samples, output, indices);
+}
+// aten::fractional_max_pool3d.output(Tensor self, int[3] kernel_size, int[3] output_size, Tensor random_samples, *, Tensor(a!) output, Tensor(b!) indices) -> (Tensor(a!), Tensor(b!))
+inline ::std::tuple<at::Tensor &,at::Tensor &> fractional_max_pool3d_outf(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & random_samples, at::Tensor & output, at::Tensor & indices) {
+    return at::_ops::fractional_max_pool3d_output::call(self, kernel_size, output_size, random_samples, output, indices);
+}
+
+// aten::fractional_max_pool3d(Tensor self, int[3] kernel_size, int[3] output_size, Tensor random_samples) -> (Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor> fractional_max_pool3d(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & random_samples) {
+    return at::_ops::fractional_max_pool3d::call(self, kernel_size, output_size, random_samples);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fused_moving_avg_obs_fake_quant_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 fused_moving_avg_obs_fake_quant {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &, at::Tensor &, at::Tensor &, at::Tensor &, at::Tensor &, double, int64_t, int64_t, int64_t, bool, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::fused_moving_avg_obs_fake_quant")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fused_moving_avg_obs_fake_quant(Tensor self, Tensor observer_on, Tensor fake_quant_on, Tensor(a!) running_min, Tensor(b!) running_max, Tensor(c!) scale, Tensor(d!) zero_point, float averaging_const, int quant_min, int quant_max, int ch_axis, bool per_row_fake_quant=False, bool symmetric_quant=False) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const at::Tensor & observer_on, const at::Tensor & fake_quant_on, at::Tensor & running_min, at::Tensor & running_max, at::Tensor & scale, at::Tensor & zero_point, double averaging_const, int64_t quant_min, int64_t quant_max, int64_t ch_axis, bool per_row_fake_quant, bool symmetric_quant);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & observer_on, const at::Tensor & fake_quant_on, at::Tensor & running_min, at::Tensor & running_max, at::Tensor & scale, at::Tensor & zero_point, double averaging_const, int64_t quant_min, int64_t quant_max, int64_t ch_axis, bool per_row_fake_quant, bool symmetric_quant);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu_backward_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 at::Tensor glu_backward(const at::Tensor & grad_output, const at::Tensor & self, int64_t dim);
+TORCH_API at::Tensor & glu_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, int64_t dim);
+TORCH_API at::Tensor & glu_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, int64_t dim, at::Tensor & grad_input);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_3d_compositeexplicitautograd_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 compositeexplicitautograd {
+
+TORCH_API at::Tensor & grid_sampler_3d_out(at::Tensor & out, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners);
+TORCH_API at::Tensor & grid_sampler_3d_outf(const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gru_cell_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 gru_cell(const at::Tensor & input, const at::Tensor & hx, const at::Tensor & w_ih, const at::Tensor & w_hh, const c10::optional<at::Tensor> & b_ih={}, const c10::optional<at::Tensor> & b_hh={});
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

@@ -0,0 +1,26 @@
+#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 hardtanh(const at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1);
+TORCH_API at::Tensor & hardtanh_out(const at::Tensor & self, const at::Scalar & min_val, const at::Scalar & max_val, at::Tensor & out);
+TORCH_API at::Tensor & hardtanh_(at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1);
+TORCH_API at::Tensor hardtanh_quantized_cpu(const at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1);
+TORCH_API at::Tensor & hardtanh_out_quantized_cpu(const at::Tensor & self, const at::Scalar & min_val, const at::Scalar & max_val, at::Tensor & out);
+TORCH_API at::Tensor & hardtanh_quantized_cpu_(at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,72 @@
+#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 index_copy_out {
+  using schema = at::Tensor & (const at::Tensor &, int64_t, 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::index_copy")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_copy.out(Tensor self, int dim, Tensor index, Tensor source, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, at::Tensor & out);
+};
+
+struct TORCH_API index_copy_ {
+  using schema = at::Tensor & (at::Tensor &, int64_t, 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::index_copy_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_copy_(Tensor(a!) self, int dim, Tensor index, Tensor source) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source);
+};
+
+struct TORCH_API index_copy {
+  using schema = at::Tensor (const at::Tensor &, int64_t, 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::index_copy")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_copy(Tensor self, int dim, Tensor index, Tensor source) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source);
+};
+
+struct TORCH_API index_copy__dimname {
+  using schema = at::Tensor & (at::Tensor &, at::Dimname, 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::index_copy_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dimname")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_copy_.dimname(Tensor(a!) self, Dimname dim, Tensor index, Tensor source) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & source);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & source);
+};
+
+struct TORCH_API index_copy_dimname {
+  using schema = at::Tensor (const at::Tensor &, at::Dimname, 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::index_copy")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dimname")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_copy.dimname(Tensor self, Dimname dim, Tensor index, Tensor source) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & source);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & source);
+};
+
+}} // namespace at::_ops

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

@@ -0,0 +1,67 @@
+#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/isin_ops.h>
+
+namespace at {
+
+
+// aten::isin.Tensor_Tensor_out(Tensor elements, Tensor test_elements, *, bool assume_unique=False, bool invert=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & isin_out(at::Tensor & out, const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false) {
+    return at::_ops::isin_Tensor_Tensor_out::call(elements, test_elements, assume_unique, invert, out);
+}
+// aten::isin.Tensor_Tensor_out(Tensor elements, Tensor test_elements, *, bool assume_unique=False, bool invert=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & isin_outf(const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out) {
+    return at::_ops::isin_Tensor_Tensor_out::call(elements, test_elements, assume_unique, invert, out);
+}
+
+// aten::isin.Tensor_Tensor(Tensor elements, Tensor test_elements, *, bool assume_unique=False, bool invert=False) -> Tensor
+inline at::Tensor isin(const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false) {
+    return at::_ops::isin_Tensor_Tensor::call(elements, test_elements, assume_unique, invert);
+}
+
+// aten::isin.Tensor_Scalar_out(Tensor elements, Scalar test_element, *, bool assume_unique=False, bool invert=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & isin_out(at::Tensor & out, const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique=false, bool invert=false) {
+    return at::_ops::isin_Tensor_Scalar_out::call(elements, test_element, assume_unique, invert, out);
+}
+// aten::isin.Tensor_Scalar_out(Tensor elements, Scalar test_element, *, bool assume_unique=False, bool invert=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & isin_outf(const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique, bool invert, at::Tensor & out) {
+    return at::_ops::isin_Tensor_Scalar_out::call(elements, test_element, assume_unique, invert, out);
+}
+
+// aten::isin.Tensor_Scalar(Tensor elements, Scalar test_element, *, bool assume_unique=False, bool invert=False) -> Tensor
+inline at::Tensor isin(const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique=false, bool invert=false) {
+    return at::_ops::isin_Tensor_Scalar::call(elements, test_element, assume_unique, invert);
+}
+
+// aten::isin.Scalar_Tensor_out(Scalar element, Tensor test_elements, *, bool assume_unique=False, bool invert=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & isin_out(at::Tensor & out, const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false) {
+    return at::_ops::isin_Scalar_Tensor_out::call(element, test_elements, assume_unique, invert, out);
+}
+// aten::isin.Scalar_Tensor_out(Scalar element, Tensor test_elements, *, bool assume_unique=False, bool invert=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & isin_outf(const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out) {
+    return at::_ops::isin_Scalar_Tensor_out::call(element, test_elements, assume_unique, invert, out);
+}
+
+// aten::isin.Scalar_Tensor(Scalar element, Tensor test_elements, *, bool assume_unique=False, bool invert=False) -> Tensor
+inline at::Tensor isin(const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false) {
+    return at::_ops::isin_Scalar_Tensor::call(element, test_elements, assume_unique, invert);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isin_compositeexplicitautogradnonfunctional_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 compositeexplicitautogradnonfunctional {
+
+TORCH_API at::Tensor isin(const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false);
+TORCH_API at::Tensor isin(const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique=false, bool invert=false);
+TORCH_API at::Tensor isin(const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isposinf_meta_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 meta {
+
+TORCH_API at::Tensor isposinf(const at::Tensor & self);
+TORCH_API at::Tensor & isposinf_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & isposinf_outf(const at::Tensor & self, at::Tensor & out);
+
+} // namespace meta
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/istft_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 istft(const at::Tensor & self, int64_t n_fft, c10::optional<int64_t> hop_length=c10::nullopt, c10::optional<int64_t> win_length=c10::nullopt, const c10::optional<at::Tensor> & window={}, bool center=true, bool normalized=false, c10::optional<bool> onesided=c10::nullopt, c10::optional<int64_t> length=c10::nullopt, bool return_complex=false);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lcm_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 lcm(const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & lcm_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & lcm_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out);
+TORCH_API at::Tensor & lcm_(at::Tensor & self, const at::Tensor & other);
+
+} // namespace meta
+} // namespace at