Add files using upload-large-folder tool

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convert_indices_from_csr_to_coo_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>
+#include <ATen/ops/_convert_indices_from_csr_to_coo_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured__convert_indices_from_csr_to_coo_structured_cpu : public at::meta::structured__convert_indices_from_csr_to_coo {
+void impl(const at::Tensor & crow_indices, const at::Tensor & col_indices, bool out_int32, bool transpose, const at::Tensor & out);
+};
+struct TORCH_API structured__convert_indices_from_csr_to_coo_structured_cuda : public at::meta::structured__convert_indices_from_csr_to_coo {
+void impl(const at::Tensor & crow_indices, const at::Tensor & col_indices, bool out_int32, bool transpose, const at::Tensor & out);
+};
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convert_weight_to_int4pack_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 _convert_weight_to_int4pack_cpu(const at::Tensor & self, int64_t innerKTiles);
+TORCH_API at::Tensor _convert_weight_to_int4pack_cuda(const at::Tensor & self, int64_t innerKTiles);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cummax_helper_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 void _cummax_helper(const at::Tensor & self, at::Tensor & values, at::Tensor & indices, int64_t dim);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fake_quantize_learnable_per_channel_affine_backward_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 ::std::tuple<at::Tensor,at::Tensor,at::Tensor> _fake_quantize_learnable_per_channel_affine_backward(const at::Tensor & grad, const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis, int64_t quant_min, int64_t quant_max, double grad_factor=1.0);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sqrt_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 void _foreach_sqrt_out(at::TensorList out, at::TensorList self);
+TORCH_API void _foreach_sqrt_outf(at::TensorList self, at::TensorList out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_grid_sampler_2d_cpu_fallback_backward.h

@@ -0,0 +1,30 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/_grid_sampler_2d_cpu_fallback_backward_ops.h>
+
+namespace at {
+
+
+// aten::_grid_sampler_2d_cpu_fallback_backward(Tensor grad_output, Tensor input, Tensor grid, int interpolation_mode, int padding_mode, bool align_corners) -> (Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor> _grid_sampler_2d_cpu_fallback_backward(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners) {
+    return at::_ops::_grid_sampler_2d_cpu_fallback_backward::call(grad_output, input, grid, interpolation_mode, padding_mode, align_corners);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_check_errors.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/_linalg_check_errors_ops.h>
+
+namespace at {
+
+
+// aten::_linalg_check_errors(Tensor info, str api_name, *, bool is_matrix) -> ()
+inline void _linalg_check_errors(const at::Tensor & info, c10::string_view api_name, bool is_matrix) {
+    return at::_ops::_linalg_check_errors::call(info, api_name, is_matrix);
+}
+
+}

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

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

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

@@ -0,0 +1,20 @@
+#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 {
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_from_mask_left_aligned_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 bool _nested_tensor_from_mask_left_aligned(const at::Tensor & t, const at::Tensor & mask);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_pdist_forward.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/_pdist_forward_ops.h>
+
+namespace at {
+
+
+// aten::_pdist_forward(Tensor self, float p=2) -> Tensor
+inline at::Tensor _pdist_forward(const at::Tensor & self, double p=2) {
+    return at::_ops::_pdist_forward::call(self, p);
+}
+
+// aten::_pdist_forward.out(Tensor self, float p=2, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _pdist_forward_out(at::Tensor & out, const at::Tensor & self, double p=2) {
+    return at::_ops::_pdist_forward_out::call(self, p, out);
+}
+// aten::_pdist_forward.out(Tensor self, float p=2, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _pdist_forward_outf(const at::Tensor & self, double p, at::Tensor & out) {
+    return at::_ops::_pdist_forward_out::call(self, p, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_reshape_alias_copy_compositeexplicitautograd_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 compositeexplicitautograd {
+
+TORCH_API at::Tensor & _reshape_alias_copy_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef size, at::IntArrayRef stride);
+TORCH_API at::Tensor & _reshape_alias_copy_outf(const at::Tensor & self, at::IntArrayRef size, at::IntArrayRef stride, at::Tensor & out);
+TORCH_API at::Tensor & _reshape_alias_copy_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef size, c10::SymIntArrayRef stride);
+TORCH_API at::Tensor & _reshape_alias_copy_symint_outf(const at::Tensor & self, c10::SymIntArrayRef size, c10::SymIntArrayRef stride, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_bsc_tensor_unsafe_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 _sparse_bsc_tensor_unsafe {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &, at::IntArrayRef, c10::optional<at::ScalarType>, c10::optional<at::Layout>, c10::optional<at::Device>, c10::optional<bool>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_sparse_bsc_tensor_unsafe")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_sparse_bsc_tensor_unsafe(Tensor ccol_indices, Tensor row_indices, Tensor values, int[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & ccol_indices, const at::Tensor & row_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);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & ccol_indices, const at::Tensor & row_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);
+};
+
+}} // namespace at::_ops

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

@@ -0,0 +1,24 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor _stack(at::TensorList tensors, int64_t dim=0);
+TORCH_API at::Tensor & _stack_out(at::TensorList tensors, int64_t dim, at::Tensor & out);
+TORCH_API at::Tensor _stack_cpu(at::TensorList tensors, int64_t dim=0);
+TORCH_API at::Tensor & _stack_out_cpu(at::TensorList tensors, int64_t dim, at::Tensor & out);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_optional_filled_intlist_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 _test_optional_filled_intlist(const at::Tensor & values, at::OptionalIntArrayRef addends);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_sparse_semi_structured.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/_to_sparse_semi_structured_ops.h>
+
+namespace at {
+
+
+// aten::_to_sparse_semi_structured(Tensor dense) -> (Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor> _to_sparse_semi_structured(const at::Tensor & dense) {
+    return at::_ops::_to_sparse_semi_structured::call(dense);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique2_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 _unique2 {
+  using schema = ::std::tuple<at::Tensor,at::Tensor,at::Tensor> (const at::Tensor &, 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::_unique2")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_unique2(Tensor self, bool sorted=True, bool return_inverse=False, bool return_counts=False) -> (Tensor, Tensor, Tensor)")
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor> call(const at::Tensor & self, bool sorted, bool return_inverse, bool return_counts);
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, bool sorted, bool return_inverse, bool return_counts);
+};
+
+struct TORCH_API _unique2_out {
+  using schema = ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> (const at::Tensor &, bool, bool, bool, 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::_unique2")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_unique2.out(Tensor self, bool sorted=True, bool return_inverse=False, bool return_counts=False, *, 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 & self, bool sorted, bool return_inverse, bool return_counts, 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 & self, bool sorted, bool return_inverse, bool return_counts, 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/_validate_sparse_bsr_tensor_args.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/_validate_sparse_bsr_tensor_args_ops.h>
+
+namespace at {
+
+
+// aten::_validate_sparse_bsr_tensor_args(Tensor crow_indices, Tensor col_indices, Tensor values, int[] size) -> ()
+inline void _validate_sparse_bsr_tensor_args(const at::Tensor & crow_indices, const at::Tensor & col_indices, const at::Tensor & values, at::IntArrayRef size) {
+    return at::_ops::_validate_sparse_bsr_tensor_args::call(crow_indices, col_indices, values, size);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_validate_sparse_csc_tensor_args_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 void _validate_sparse_csc_tensor_args(const at::Tensor & ccol_indices, const at::Tensor & row_indices, const at::Tensor & values, at::IntArrayRef size);
+} // namespace native
+} // namespace at

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool3d_cuda_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 cuda {
+
+TORCH_API at::Tensor & adaptive_avg_pool3d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size);
+TORCH_API at::Tensor & adaptive_avg_pool3d_outf(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out);
+TORCH_API at::Tensor & adaptive_avg_pool3d_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size);
+TORCH_API at::Tensor & adaptive_avg_pool3d_symint_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, at::Tensor & out);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/affine_grid_generator_compositeexplicitautograd_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 compositeexplicitautograd {
+
+TORCH_API at::Tensor affine_grid_generator(const at::Tensor & theta, at::IntArrayRef size, bool align_corners);
+TORCH_API at::Tensor affine_grid_generator_symint(const at::Tensor & theta, c10::SymIntArrayRef size, bool align_corners);
+TORCH_API at::Tensor & affine_grid_generator_out(at::Tensor & out, const at::Tensor & theta, at::IntArrayRef size, bool align_corners);
+TORCH_API at::Tensor & affine_grid_generator_outf(const at::Tensor & theta, at::IntArrayRef size, bool align_corners, at::Tensor & out);
+TORCH_API at::Tensor & affine_grid_generator_symint_out(at::Tensor & out, const at::Tensor & theta, c10::SymIntArrayRef size, bool align_corners);
+TORCH_API at::Tensor & affine_grid_generator_symint_outf(const at::Tensor & theta, c10::SymIntArrayRef size, bool align_corners, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/and_compositeimplicitautograd_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 compositeimplicitautograd {
+
+TORCH_API at::Tensor __and__(const at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor & __iand__(at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor __and__(const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & __iand__(at::Tensor & self, const at::Tensor & other);
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor arccos(const at::Tensor & self);
+TORCH_API at::Tensor & arccos_out(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & arccos_(at::Tensor & self);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool2d_backward_compositeexplicitautogradnonfunctional_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeexplicitautogradnonfunctional {
+
+TORCH_API at::Tensor avg_pool2d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional<int64_t> divisor_override);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_cpu_dispatch.h

@@ -0,0 +1,25 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API at::Tensor avg_pool3d(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, bool ceil_mode=false, bool count_include_pad=true, c10::optional<int64_t> divisor_override=c10::nullopt);
+TORCH_API at::Tensor & avg_pool3d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, bool ceil_mode=false, bool count_include_pad=true, c10::optional<int64_t> divisor_override=c10::nullopt);
+TORCH_API at::Tensor & avg_pool3d_outf(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional<int64_t> divisor_override, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bernoulli_compositeexplicitautograd_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 compositeexplicitautograd {
+
+TORCH_API at::Tensor bernoulli(const at::Tensor & self, c10::optional<at::Generator> generator=c10::nullopt);
+TORCH_API at::Tensor bernoulli(const at::Tensor & self, const at::Tensor & p, c10::optional<at::Generator> generator=c10::nullopt);
+TORCH_API at::Tensor & bernoulli_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & p, c10::optional<at::Generator> generator=c10::nullopt);
+TORCH_API at::Tensor & bernoulli_outf(const at::Tensor & self, const at::Tensor & p, c10::optional<at::Generator> generator, at::Tensor & out);
+TORCH_API at::Tensor & bernoulli_out(at::Tensor & out, const at::Tensor & self, double p=0.5, c10::optional<at::Generator> generator=c10::nullopt);
+TORCH_API at::Tensor & bernoulli_outf(const at::Tensor & self, double p, c10::optional<at::Generator> generator, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_left_shift_cuda_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 cuda {
+
+TORCH_API at::Tensor bitwise_left_shift(const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & bitwise_left_shift_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & bitwise_left_shift_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out);
+TORCH_API at::Tensor & bitwise_left_shift_(at::Tensor & self, const at::Tensor & other);
+
+} // namespace cuda
+} // namespace at

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/col2im.h

@@ -0,0 +1,91 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/col2im_ops.h>
+
+namespace at {
+
+
+// aten::col2im.out(Tensor self, SymInt[2] output_size, int[2] kernel_size, int[2] dilation, int[2] padding, int[2] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & col2im_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::col2im_out::call(self, c10::fromIntArrayRefSlow(output_size), kernel_size, dilation, padding, stride, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & col2im_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::col2im_out::call(self, c10::fromIntArrayRefSlow(output_size), kernel_size, dilation, padding, stride, out);
+  }
+}
+
+// aten::col2im.out(Tensor self, SymInt[2] output_size, int[2] kernel_size, int[2] dilation, int[2] padding, int[2] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & col2im_outf(const at::Tensor & self, at::IntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride, at::Tensor & out) {
+    return at::_ops::col2im_out::call(self, c10::fromIntArrayRefSlow(output_size), kernel_size, dilation, padding, stride, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & col2im_outf(const at::Tensor & self, at::IntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride, at::Tensor & out) {
+    return at::_ops::col2im_out::call(self, c10::fromIntArrayRefSlow(output_size), kernel_size, dilation, padding, stride, out);
+  }
+}
+
+// aten::col2im.out(Tensor self, SymInt[2] output_size, int[2] kernel_size, int[2] dilation, int[2] padding, int[2] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & col2im_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::col2im_out::call(self, output_size, kernel_size, dilation, padding, stride, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & col2im_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::col2im_out::call(self, output_size, kernel_size, dilation, padding, stride, out);
+  }
+}
+
+// aten::col2im.out(Tensor self, SymInt[2] output_size, int[2] kernel_size, int[2] dilation, int[2] padding, int[2] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & col2im_symint_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride, at::Tensor & out) {
+    return at::_ops::col2im_out::call(self, output_size, kernel_size, dilation, padding, stride, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & col2im_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride, at::Tensor & out) {
+    return at::_ops::col2im_out::call(self, output_size, kernel_size, dilation, padding, stride, out);
+  }
+}
+
+// aten::col2im(Tensor self, SymInt[2] output_size, int[2] kernel_size, int[2] dilation, int[2] padding, int[2] stride) -> Tensor
+inline at::Tensor col2im(const at::Tensor & self, at::IntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::col2im::call(self, c10::fromIntArrayRefSlow(output_size), kernel_size, dilation, padding, stride);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor col2im(const at::Tensor & self, at::IntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::col2im::call(self, c10::fromIntArrayRefSlow(output_size), kernel_size, dilation, padding, stride);
+  }
+}
+
+// aten::col2im(Tensor self, SymInt[2] output_size, int[2] kernel_size, int[2] dilation, int[2] padding, int[2] stride) -> Tensor
+inline at::Tensor col2im_symint(const at::Tensor & self, c10::SymIntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::col2im::call(self, output_size, kernel_size, dilation, padding, stride);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor col2im(const at::Tensor & self, c10::SymIntArrayRef output_size, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::col2im::call(self, output_size, kernel_size, dilation, padding, stride);
+  }
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/conv_depthwise3d_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 at::Tensor conv_depthwise3d(const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef kernel_size, const c10::optional<at::Tensor> & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation);
+TORCH_API at::Tensor conv_depthwise3d_symint(const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, const c10::optional<at::Tensor> & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cosh_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 cosh(const at::Tensor & self);
+TORCH_API at::Tensor & cosh_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & cosh_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & cosh_(at::Tensor & self);
+
+} // namespace meta
+} // namespace at

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

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

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+#include <ATen/ops/elu_backward_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_elu_backward_out : public at::meta::structured_elu_backward {
+void impl(const at::Tensor & grad_output, const at::Scalar & alpha, const at::Scalar & scale, const at::Scalar & input_scale, bool is_result, const at::Tensor & self_or_result, const at::Tensor & grad_input);
+};
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,47 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+
+
+
+#include <ATen/ops/embedding_backward_ops.h>
+
+namespace at {
+
+
+// aten::embedding_backward(Tensor grad, Tensor indices, SymInt num_weights, SymInt padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor
+inline at::Tensor embedding_backward(const at::Tensor & grad, const at::Tensor & indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) {
+    return at::_ops::embedding_backward::call(grad, indices, num_weights, padding_idx, scale_grad_by_freq, sparse);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor embedding_backward(const at::Tensor & grad, const at::Tensor & indices, int64_t num_weights, int64_t padding_idx, bool scale_grad_by_freq, bool sparse) {
+    return at::_ops::embedding_backward::call(grad, indices, num_weights, padding_idx, scale_grad_by_freq, sparse);
+  }
+}
+
+// aten::embedding_backward(Tensor grad, Tensor indices, SymInt num_weights, SymInt padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor
+inline at::Tensor embedding_backward_symint(const at::Tensor & grad, const at::Tensor & indices, c10::SymInt num_weights, c10::SymInt padding_idx, bool scale_grad_by_freq, bool sparse) {
+    return at::_ops::embedding_backward::call(grad, indices, num_weights, padding_idx, scale_grad_by_freq, sparse);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor embedding_backward(const at::Tensor & grad, const at::Tensor & indices, c10::SymInt num_weights, c10::SymInt padding_idx, bool scale_grad_by_freq, bool sparse) {
+    return at::_ops::embedding_backward::call(grad, indices, num_weights, padding_idx, scale_grad_by_freq, sparse);
+  }
+}
+
+}

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

@@ -0,0 +1,43 @@
+#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_quantized_ops.h>
+
+namespace at {
+
+
+// aten::empty_quantized(int[] size, Tensor qtensor, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor
+inline at::Tensor empty_quantized(at::IntArrayRef size, const at::Tensor & qtensor, at::TensorOptions options={}, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_quantized::call(size, qtensor, 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_quantized(int[] size, Tensor qtensor, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor
+inline at::Tensor empty_quantized(at::IntArrayRef size, const at::Tensor & qtensor, 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_quantized::call(size, qtensor, dtype, layout, device, pin_memory, memory_format);
+}
+
+// aten::empty_quantized.out(int[] size, Tensor qtensor, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_quantized_out(at::Tensor & out, at::IntArrayRef size, const at::Tensor & qtensor, c10::optional<at::MemoryFormat> memory_format=c10::nullopt) {
+    return at::_ops::empty_quantized_out::call(size, qtensor, memory_format, out);
+}
+// aten::empty_quantized.out(int[] size, Tensor qtensor, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_quantized_outf(at::IntArrayRef size, const at::Tensor & qtensor, c10::optional<at::MemoryFormat> memory_format, at::Tensor & out) {
+    return at::_ops::empty_quantized_out::call(size, qtensor, memory_format, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/exp2_cuda_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 cuda {
+
+TORCH_API at::Tensor exp2(const at::Tensor & self);
+TORCH_API at::Tensor & exp2_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & exp2_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & exp2_(at::Tensor & self);
+
+} // namespace cuda
+} // namespace at

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_2d_backward_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 ::std::tuple<at::Tensor,at::Tensor> grid_sampler_2d_backward(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, ::std::array<bool,2> output_mask);
+
+} // namespace cpu
+} // namespace at

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardsigmoid_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 hardsigmoid(const at::Tensor & self);
+TORCH_API at::Tensor & hardsigmoid_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & hardsigmoid_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & hardsigmoid_(at::Tensor & self);
+
+} // namespace meta
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_add_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 index_add(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha=1);
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

@@ -0,0 +1,50 @@
+#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_index_Tensor : public TensorIteratorBase {
+    
+                template <bool SIZES = false, bool STRIDES = false>
+                struct TORCH_API precompute_out {
+                    
+                    precompute_out<true, STRIDES> set_sizes(at::DimVector value) {
+                        static_assert(SIZES == false, "sizes already set");
+                        precompute_out<true, STRIDES> ret;
+ret.sizes = value;
+ret.strides = this->strides;
+return ret;
+                    }
+                
+
+                    precompute_out<SIZES, true> set_strides(at::DimVector value) {
+                        static_assert(STRIDES == false, "strides already set");
+                        precompute_out<SIZES, true> ret;
+ret.sizes = this->sizes;
+ret.strides = value;
+return ret;
+                    }
+                
+                    at::DimVector sizes;
+at::DimVector strides;
+            };
+    using meta_return_ty = precompute_out <true, true>;
+    meta_return_ty meta(const at::Tensor & self, at::IOptTensorListRef indices);
+};
+
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor index_put(const at::Tensor & self, const c10::List<c10::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate=false);
+TORCH_API at::Tensor & index_put_out(const at::Tensor & self, const c10::List<c10::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate, at::Tensor & out);
+TORCH_API at::Tensor & index_put_(at::Tensor & self, const c10::List<c10::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate=false);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,32 @@
+#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_lerp_Scalar : public TensorIteratorBase {
+    
+    
+    void meta(const at::Tensor & self, const at::Tensor & end, const at::Scalar & weight);
+};
+struct TORCH_API structured_lerp_Tensor : public TensorIteratorBase {
+    
+    
+    void meta(const at::Tensor & self, const at::Tensor & end, const at::Tensor & weight);
+};
+
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,24 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor linalg_matrix_norm(const at::Tensor & self, const at::Scalar & ord, at::IntArrayRef dim={-2,-1}, bool keepdim=false, c10::optional<at::ScalarType> dtype=c10::nullopt);
+TORCH_API at::Tensor & linalg_matrix_norm_out(const at::Tensor & self, const at::Scalar & ord, at::IntArrayRef dim, bool keepdim, c10::optional<at::ScalarType> dtype, at::Tensor & out);
+TORCH_API at::Tensor linalg_matrix_norm(const at::Tensor & self, c10::string_view ord="fro", at::IntArrayRef dim={-2,-1}, bool keepdim=false, c10::optional<at::ScalarType> dtype=c10::nullopt);
+TORCH_API at::Tensor & linalg_matrix_norm_out(const at::Tensor & self, c10::string_view ord, at::IntArrayRef dim, bool keepdim, c10::optional<at::ScalarType> dtype, at::Tensor & out);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,81 @@
+#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/linalg_pinv_ops.h>
+
+namespace at {
+
+
+// aten::linalg_pinv.atol_rtol_tensor(Tensor self, *, Tensor? atol=None, Tensor? rtol=None, bool hermitian=False) -> Tensor
+inline at::Tensor linalg_pinv(const at::Tensor & self, const c10::optional<at::Tensor> & atol={}, const c10::optional<at::Tensor> & rtol={}, bool hermitian=false) {
+    return at::_ops::linalg_pinv_atol_rtol_tensor::call(self, atol, rtol, hermitian);
+}
+
+// aten::linalg_pinv.atol_rtol_tensor_out(Tensor self, *, Tensor? atol=None, Tensor? rtol=None, bool hermitian=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & linalg_pinv_out(at::Tensor & out, const at::Tensor & self, const c10::optional<at::Tensor> & atol={}, const c10::optional<at::Tensor> & rtol={}, bool hermitian=false) {
+    return at::_ops::linalg_pinv_atol_rtol_tensor_out::call(self, atol, rtol, hermitian, out);
+}
+// aten::linalg_pinv.atol_rtol_tensor_out(Tensor self, *, Tensor? atol=None, Tensor? rtol=None, bool hermitian=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & linalg_pinv_outf(const at::Tensor & self, const c10::optional<at::Tensor> & atol, const c10::optional<at::Tensor> & rtol, bool hermitian, at::Tensor & out) {
+    return at::_ops::linalg_pinv_atol_rtol_tensor_out::call(self, atol, rtol, hermitian, out);
+}
+
+// aten::linalg_pinv.atol_rtol_float(Tensor self, *, float? atol=None, float? rtol=None, bool hermitian=False) -> Tensor
+inline at::Tensor linalg_pinv(const at::Tensor & self, c10::optional<double> atol, c10::optional<double> rtol, bool hermitian=false) {
+    return at::_ops::linalg_pinv_atol_rtol_float::call(self, atol, rtol, hermitian);
+}
+
+// aten::linalg_pinv.atol_rtol_float_out(Tensor self, *, float? atol=None, float? rtol=None, bool hermitian=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & linalg_pinv_out(at::Tensor & out, const at::Tensor & self, c10::optional<double> atol, c10::optional<double> rtol, bool hermitian=false) {
+    return at::_ops::linalg_pinv_atol_rtol_float_out::call(self, atol, rtol, hermitian, out);
+}
+// aten::linalg_pinv.atol_rtol_float_out(Tensor self, *, float? atol=None, float? rtol=None, bool hermitian=False, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & linalg_pinv_outf(const at::Tensor & self, c10::optional<double> atol, c10::optional<double> rtol, bool hermitian, at::Tensor & out) {
+    return at::_ops::linalg_pinv_atol_rtol_float_out::call(self, atol, rtol, hermitian, out);
+}
+
+// aten::linalg_pinv(Tensor self, float rcond, bool hermitian=False) -> Tensor
+inline at::Tensor linalg_pinv(const at::Tensor & self, double rcond, bool hermitian=false) {
+    return at::_ops::linalg_pinv::call(self, rcond, hermitian);
+}
+
+// aten::linalg_pinv.rcond_tensor(Tensor self, Tensor rcond, bool hermitian=False) -> Tensor
+inline at::Tensor linalg_pinv(const at::Tensor & self, const at::Tensor & rcond, bool hermitian=false) {
+    return at::_ops::linalg_pinv_rcond_tensor::call(self, rcond, hermitian);
+}
+
+// aten::linalg_pinv.out(Tensor self, float rcond, bool hermitian=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & linalg_pinv_out(at::Tensor & out, const at::Tensor & self, double rcond, bool hermitian=false) {
+    return at::_ops::linalg_pinv_out::call(self, rcond, hermitian, out);
+}
+// aten::linalg_pinv.out(Tensor self, float rcond, bool hermitian=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & linalg_pinv_outf(const at::Tensor & self, double rcond, bool hermitian, at::Tensor & out) {
+    return at::_ops::linalg_pinv_out::call(self, rcond, hermitian, out);
+}
+
+// aten::linalg_pinv.out_rcond_tensor(Tensor self, Tensor rcond, bool hermitian=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & linalg_pinv_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & rcond, bool hermitian=false) {
+    return at::_ops::linalg_pinv_out_rcond_tensor::call(self, rcond, hermitian, out);
+}
+// aten::linalg_pinv.out_rcond_tensor(Tensor self, Tensor rcond, bool hermitian=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & linalg_pinv_outf(const at::Tensor & self, const at::Tensor & rcond, bool hermitian, at::Tensor & out) {
+    return at::_ops::linalg_pinv_out_rcond_tensor::call(self, rcond, hermitian, out);
+}
+
+}

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

@@ -0,0 +1,81 @@
+#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/max_ops.h>
+
+namespace at {
+
+
+// aten::max.dim(Tensor self, int dim, bool keepdim=False) -> (Tensor values, Tensor indices)
+inline ::std::tuple<at::Tensor,at::Tensor> max(const at::Tensor & self, int64_t dim, bool keepdim=false) {
+    return at::_ops::max_dim::call(self, dim, keepdim);
+}
+
+// aten::max.dim_max(Tensor self, int dim, bool keepdim=False, *, Tensor(a!) max, Tensor(b!) max_values) -> (Tensor(a!) values, Tensor(b!) indices)
+inline ::std::tuple<at::Tensor &,at::Tensor &> max_out(at::Tensor & max, at::Tensor & max_values, const at::Tensor & self, int64_t dim, bool keepdim=false) {
+    return at::_ops::max_dim_max::call(self, dim, keepdim, max, max_values);
+}
+// aten::max.dim_max(Tensor self, int dim, bool keepdim=False, *, Tensor(a!) max, Tensor(b!) max_values) -> (Tensor(a!) values, Tensor(b!) indices)
+inline ::std::tuple<at::Tensor &,at::Tensor &> max_outf(const at::Tensor & self, int64_t dim, bool keepdim, at::Tensor & max, at::Tensor & max_values) {
+    return at::_ops::max_dim_max::call(self, dim, keepdim, max, max_values);
+}
+
+// aten::max.names_dim(Tensor self, Dimname dim, bool keepdim=False) -> (Tensor values, Tensor indices)
+inline ::std::tuple<at::Tensor,at::Tensor> max(const at::Tensor & self, at::Dimname dim, bool keepdim=false) {
+    return at::_ops::max_names_dim::call(self, dim, keepdim);
+}
+
+// aten::max.names_dim_max(Tensor self, Dimname dim, bool keepdim=False, *, Tensor(a!) max, Tensor(b!) max_values) -> (Tensor(a!) values, Tensor(b!) indices)
+inline ::std::tuple<at::Tensor &,at::Tensor &> max_out(at::Tensor & max, at::Tensor & max_values, const at::Tensor & self, at::Dimname dim, bool keepdim=false) {
+    return at::_ops::max_names_dim_max::call(self, dim, keepdim, max, max_values);
+}
+// aten::max.names_dim_max(Tensor self, Dimname dim, bool keepdim=False, *, Tensor(a!) max, Tensor(b!) max_values) -> (Tensor(a!) values, Tensor(b!) indices)
+inline ::std::tuple<at::Tensor &,at::Tensor &> max_outf(const at::Tensor & self, at::Dimname dim, bool keepdim, at::Tensor & max, at::Tensor & max_values) {
+    return at::_ops::max_names_dim_max::call(self, dim, keepdim, max, max_values);
+}
+
+// aten::max(Tensor self) -> Tensor
+inline at::Tensor max(const at::Tensor & self) {
+    return at::_ops::max::call(self);
+}
+
+// aten::max.other(Tensor self, Tensor other) -> Tensor
+inline at::Tensor max(const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::max_other::call(self, other);
+}
+
+// aten::max.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & max_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::max_out::call(self, other, out);
+}
+// aten::max.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & max_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) {
+    return at::_ops::max_out::call(self, other, out);
+}
+
+// aten::max.unary_out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & max_out(at::Tensor & out, const at::Tensor & self) {
+    return at::_ops::max_unary_out::call(self, out);
+}
+// aten::max.unary_out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & max_outf(const at::Tensor & self, at::Tensor & out) {
+    return at::_ops::max_unary_out::call(self, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool1d_with_indices_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 ::std::tuple<at::Tensor,at::Tensor> max_pool1d_with_indices(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, at::IntArrayRef dilation=1, bool ceil_mode=false);
+
+} // namespace compositeimplicitautograd
+} // namespace at