Add files using upload-large-folder tool

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cholesky_solve_helper.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/_cholesky_solve_helper_ops.h>
+
+namespace at {
+
+
+// aten::_cholesky_solve_helper(Tensor self, Tensor A, bool upper) -> Tensor
+inline at::Tensor _cholesky_solve_helper(const at::Tensor & self, const at::Tensor & A, bool upper) {
+    return at::_ops::_cholesky_solve_helper::call(self, A, upper);
+}
+
+// aten::_cholesky_solve_helper.out(Tensor self, Tensor A, bool upper, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _cholesky_solve_helper_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & A, bool upper) {
+    return at::_ops::_cholesky_solve_helper_out::call(self, A, upper, out);
+}
+// aten::_cholesky_solve_helper.out(Tensor self, Tensor A, bool upper, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _cholesky_solve_helper_outf(const at::Tensor & self, const at::Tensor & A, bool upper, at::Tensor & out) {
+    return at::_ops::_cholesky_solve_helper_out::call(self, A, upper, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_embedding_bag_forward_only_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 _embedding_bag_forward_only {
+  using schema = ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor> (const at::Tensor &, const at::Tensor &, const at::Tensor &, bool, int64_t, bool, const c10::optional<at::Tensor> &, bool, int64_t);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_embedding_bag_forward_only")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_embedding_bag_forward_only(Tensor weight, Tensor indices, Tensor offsets, bool scale_grad_by_freq=False, int mode=0, bool sparse=False, Tensor? per_sample_weights=None, bool include_last_offset=False, int padding_idx=-1) -> (Tensor, Tensor, Tensor, Tensor)")
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor> call(const at::Tensor & weight, const at::Tensor & indices, const at::Tensor & offsets, bool scale_grad_by_freq, int64_t mode, bool sparse, const c10::optional<at::Tensor> & per_sample_weights, bool include_last_offset, int64_t padding_idx);
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & weight, const at::Tensor & indices, const at::Tensor & offsets, bool scale_grad_by_freq, int64_t mode, bool sparse, const c10::optional<at::Tensor> & per_sample_weights, bool include_last_offset, int64_t padding_idx);
+};
+
+struct TORCH_API _embedding_bag_forward_only_out {
+  using schema = ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &,at::Tensor &> (const at::Tensor &, const at::Tensor &, const at::Tensor &, bool, int64_t, bool, const c10::optional<at::Tensor> &, bool, int64_t, at::Tensor &, 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::_embedding_bag_forward_only")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_embedding_bag_forward_only.out(Tensor weight, Tensor indices, Tensor offsets, bool scale_grad_by_freq=False, int mode=0, bool sparse=False, Tensor? per_sample_weights=None, bool include_last_offset=False, int padding_idx=-1, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2, Tensor(d!) out3) -> (Tensor(a!), Tensor(b!), Tensor(c!), Tensor(d!))")
+  static ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &,at::Tensor &> call(const at::Tensor & weight, const at::Tensor & indices, const at::Tensor & offsets, bool scale_grad_by_freq, int64_t mode, bool sparse, const c10::optional<at::Tensor> & per_sample_weights, bool include_last_offset, int64_t padding_idx, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3);
+  static ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &,at::Tensor &> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & weight, const at::Tensor & indices, const at::Tensor & offsets, bool scale_grad_by_freq, int64_t mode, bool sparse, const c10::optional<at::Tensor> & per_sample_weights, bool include_last_offset, int64_t padding_idx, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_acos_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_acos(at::TensorList self);
+TORCH_API void _foreach_acos_(at::TensorList self);
+
+} // namespace cuda
+} // namespace at

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log10_cpu_dispatch.h

@@ -0,0 +1,24 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API ::std::vector<at::Tensor> _foreach_log10(at::TensorList self);
+TORCH_API void _foreach_log10_(at::TensorList self);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_histogramdd_bin_edges_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 _histogramdd_bin_edges_out(at::TensorList out, const at::Tensor & self, at::IntArrayRef bins, c10::optional<at::ArrayRef<double>> range=c10::nullopt, const c10::optional<at::Tensor> & weight={}, bool density=false);
+TORCH_API void _histogramdd_bin_edges_outf(const at::Tensor & self, at::IntArrayRef bins, c10::optional<at::ArrayRef<double>> range, const c10::optional<at::Tensor> & weight, bool density, at::TensorList out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_histogramdd_from_bin_tensors_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 _histogramdd_from_bin_tensors(const at::Tensor & self, at::TensorList bins, const c10::optional<at::Tensor> & weight={}, bool density=false);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_from_mask_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 _nested_tensor_from_mask {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_nested_tensor_from_mask")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_tensor_from_mask(Tensor t, Tensor mask, bool mask_check=True) -> Tensor")
+  static at::Tensor call(const at::Tensor & t, const at::Tensor & mask, bool mask_check);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & t, const at::Tensor & mask, bool mask_check);
+};
+
+struct TORCH_API _nested_tensor_from_mask_out {
+  using schema = at::Tensor & (const 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::_nested_tensor_from_mask")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_tensor_from_mask.out(Tensor t, Tensor mask, bool mask_check=True, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & t, const at::Tensor & mask, bool mask_check, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & t, const at::Tensor & mask, bool mask_check, at::Tensor & out);
+};
+
+}} // namespace at::_ops

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_pack_padded_sequence_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 ::std::tuple<at::Tensor,at::Tensor> _pack_padded_sequence(const at::Tensor & input, const at::Tensor & lengths, bool batch_first);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> _pack_padded_sequence_out(at::Tensor & out0, at::Tensor & out1, const at::Tensor & input, const at::Tensor & lengths, bool batch_first);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> _pack_padded_sequence_outf(const at::Tensor & input, const at::Tensor & lengths, bool batch_first, at::Tensor & out0, at::Tensor & out1);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_pad_packed_sequence_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> _pad_packed_sequence(const at::Tensor & data, const at::Tensor & batch_sizes, bool batch_first, const at::Scalar & padding_value, int64_t total_length);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_mm.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/_scaled_mm_ops.h>
+
+namespace at {
+
+
+// aten::_scaled_mm(Tensor self, Tensor mat2, *, Tensor? bias=None, ScalarType? out_dtype=None, Tensor? scale_a=None, Tensor? scale_b=None, Tensor? scale_result=None, bool use_fast_accum=False) -> (Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor> _scaled_mm(const at::Tensor & self, const at::Tensor & mat2, const c10::optional<at::Tensor> & bias={}, c10::optional<at::ScalarType> out_dtype=c10::nullopt, const c10::optional<at::Tensor> & scale_a={}, const c10::optional<at::Tensor> & scale_b={}, const c10::optional<at::Tensor> & scale_result={}, bool use_fast_accum=false) {
+    return at::_ops::_scaled_mm::call(self, mat2, bias, out_dtype, scale_a, scale_b, scale_result, use_fast_accum);
+}
+
+// aten::_scaled_mm.out(Tensor self, Tensor mat2, *, Tensor? bias=None, ScalarType? out_dtype=None, Tensor? scale_a=None, Tensor? scale_b=None, Tensor? scale_result=None, bool use_fast_accum=False, Tensor(a!) out, Tensor(b!) out_amax) -> (Tensor(a!), Tensor(b!))
+inline ::std::tuple<at::Tensor &,at::Tensor &> _scaled_mm_out(at::Tensor & out, at::Tensor & out_amax, const at::Tensor & self, const at::Tensor & mat2, const c10::optional<at::Tensor> & bias={}, c10::optional<at::ScalarType> out_dtype=c10::nullopt, const c10::optional<at::Tensor> & scale_a={}, const c10::optional<at::Tensor> & scale_b={}, const c10::optional<at::Tensor> & scale_result={}, bool use_fast_accum=false) {
+    return at::_ops::_scaled_mm_out::call(self, mat2, bias, out_dtype, scale_a, scale_b, scale_result, use_fast_accum, out, out_amax);
+}
+// aten::_scaled_mm.out(Tensor self, Tensor mat2, *, Tensor? bias=None, ScalarType? out_dtype=None, Tensor? scale_a=None, Tensor? scale_b=None, Tensor? scale_result=None, bool use_fast_accum=False, Tensor(a!) out, Tensor(b!) out_amax) -> (Tensor(a!), Tensor(b!))
+inline ::std::tuple<at::Tensor &,at::Tensor &> _scaled_mm_outf(const at::Tensor & self, const at::Tensor & mat2, const c10::optional<at::Tensor> & bias, c10::optional<at::ScalarType> out_dtype, const c10::optional<at::Tensor> & scale_a, const c10::optional<at::Tensor> & scale_b, const c10::optional<at::Tensor> & scale_result, bool use_fast_accum, at::Tensor & out, at::Tensor & out_amax) {
+    return at::_ops::_scaled_mm_out::call(self, mat2, bias, out_dtype, scale_a, scale_b, scale_result, use_fast_accum, out, out_amax);
+}
+
+}

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

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API _segment_reduce_backward {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &, c10::string_view, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, int64_t, const c10::optional<at::Scalar> &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_segment_reduce_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_segment_reduce_backward(Tensor grad, Tensor output, Tensor data, str reduce, *, Tensor? lengths=None, Tensor? offsets=None, int axis=0, Scalar? initial=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & grad, const at::Tensor & output, const at::Tensor & data, c10::string_view reduce, const c10::optional<at::Tensor> & lengths, const c10::optional<at::Tensor> & offsets, int64_t axis, const c10::optional<at::Scalar> & initial);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad, const at::Tensor & output, const at::Tensor & data, c10::string_view reduce, const c10::optional<at::Tensor> & lengths, const c10::optional<at::Tensor> & offsets, int64_t axis, const c10::optional<at::Scalar> & initial);
+};
+
+struct TORCH_API _segment_reduce_backward_out {
+  using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const at::Tensor &, c10::string_view, const c10::optional<at::Tensor> &, const c10::optional<at::Tensor> &, int64_t, const c10::optional<at::Scalar> &, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_segment_reduce_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_segment_reduce_backward.out(Tensor grad, Tensor output, Tensor data, str reduce, *, Tensor? lengths=None, Tensor? offsets=None, int axis=0, Scalar? initial=None, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & grad, const at::Tensor & output, const at::Tensor & data, c10::string_view reduce, const c10::optional<at::Tensor> & lengths, const c10::optional<at::Tensor> & offsets, int64_t axis, const c10::optional<at::Scalar> & initial, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad, const at::Tensor & output, const at::Tensor & data, c10::string_view reduce, const c10::optional<at::Tensor> & lengths, const c10::optional<at::Tensor> & offsets, int64_t axis, const c10::optional<at::Scalar> & initial, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_softmax_backward_data_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 & _sparse_softmax_backward_data_out(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor softmax_backward_sparse_cpu(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, const at::Tensor & self);
+TORCH_API at::Tensor softmax_backward_sparse_cuda(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, const at::Tensor & self);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_parallel_materialize.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/_test_parallel_materialize_ops.h>
+
+namespace at {
+
+
+// aten::_test_parallel_materialize(Tensor self, int num_parallel, bool skip_first=False) -> Tensor
+inline at::Tensor _test_parallel_materialize(const at::Tensor & self, int64_t num_parallel, bool skip_first=false) {
+    return at::_ops::_test_parallel_materialize::call(self, num_parallel, skip_first);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_string_default_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 _test_string_default {
+  using schema = at::Tensor (const at::Tensor &, c10::string_view, c10::string_view);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_test_string_default")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_test_string_default(Tensor dummy, str a=\"\\\"'\\\\\", str b='\"\\'\\\\') -> Tensor")
+  static at::Tensor call(const at::Tensor & dummy, c10::string_view a, c10::string_view b);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & dummy, c10::string_view a, c10::string_view b);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_cpu_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::vector<at::Tensor> _to_cpu(at::TensorList tensors);
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_sparse_csc_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 & _to_sparse_csc_out(at::Tensor & out, const at::Tensor & self, c10::optional<int64_t> dense_dim=c10::nullopt);
+TORCH_API at::Tensor & _to_sparse_csc_outf(const at::Tensor & self, c10::optional<int64_t> dense_dim, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_sparse_csc_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 & _to_sparse_csc_out(const at::Tensor & self, c10::optional<int64_t> dense_dim, at::Tensor & out);
+TORCH_API at::Tensor dense_to_sparse_csc(const at::Tensor & self, c10::optional<int64_t> dense_dim=c10::nullopt);
+TORCH_API at::Tensor coo_to_sparse_csc(const at::Tensor & self, c10::optional<int64_t> dense_dim=c10::nullopt);
+TORCH_API at::Tensor sparse_compressed_to_sparse_csc(const at::Tensor & self, c10::optional<int64_t> dense_dim=c10::nullopt);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_transformer_encoder_layer_fwd_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 & _transformer_encoder_layer_fwd_out(at::Tensor & out, const at::Tensor & src, int64_t embed_dim, int64_t num_heads, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, bool use_gelu, bool norm_first, double eps, const at::Tensor & norm_weight_1, const at::Tensor & norm_bias_1, const at::Tensor & norm_weight_2, const at::Tensor & norm_bias_2, const at::Tensor & ffn_weight_1, const at::Tensor & ffn_bias_1, const at::Tensor & ffn_weight_2, const at::Tensor & ffn_bias_2, const c10::optional<at::Tensor> & mask={}, c10::optional<int64_t> mask_type=c10::nullopt);
+TORCH_API at::Tensor & _transformer_encoder_layer_fwd_outf(const at::Tensor & src, int64_t embed_dim, int64_t num_heads, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, bool use_gelu, bool norm_first, double eps, const at::Tensor & norm_weight_1, const at::Tensor & norm_bias_1, const at::Tensor & norm_weight_2, const at::Tensor & norm_bias_2, const at::Tensor & ffn_weight_1, const at::Tensor & ffn_bias_1, const at::Tensor & ffn_weight_2, const at::Tensor & ffn_bias_2, const c10::optional<at::Tensor> & mask, c10::optional<int64_t> mask_type, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_use_cudnn_ctc_loss_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 bool _use_cudnn_ctc_loss(const at::Tensor & log_probs, const at::Tensor & targets, at::IntArrayRef input_lengths, at::IntArrayRef target_lengths, int64_t blank);
+TORCH_API bool _use_cudnn_ctc_loss(const at::Tensor & log_probs, const at::Tensor & targets, const at::Tensor & input_lengths, const at::Tensor & target_lengths, int64_t blank);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/angle_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 angle(const at::Tensor & self);
+TORCH_API at::Tensor & angle_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & angle_outf(const at::Tensor & self, at::Tensor & out);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/arange_meta_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 meta {
+
+TORCH_API at::Tensor & arange_out(at::Tensor & out, const at::Scalar & start, const at::Scalar & end, const at::Scalar & step);
+TORCH_API at::Tensor & arange_outf(const at::Scalar & start, const at::Scalar & end, const at::Scalar & step, at::Tensor & out);
+
+} // namespace meta
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/atan2.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/atan2_ops.h>
+
+namespace at {
+
+
+// aten::atan2.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & atan2_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::atan2_out::call(self, other, out);
+}
+// aten::atan2.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & atan2_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) {
+    return at::_ops::atan2_out::call(self, other, out);
+}
+
+// aten::atan2(Tensor self, Tensor other) -> Tensor
+inline at::Tensor atan2(const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::atan2::call(self, other);
+}
+
+}

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

@@ -0,0 +1,61 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API cat {
+  using schema = at::Tensor (const at::ITensorListRef &, int64_t);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cat")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cat(Tensor[] tensors, int dim=0) -> Tensor")
+  static at::Tensor call(const at::ITensorListRef & tensors, int64_t dim);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::ITensorListRef & tensors, int64_t dim);
+};
+
+struct TORCH_API cat_out {
+  using schema = at::Tensor & (const at::ITensorListRef &, int64_t, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cat")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cat.out(Tensor[] tensors, int dim=0, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::ITensorListRef & tensors, int64_t dim, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::ITensorListRef & tensors, int64_t dim, at::Tensor & out);
+};
+
+struct TORCH_API cat_names {
+  using schema = at::Tensor (at::TensorList, at::Dimname);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cat")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "names")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cat.names(Tensor[] tensors, Dimname dim) -> Tensor")
+  static at::Tensor call(at::TensorList tensors, at::Dimname dim);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList tensors, at::Dimname dim);
+};
+
+struct TORCH_API cat_names_out {
+  using schema = at::Tensor & (at::TensorList, at::Dimname, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cat")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "names_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cat.names_out(Tensor[] tensors, Dimname dim, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(at::TensorList tensors, at::Dimname dim, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList tensors, at::Dimname dim, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cauchy_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 cauchy_ {
+  using schema = at::Tensor & (at::Tensor &, double, double, c10::optional<at::Generator>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cauchy_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cauchy_(Tensor(a!) self, float median=0, float sigma=1, *, Generator? generator=None) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, double median, double sigma, c10::optional<at::Generator> generator);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, double median, double sigma, c10::optional<at::Generator> generator);
+};
+
+struct TORCH_API cauchy_out {
+  using schema = at::Tensor & (const at::Tensor &, double, double, c10::optional<at::Generator>, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cauchy")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cauchy.out(Tensor self, float median=0, float sigma=1, *, Generator? generator=None, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, double median, double sigma, c10::optional<at::Generator> generator, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, double median, double sigma, c10::optional<at::Generator> generator, at::Tensor & out);
+};
+
+struct TORCH_API cauchy {
+  using schema = at::Tensor (const at::Tensor &, double, double, c10::optional<at::Generator>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cauchy")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cauchy(Tensor self, float median=0, float sigma=1, *, Generator? generator=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, double median, double sigma, c10::optional<at::Generator> generator);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, double median, double sigma, c10::optional<at::Generator> generator);
+};
+
+}} // namespace at::_ops

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

@@ -0,0 +1,28 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API cummaxmin_backward {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &, int64_t);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cummaxmin_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cummaxmin_backward(Tensor grad, Tensor input, Tensor indices, int dim) -> Tensor")
+  static at::Tensor call(const at::Tensor & grad, const at::Tensor & input, const at::Tensor & indices, int64_t dim);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad, const at::Tensor & input, const at::Tensor & indices, int64_t dim);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_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/cumsum_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_cumsum_out : public at::meta::structured_cumsum {
+void impl(const at::Tensor & self, int64_t dim, c10::optional<at::ScalarType> dtype, const at::Tensor & out);
+};
+TORCH_API at::Tensor cumsum(const at::Tensor & self, at::Dimname dim, c10::optional<at::ScalarType> dtype=c10::nullopt);
+TORCH_API at::Tensor & cumsum_out(const at::Tensor & self, at::Dimname dim, c10::optional<at::ScalarType> dtype, at::Tensor & out);
+TORCH_API at::Tensor & cumsum_(at::Tensor & self, at::Dimname dim, c10::optional<at::ScalarType> dtype=c10::nullopt);
+} // namespace native
+} // namespace at

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/empty_strided.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_strided_ops.h>
+
+namespace at {
+
+
+// aten::empty_strided(SymInt[] size, SymInt[] stride, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor empty_strided(at::IntArrayRef size, at::IntArrayRef stride, at::TensorOptions options={}) {
+    return at::_ops::empty_strided::call(c10::fromIntArrayRefSlow(size), c10::fromIntArrayRefSlow(stride), 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_strided(at::IntArrayRef size, at::IntArrayRef stride, at::TensorOptions options={}) {
+    return at::_ops::empty_strided::call(c10::fromIntArrayRefSlow(size), c10::fromIntArrayRefSlow(stride), c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+  }
+}
+
+// aten::empty_strided(SymInt[] size, SymInt[] stride, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor empty_strided(at::IntArrayRef size, at::IntArrayRef stride, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory) {
+    return at::_ops::empty_strided::call(c10::fromIntArrayRefSlow(size), c10::fromIntArrayRefSlow(stride), 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_strided(at::IntArrayRef size, at::IntArrayRef stride, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory) {
+    return at::_ops::empty_strided::call(c10::fromIntArrayRefSlow(size), c10::fromIntArrayRefSlow(stride), dtype, layout, device, pin_memory);
+  }
+}
+
+// aten::empty_strided(SymInt[] size, SymInt[] stride, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor empty_strided_symint(c10::SymIntArrayRef size, c10::SymIntArrayRef stride, at::TensorOptions options={}) {
+    return at::_ops::empty_strided::call(size, stride, 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_strided(c10::SymIntArrayRef size, c10::SymIntArrayRef stride, at::TensorOptions options={}) {
+    return at::_ops::empty_strided::call(size, stride, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+  }
+}
+
+// aten::empty_strided(SymInt[] size, SymInt[] stride, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor empty_strided_symint(c10::SymIntArrayRef size, c10::SymIntArrayRef stride, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory) {
+    return at::_ops::empty_strided::call(size, stride, 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_strided(c10::SymIntArrayRef size, c10::SymIntArrayRef stride, c10::optional<at::ScalarType> dtype, c10::optional<at::Layout> layout, c10::optional<at::Device> device, c10::optional<bool> pin_memory) {
+    return at::_ops::empty_strided::call(size, stride, dtype, layout, device, pin_memory);
+  }
+}
+
+// aten::empty_strided.out(SymInt[] size, SymInt[] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_strided_out(at::Tensor & out, at::IntArrayRef size, at::IntArrayRef stride) {
+    return at::_ops::empty_strided_out::call(c10::fromIntArrayRefSlow(size), c10::fromIntArrayRefSlow(stride), out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & empty_strided_out(at::Tensor & out, at::IntArrayRef size, at::IntArrayRef stride) {
+    return at::_ops::empty_strided_out::call(c10::fromIntArrayRefSlow(size), c10::fromIntArrayRefSlow(stride), out);
+  }
+}
+
+// aten::empty_strided.out(SymInt[] size, SymInt[] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_strided_outf(at::IntArrayRef size, at::IntArrayRef stride, at::Tensor & out) {
+    return at::_ops::empty_strided_out::call(c10::fromIntArrayRefSlow(size), c10::fromIntArrayRefSlow(stride), out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & empty_strided_outf(at::IntArrayRef size, at::IntArrayRef stride, at::Tensor & out) {
+    return at::_ops::empty_strided_out::call(c10::fromIntArrayRefSlow(size), c10::fromIntArrayRefSlow(stride), out);
+  }
+}
+
+// aten::empty_strided.out(SymInt[] size, SymInt[] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_strided_symint_out(at::Tensor & out, c10::SymIntArrayRef size, c10::SymIntArrayRef stride) {
+    return at::_ops::empty_strided_out::call(size, stride, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & empty_strided_out(at::Tensor & out, c10::SymIntArrayRef size, c10::SymIntArrayRef stride) {
+    return at::_ops::empty_strided_out::call(size, stride, out);
+  }
+}
+
+// aten::empty_strided.out(SymInt[] size, SymInt[] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & empty_strided_symint_outf(c10::SymIntArrayRef size, c10::SymIntArrayRef stride, at::Tensor & out) {
+    return at::_ops::empty_strided_out::call(size, stride, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & empty_strided_outf(c10::SymIntArrayRef size, c10::SymIntArrayRef stride, at::Tensor & out) {
+    return at::_ops::empty_strided_out::call(size, stride, out);
+  }
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/floor_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 floor(const at::Tensor & self);
+TORCH_API at::Tensor & floor_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & floor_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & floor_(at::Tensor & self);
+
+} // namespace cuda
+} // namespace at

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

@@ -0,0 +1,29 @@
+#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 floor_divide(const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & floor_divide_out(const at::Tensor & self, const at::Tensor & other, at::Tensor & out);
+TORCH_API at::Tensor & floor_divide_(at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor floor_divide_sparse(const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & floor_divide_out_sparse_zerodim(const at::Tensor & self, const at::Tensor & other, at::Tensor & out);
+TORCH_API at::Tensor & floor_divide_sparse_(at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor floor_divide(const at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor & floor_divide_Scalar_out(const at::Tensor & self, const at::Scalar & other, at::Tensor & out);
+TORCH_API at::Tensor & floor_divide_(at::Tensor & self, const at::Scalar & other);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gather_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/gather_backward_ops.h>
+
+namespace at {
+
+
+// aten::gather_backward(Tensor grad, Tensor self, int dim, Tensor index, bool sparse_grad) -> Tensor
+inline at::Tensor gather_backward(const at::Tensor & grad, const at::Tensor & self, int64_t dim, const at::Tensor & index, bool sparse_grad) {
+    return at::_ops::gather_backward::call(grad, self, dim, index, sparse_grad);
+}
+
+}

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu_jvp_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 glu_jvp(const at::Tensor & glu, const at::Tensor & x, const at::Tensor & dx, int64_t dim);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardsigmoid_backward_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 hardsigmoid_backward(const at::Tensor & grad_output, const at::Tensor & self);
+TORCH_API at::Tensor & hardsigmoid_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self);
+TORCH_API at::Tensor & hardsigmoid_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & grad_input);
+
+} // namespace meta
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hstack_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 at::Tensor hstack(at::TensorList tensors);
+TORCH_API at::Tensor & hstack_out(at::Tensor & out, at::TensorList tensors);
+TORCH_API at::Tensor & hstack_outf(at::TensorList tensors, at::Tensor & out);
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_neg.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/is_neg_ops.h>
+
+namespace at {
+
+
+// aten::is_neg(Tensor self) -> bool
+inline bool __dispatch_is_neg(const at::Tensor & self) {
+    return at::_ops::is_neg::call(self);
+}
+
+}

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

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

@@ -0,0 +1,27 @@
+#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/isposinf_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_isposinf_out : public at::meta::structured_isposinf {
+void impl(const at::Tensor & self, const at::Tensor & out);
+};
+TORCH_API at::Tensor isposinf_sparse(const at::Tensor & self);
+TORCH_API at::Tensor & isposinf_sparse_out(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor isposinf_sparse_csr(const at::Tensor & self);
+TORCH_API at::Tensor & isposinf_sparse_csr_out(const at::Tensor & self, at::Tensor & out);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/leaky_relu_backward.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/leaky_relu_backward_ops.h>
+
+namespace at {
+
+
+// aten::leaky_relu_backward.grad_input(Tensor grad_output, Tensor self, Scalar negative_slope, bool self_is_result, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & leaky_relu_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result) {
+    return at::_ops::leaky_relu_backward_grad_input::call(grad_output, self, negative_slope, self_is_result, grad_input);
+}
+// aten::leaky_relu_backward.grad_input(Tensor grad_output, Tensor self, Scalar negative_slope, bool self_is_result, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & leaky_relu_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result, at::Tensor & grad_input) {
+    return at::_ops::leaky_relu_backward_grad_input::call(grad_output, self, negative_slope, self_is_result, grad_input);
+}
+
+// aten::leaky_relu_backward(Tensor grad_output, Tensor self, Scalar negative_slope, bool self_is_result) -> Tensor
+inline at::Tensor leaky_relu_backward(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result) {
+    return at::_ops::leaky_relu_backward::call(grad_output, self, negative_slope, self_is_result);
+}
+
+}

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_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_linalg_lu_solve : public at::impl::MetaBase {
+    
+    
+    void meta(const at::Tensor & LU, const at::Tensor & pivots, const at::Tensor & B, bool left, bool adjoint);
+};
+
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_vander.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/linalg_vander_ops.h>
+
+namespace at {
+
+
+// aten::linalg_vander(Tensor x, *, SymInt? N=None) -> Tensor
+inline at::Tensor linalg_vander(const at::Tensor & x, c10::optional<int64_t> N=c10::nullopt) {
+    return at::_ops::linalg_vander::call(x, N.has_value() ? c10::make_optional(c10::SymInt(*N)) : c10::nullopt);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor linalg_vander(const at::Tensor & x, c10::optional<int64_t> N=c10::nullopt) {
+    return at::_ops::linalg_vander::call(x, N.has_value() ? c10::make_optional(c10::SymInt(*N)) : c10::nullopt);
+  }
+}
+
+// aten::linalg_vander(Tensor x, *, SymInt? N=None) -> Tensor
+inline at::Tensor linalg_vander_symint(const at::Tensor & x, c10::optional<c10::SymInt> N=c10::nullopt) {
+    return at::_ops::linalg_vander::call(x, N);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor linalg_vander(const at::Tensor & x, c10::optional<c10::SymInt> N=c10::nullopt) {
+    return at::_ops::linalg_vander::call(x, N);
+  }
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linspace_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 & linspace_out(at::Tensor & out, const at::Scalar & start, const at::Scalar & end, int64_t steps);
+TORCH_API at::Tensor & linspace_outf(const at::Scalar & start, const at::Scalar & end, int64_t steps, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log_sigmoid_backward.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/log_sigmoid_backward_ops.h>
+
+namespace at {
+
+
+// aten::log_sigmoid_backward.grad_input(Tensor grad_output, Tensor self, Tensor buffer, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & log_sigmoid_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & buffer) {
+    return at::_ops::log_sigmoid_backward_grad_input::call(grad_output, self, buffer, grad_input);
+}
+// aten::log_sigmoid_backward.grad_input(Tensor grad_output, Tensor self, Tensor buffer, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & log_sigmoid_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & buffer, at::Tensor & grad_input) {
+    return at::_ops::log_sigmoid_backward_grad_input::call(grad_output, self, buffer, grad_input);
+}
+
+// aten::log_sigmoid_backward(Tensor grad_output, Tensor self, Tensor buffer) -> Tensor
+inline at::Tensor log_sigmoid_backward(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & buffer) {
+    return at::_ops::log_sigmoid_backward::call(grad_output, self, buffer);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_with_indices_backward_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 max_pool2d_with_indices_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, const at::Tensor & indices);
+TORCH_API at::Tensor & max_pool2d_with_indices_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, const at::Tensor & indices);
+TORCH_API at::Tensor & max_pool2d_with_indices_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, const at::Tensor & indices, at::Tensor & grad_input);
+
+} // namespace meta
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool3d_with_indices_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 max_pool3d_with_indices_out {
+  using schema = ::std::tuple<at::Tensor &,at::Tensor &> (const at::Tensor &, at::IntArrayRef, at::IntArrayRef, at::IntArrayRef, at::IntArrayRef, bool, at::Tensor &, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max_pool3d_with_indices")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max_pool3d_with_indices.out(Tensor self, int[3] kernel_size, int[3] stride=[], int[3] padding=0, int[3] dilation=1, bool ceil_mode=False, *, Tensor(a!) out, Tensor(b!) indices) -> (Tensor(a!), Tensor(b!))")
+  static ::std::tuple<at::Tensor &,at::Tensor &> call(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, at::Tensor & out, at::Tensor & indices);
+  static ::std::tuple<at::Tensor &,at::Tensor &> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, at::Tensor & out, at::Tensor & indices);
+};
+
+struct TORCH_API max_pool3d_with_indices {
+  using schema = ::std::tuple<at::Tensor,at::Tensor> (const at::Tensor &, at::IntArrayRef, at::IntArrayRef, at::IntArrayRef, at::IntArrayRef, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max_pool3d_with_indices")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max_pool3d_with_indices(Tensor self, int[3] kernel_size, int[3] stride=[], int[3] padding=0, int[3] dilation=1, bool ceil_mode=False) -> (Tensor, Tensor)")
+  static ::std::tuple<at::Tensor,at::Tensor> call(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode);
+  static ::std::tuple<at::Tensor,at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode);
+};
+
+}} // namespace at::_ops