Add files using upload-large-folder tool

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_dirichlet_grad_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 _dirichlet_grad(const at::Tensor & x, const at::Tensor & alpha, const at::Tensor & total);
+
+} // namespace cuda
+} // namespace at

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

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor & _embedding_bag_per_sample_weights_backward_out(const at::Tensor & grad, const at::Tensor & weight, const at::Tensor & indices, const at::Tensor & offsets, const at::Tensor & offset2bag, int64_t mode, int64_t padding_idx, at::Tensor & out);
+TORCH_API at::Tensor _embedding_bag_per_sample_weights_backward_cpu(const at::Tensor & grad, const at::Tensor & weight, const at::Tensor & indices, const at::Tensor & offsets, const at::Tensor & offset2bag, int64_t mode, int64_t padding_idx=-1);
+TORCH_API at::Tensor _embedding_bag_per_sample_weights_backward_cuda(const at::Tensor & grad, const at::Tensor & weight, const at::Tensor & indices, const at::Tensor & offsets, const at::Tensor & offset2bag, int64_t mode, int64_t padding_idx=-1);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,25 @@
+#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 _foreach_cos_out(at::TensorList self, at::TensorList out);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_cos_slow(at::TensorList self);
+TORCH_API void foreach_tensor_cos_slow_(at::TensorList self);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_cos_cuda(at::TensorList self);
+TORCH_API void foreach_tensor_cos_cuda_(at::TensorList self);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,25 @@
+#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 _foreach_log_out(at::TensorList self, at::TensorList out);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_log_slow(at::TensorList self);
+TORCH_API void foreach_tensor_log_slow_(at::TensorList self);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_log_cuda(at::TensorList self);
+TORCH_API void foreach_tensor_log_cuda_(at::TensorList self);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_functional_assert_scalar_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 _functional_assert_scalar {
+  using schema = at::Tensor (const at::Scalar &, c10::string_view, 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::_functional_assert_scalar")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_functional_assert_scalar(Scalar self, str assert_msg, Tensor dep_token) -> Tensor")
+  static at::Tensor call(const at::Scalar & self, c10::string_view assert_msg, const at::Tensor & dep_token);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & self, c10::string_view assert_msg, const at::Tensor & dep_token);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_functional_sym_constrain_range_for_size.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/_functional_sym_constrain_range_for_size_ops.h>
+
+namespace at {
+
+
+// aten::_functional_sym_constrain_range_for_size(Scalar size, int? min, int? max, Tensor dep_token) -> Tensor
+inline at::Tensor _functional_sym_constrain_range_for_size(const at::Scalar & size, c10::optional<int64_t> min, c10::optional<int64_t> max, const at::Tensor & dep_token) {
+    return at::_ops::_functional_sym_constrain_range_for_size::call(size, min, max, dep_token);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_backward_data_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 _log_softmax_backward_data(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype);
+TORCH_API at::Tensor & _log_softmax_backward_data_out(at::Tensor & out, const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype);
+TORCH_API at::Tensor & _log_softmax_backward_data_outf(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_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 _log_softmax(const at::Tensor & self, int64_t dim, bool half_to_float);
+TORCH_API at::Tensor & _log_softmax_out(at::Tensor & out, const at::Tensor & self, int64_t dim, bool half_to_float);
+TORCH_API at::Tensor & _log_softmax_outf(const at::Tensor & self, int64_t dim, bool half_to_float, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_make_per_tensor_quantized_tensor_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 _make_per_tensor_quantized_tensor(const at::Tensor & self, double scale, int64_t zero_point);
+
+} // namespace cuda
+} // namespace at

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_pdist_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 at::Tensor _pdist_backward(const at::Tensor & grad, const at::Tensor & self, double p, const at::Tensor & pdist);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_pdist_forward_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 _pdist_forward(const at::Tensor & self, double p=2);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_triton_scaled_dot_attention.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/_triton_scaled_dot_attention_ops.h>
+
+namespace at {
+
+
+// aten::_triton_scaled_dot_attention(Tensor q, Tensor k, Tensor v, float dropout_p=0.0) -> Tensor
+inline at::Tensor _triton_scaled_dot_attention(const at::Tensor & q, const at::Tensor & k, const at::Tensor & v, double dropout_p=0.0) {
+    return at::_ops::_triton_scaled_dot_attention::call(q, k, v, dropout_p);
+}
+
+// aten::_triton_scaled_dot_attention.out(Tensor q, Tensor k, Tensor v, float dropout_p=0.0, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _triton_scaled_dot_attention_out(at::Tensor & out, const at::Tensor & q, const at::Tensor & k, const at::Tensor & v, double dropout_p=0.0) {
+    return at::_ops::_triton_scaled_dot_attention_out::call(q, k, v, dropout_p, out);
+}
+// aten::_triton_scaled_dot_attention.out(Tensor q, Tensor k, Tensor v, float dropout_p=0.0, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _triton_scaled_dot_attention_outf(const at::Tensor & q, const at::Tensor & k, const at::Tensor & v, double dropout_p, at::Tensor & out) {
+    return at::_ops::_triton_scaled_dot_attention_out::call(q, k, v, dropout_p, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bicubic2d_aa_backward_cpu_dispatch.h

@@ -0,0 +1,28 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API at::Tensor _upsample_bicubic2d_aa_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional<double> scales_h=c10::nullopt, c10::optional<double> scales_w=c10::nullopt);
+TORCH_API at::Tensor _upsample_bicubic2d_aa_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional<double> scales_h=c10::nullopt, c10::optional<double> scales_w=c10::nullopt);
+TORCH_API at::Tensor & _upsample_bicubic2d_aa_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional<double> scales_h=c10::nullopt, c10::optional<double> scales_w=c10::nullopt);
+TORCH_API at::Tensor & _upsample_bicubic2d_aa_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w, at::Tensor & grad_input);
+TORCH_API at::Tensor & _upsample_bicubic2d_aa_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional<double> scales_h=c10::nullopt, c10::optional<double> scales_w=c10::nullopt);
+TORCH_API at::Tensor & _upsample_bicubic2d_aa_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w, at::Tensor & grad_input);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bicubic2d_aa_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 _upsample_bicubic2d_aa_backward_grad_input {
+  using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, bool, c10::optional<double>, c10::optional<double>, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_upsample_bicubic2d_aa_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_upsample_bicubic2d_aa_backward.grad_input(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, bool align_corners, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w, at::Tensor & grad_input);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w, at::Tensor & grad_input);
+};
+
+struct TORCH_API _upsample_bicubic2d_aa_backward {
+  using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, bool, c10::optional<double>, c10::optional<double>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_upsample_bicubic2d_aa_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_upsample_bicubic2d_aa_backward(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, bool align_corners, float? scales_h=None, float? scales_w=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bilinear2d_aa_backward_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/_upsample_bilinear2d_aa_backward_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured__upsample_bilinear2d_aa_backward_out_cpu : public at::meta::structured__upsample_bilinear2d_aa_backward {
+void impl(const at::Tensor & grad_output, at::ArrayRef<int64_t> output_size, at::ArrayRef<int64_t> input_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w, const at::Tensor & grad_input);
+};
+struct TORCH_API structured__upsample_bilinear2d_aa_backward_out_cuda : public at::meta::structured__upsample_bilinear2d_aa_backward {
+void impl(const at::Tensor & grad_output, at::ArrayRef<int64_t> output_size, at::ArrayRef<int64_t> input_size, bool align_corners, c10::optional<double> scales_h, c10::optional<double> scales_w, const at::Tensor & grad_input);
+};
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_validate_sparse_bsr_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_bsr_tensor_args(const at::Tensor & crow_indices, const at::Tensor & col_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/_weight_norm_interface_backward_native.h

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <c10/util/Optional.h>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> _weight_norm_interface_backward_out(const at::Tensor & grad_w, const at::Tensor & saved_v, const at::Tensor & saved_g, const at::Tensor & saved_norms, int64_t dim, at::Tensor & out0, at::Tensor & out1);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> weight_norm_backward_cpu(const at::Tensor & grad_w, const at::Tensor & saved_v, const at::Tensor & saved_g, const at::Tensor & saved_norms, int64_t dim);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> weight_norm_backward_cuda(const at::Tensor & grad_w, const at::Tensor & saved_v, const at::Tensor & saved_g, const at::Tensor & saved_norms, int64_t dim);
+} // namespace native
+} // namespace at

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/align_to_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 align_to {
+  using schema = at::Tensor (const at::Tensor &, at::DimnameList);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::align_to")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "align_to(Tensor(a) self, Dimname[] names) -> Tensor(a)")
+  static at::Tensor call(const at::Tensor & self, at::DimnameList names);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::DimnameList names);
+};
+
+struct TORCH_API align_to_ellipsis_idx {
+  using schema = at::Tensor (const at::Tensor &, at::DimnameList, int64_t);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::align_to")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "ellipsis_idx")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "align_to.ellipsis_idx(Tensor(a) self, Dimname[] order, int ellipsis_idx) -> Tensor(a)")
+  static at::Tensor call(const at::Tensor & self, at::DimnameList order, int64_t ellipsis_idx);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::DimnameList order, int64_t ellipsis_idx);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/amin_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 amin(const at::Tensor & self, at::IntArrayRef dim={}, bool keepdim=false);
+TORCH_API at::Tensor & amin_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef dim={}, bool keepdim=false);
+TORCH_API at::Tensor & amin_outf(const at::Tensor & self, at::IntArrayRef dim, bool keepdim, at::Tensor & out);
+
+} // namespace meta
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/any.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/any_ops.h>
+
+namespace at {
+
+
+// aten::any.dim(Tensor self, int dim, bool keepdim=False) -> Tensor
+inline at::Tensor any(const at::Tensor & self, int64_t dim, bool keepdim=false) {
+    return at::_ops::any_dim::call(self, dim, keepdim);
+}
+
+// aten::any.dims(Tensor self, int[]? dim=None, bool keepdim=False) -> Tensor
+inline at::Tensor any(const at::Tensor & self, at::OptionalIntArrayRef dim, bool keepdim=false) {
+    return at::_ops::any_dims::call(self, dim, keepdim);
+}
+
+// aten::any.out(Tensor self, int dim, bool keepdim=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & any_out(at::Tensor & out, const at::Tensor & self, int64_t dim, bool keepdim=false) {
+    return at::_ops::any_out::call(self, dim, keepdim, out);
+}
+// aten::any.out(Tensor self, int dim, bool keepdim=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & any_outf(const at::Tensor & self, int64_t dim, bool keepdim, at::Tensor & out) {
+    return at::_ops::any_out::call(self, dim, keepdim, out);
+}
+
+// aten::any.dims_out(Tensor self, int[]? dim=None, bool keepdim=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & any_out(at::Tensor & out, const at::Tensor & self, at::OptionalIntArrayRef dim, bool keepdim=false) {
+    return at::_ops::any_dims_out::call(self, dim, keepdim, out);
+}
+// aten::any.dims_out(Tensor self, int[]? dim=None, bool keepdim=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & any_outf(const at::Tensor & self, at::OptionalIntArrayRef dim, bool keepdim, at::Tensor & out) {
+    return at::_ops::any_dims_out::call(self, dim, keepdim, out);
+}
+
+// aten::any.dimname(Tensor self, Dimname dim, bool keepdim=False) -> Tensor
+inline at::Tensor any(const at::Tensor & self, at::Dimname dim, bool keepdim=false) {
+    return at::_ops::any_dimname::call(self, dim, keepdim);
+}
+
+// aten::any.dimname_out(Tensor self, Dimname dim, bool keepdim=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & any_out(at::Tensor & out, const at::Tensor & self, at::Dimname dim, bool keepdim=false) {
+    return at::_ops::any_dimname_out::call(self, dim, keepdim, out);
+}
+// aten::any.dimname_out(Tensor self, Dimname dim, bool keepdim=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & any_outf(const at::Tensor & self, at::Dimname dim, bool keepdim, at::Tensor & out) {
+    return at::_ops::any_dimname_out::call(self, dim, keepdim, out);
+}
+
+// aten::any(Tensor self) -> Tensor
+inline at::Tensor any(const at::Tensor & self) {
+    return at::_ops::any::call(self);
+}
+
+// aten::any.all_out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & any_out(at::Tensor & out, const at::Tensor & self) {
+    return at::_ops::any_all_out::call(self, out);
+}
+// aten::any.all_out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & any_outf(const at::Tensor & self, at::Tensor & out) {
+    return at::_ops::any_all_out::call(self, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/as_strided_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 as_strided(const at::Tensor & self, at::IntArrayRef size, at::IntArrayRef stride, c10::optional<int64_t> storage_offset=c10::nullopt);
+TORCH_API at::Tensor as_strided_symint(const at::Tensor & self, c10::SymIntArrayRef size, c10::SymIntArrayRef stride, c10::optional<c10::SymInt> storage_offset=c10::nullopt);
+
+} // namespace meta
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/atanh_cpu_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 cpu {
+
+TORCH_API at::Tensor atanh(const at::Tensor & self);
+TORCH_API at::Tensor & atanh_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & atanh_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & atanh_(at::Tensor & self);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool2d_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 avg_pool2d(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_pool2d_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_pool2d_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 meta
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/binomial_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 binomial(const at::Tensor & count, const at::Tensor & prob, c10::optional<at::Generator> generator=c10::nullopt);
+
+} // namespace cpu
+} // namespace at

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

@@ -0,0 +1,29 @@
+#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 bitwise_left_shift(const at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor & bitwise_left_shift_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor & bitwise_left_shift_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out);
+TORCH_API at::Tensor & bitwise_left_shift_(at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor bitwise_left_shift(const at::Scalar & self, const at::Tensor & other);
+TORCH_API at::Tensor & bitwise_left_shift_out(at::Tensor & out, const at::Scalar & self, const at::Tensor & other);
+TORCH_API at::Tensor & bitwise_left_shift_outf(const at::Scalar & self, const at::Tensor & other, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bmm_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 bmm(const at::Tensor & self, const at::Tensor & mat2);
+TORCH_API at::Tensor & bmm_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & mat2);
+TORCH_API at::Tensor & bmm_outf(const at::Tensor & self, const at::Tensor & mat2, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

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

@@ -0,0 +1,26 @@
+#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/chalf_ops.h>
+
+namespace at {
+
+
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/conv2d_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 conv2d_symint(const at::Tensor & input, const at::Tensor & weight, const c10::optional<at::Tensor> & bias={}, c10::SymIntArrayRef stride=c10::SymInt(1), c10::SymIntArrayRef padding=c10::SymInt(0), c10::SymIntArrayRef dilation=c10::SymInt(1), c10::SymInt groups=1);
+TORCH_API at::Tensor conv2d_padding_symint(const at::Tensor & input, const at::Tensor & weight, const c10::optional<at::Tensor> & bias={}, c10::SymIntArrayRef stride=c10::SymInt(1), c10::string_view padding="valid", c10::SymIntArrayRef dilation=c10::SymInt(1), c10::SymInt groups=1);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cross_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 cross_out {
+  using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, c10::optional<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::cross")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cross.out(Tensor self, Tensor other, int? dim=None, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, c10::optional<int64_t> dim, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, c10::optional<int64_t> dim, at::Tensor & out);
+};
+
+struct TORCH_API cross {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, c10::optional<int64_t>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cross")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cross(Tensor self, Tensor other, int? dim=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const at::Tensor & other, c10::optional<int64_t> dim);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, c10::optional<int64_t> dim);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/einsum_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 einsum(c10::string_view equation, at::TensorList tensors, at::OptionalIntArrayRef path=c10::nullopt);
+
+} // namespace compositeimplicitautograd
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/embedding_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 embedding_backward {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, c10::SymInt, c10::SymInt, bool, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::embedding_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "embedding_backward(Tensor grad, Tensor indices, SymInt num_weights, SymInt padding_idx, bool scale_grad_by_freq, bool sparse) -> Tensor")
+  static at::Tensor call(const at::Tensor & grad, const at::Tensor & indices, c10::SymInt num_weights, c10::SymInt padding_idx, bool scale_grad_by_freq, bool sparse);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad, const at::Tensor & indices, c10::SymInt num_weights, c10::SymInt padding_idx, bool scale_grad_by_freq, bool sparse);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/erfinv_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>
+#include <ATen/ops/erfinv_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_erfinv_out : public at::meta::structured_erfinv {
+void impl(const at::Tensor & self, const at::Tensor & out);
+};
+TORCH_API at::Tensor erfinv_sparse(const at::Tensor & self);
+TORCH_API at::Tensor & erfinv_sparse_out(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & erfinv_sparse_(at::Tensor & self);
+TORCH_API at::Tensor erfinv_sparse_csr(const at::Tensor & self);
+TORCH_API at::Tensor & erfinv_sparse_csr_out(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & erfinv_sparse_csr_(at::Tensor & self);
+} // namespace native
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/exp2_compositeexplicitautogradnonfunctional_dispatch.h

@@ -0,0 +1,24 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeexplicitautogradnonfunctional {
+
+TORCH_API at::Tensor exp2(const at::Tensor & self);
+TORCH_API at::Tensor & exp2_(at::Tensor & self);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_hfft_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 fft_hfft {
+  using schema = at::Tensor (const at::Tensor &, c10::optional<c10::SymInt>, int64_t, c10::optional<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::fft_hfft")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fft_hfft(Tensor self, SymInt? n=None, int dim=-1, str? norm=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, c10::optional<c10::SymInt> n, int64_t dim, c10::optional<c10::string_view> norm);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::optional<c10::SymInt> n, int64_t dim, c10::optional<c10::string_view> norm);
+};
+
+struct TORCH_API fft_hfft_out {
+  using schema = at::Tensor & (const at::Tensor &, c10::optional<c10::SymInt>, int64_t, c10::optional<c10::string_view>, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::fft_hfft")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fft_hfft.out(Tensor self, SymInt? n=None, int dim=-1, str? norm=None, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, c10::optional<c10::SymInt> n, int64_t dim, c10::optional<c10::string_view> norm, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::optional<c10::SymInt> n, int64_t dim, c10::optional<c10::string_view> norm, at::Tensor & out);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_ifft2_compositeimplicitautograd_dispatch.h

@@ -0,0 +1,28 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeimplicitautograd {
+
+TORCH_API at::Tensor fft_ifft2(const at::Tensor & self, at::OptionalIntArrayRef s=c10::nullopt, at::IntArrayRef dim={-2,-1}, c10::optional<c10::string_view> norm=c10::nullopt);
+TORCH_API at::Tensor fft_ifft2_symint(const at::Tensor & self, at::OptionalSymIntArrayRef s=c10::nullopt, at::IntArrayRef dim={-2,-1}, c10::optional<c10::string_view> norm=c10::nullopt);
+TORCH_API at::Tensor & fft_ifft2_out(at::Tensor & out, const at::Tensor & self, at::OptionalIntArrayRef s=c10::nullopt, at::IntArrayRef dim={-2,-1}, c10::optional<c10::string_view> norm=c10::nullopt);
+TORCH_API at::Tensor & fft_ifft2_outf(const at::Tensor & self, at::OptionalIntArrayRef s, at::IntArrayRef dim, c10::optional<c10::string_view> norm, at::Tensor & out);
+TORCH_API at::Tensor & fft_ifft2_symint_out(at::Tensor & out, const at::Tensor & self, at::OptionalSymIntArrayRef s=c10::nullopt, at::IntArrayRef dim={-2,-1}, c10::optional<c10::string_view> norm=c10::nullopt);
+TORCH_API at::Tensor & fft_ifft2_symint_outf(const at::Tensor & self, at::OptionalSymIntArrayRef s, at::IntArrayRef dim, c10::optional<c10::string_view> norm, at::Tensor & out);
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool3d_backward_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 fractional_max_pool3d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices);
+TORCH_API at::Tensor & fractional_max_pool3d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices);
+TORCH_API at::Tensor & fractional_max_pool3d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices, at::Tensor & grad_input);
+
+} // namespace cpu
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool3d_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 fractional_max_pool3d_backward_grad_input {
+  using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, at::IntArrayRef, at::IntArrayRef, 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::fractional_max_pool3d_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fractional_max_pool3d_backward.grad_input(Tensor grad_output, Tensor self, int[3] kernel_size, int[3] output_size, Tensor indices, *, Tensor(a!) grad_input) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices, at::Tensor & grad_input);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices, at::Tensor & grad_input);
+};
+
+struct TORCH_API fractional_max_pool3d_backward {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, at::IntArrayRef, at::IntArrayRef, 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::fractional_max_pool3d_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fractional_max_pool3d_backward(Tensor grad_output, Tensor self, int[3] kernel_size, int[3] output_size, Tensor indices) -> Tensor")
+  static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu_jvp.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/glu_jvp_ops.h>
+
+namespace at {
+
+
+// aten::glu_jvp(Tensor glu, Tensor x, Tensor dx, int dim) -> Tensor
+inline at::Tensor glu_jvp(const at::Tensor & glu, const at::Tensor & x, const at::Tensor & dx, int64_t dim) {
+    return at::_ops::glu_jvp::call(glu, x, dx, dim);
+}
+
+// aten::glu_jvp.out(Tensor glu, Tensor x, Tensor dx, int dim, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & glu_jvp_out(at::Tensor & out, const at::Tensor & glu, const at::Tensor & x, const at::Tensor & dx, int64_t dim) {
+    return at::_ops::glu_jvp_out::call(glu, x, dx, dim, out);
+}
+// aten::glu_jvp.out(Tensor glu, Tensor x, Tensor dx, int dim, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & glu_jvp_outf(const at::Tensor & glu, const at::Tensor & x, const at::Tensor & dx, int64_t dim, at::Tensor & out) {
+    return at::_ops::glu_jvp_out::call(glu, x, dx, dim, out);
+}
+
+}

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_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 hardshrink(const at::Tensor & self, const at::Scalar & lambd=0.5);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

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

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

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

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

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lu_unpack_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 ::std::tuple<at::Tensor,at::Tensor,at::Tensor> lu_unpack(const at::Tensor & LU_data, const at::Tensor & LU_pivots, bool unpack_data=true, bool unpack_pivots=true);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

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

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor,::std::vector<at::Tensor>> miopen_rnn_backward(const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const c10::optional<at::Tensor> & cx, const at::Tensor & output, const c10::optional<at::Tensor> & grad_output, const c10::optional<at::Tensor> & grad_hy, const c10::optional<at::Tensor> & grad_cy, int64_t mode, int64_t hidden_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, at::IntArrayRef batch_sizes, const c10::optional<at::Tensor> & dropout_state, const at::Tensor & reserve, ::std::array<bool,4> output_mask);
+
+} // namespace cuda
+} // namespace at

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_max_pool2d_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 mkldnn_max_pool2d {
+  using schema = 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::mkldnn_max_pool2d")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mkldnn_max_pool2d(Tensor self, int[2] kernel_size, int[2] stride=[], int[2] padding=0, int[2] dilation=1, bool ceil_mode=False) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode);
+  static 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);
+};
+
+struct TORCH_API mkldnn_max_pool2d_out {
+  using schema = at::Tensor & (const at::Tensor &, at::IntArrayRef, at::IntArrayRef, at::IntArrayRef, at::IntArrayRef, 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::mkldnn_max_pool2d")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mkldnn_max_pool2d.out(Tensor self, int[2] kernel_size, int[2] stride=[], int[2] padding=0, int[2] dilation=1, bool ceil_mode=False, *, Tensor(a!) out) -> Tensor(a!)")
+  static 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);
+  static 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);
+};
+
+}} // namespace at::_ops

tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mul_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 mul(const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & mul_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & mul_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out);
+TORCH_API at::Tensor & mul_(at::Tensor & self, const at::Tensor & other);
+
+} // namespace cuda
+} // namespace at