diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cast_Byte.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cast_Byte.h new file mode 100644 index 0000000000000000000000000000000000000000..7e767684f5fee393063085143b8e64693e32ca70 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cast_Byte.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_cast_Byte(Tensor self, bool non_blocking=False) -> Tensor +inline at::Tensor _cast_Byte(const at::Tensor & self, bool non_blocking=false) { + return at::_ops::_cast_Byte::call(self, non_blocking); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_ctc_loss_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_ctc_loss_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..74e4c342c6a77a46821240232fd18db56ae6b6e2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_ctc_loss_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 +#include +#include + +// 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 + +namespace at { + +namespace cpu { + +TORCH_API ::std::tuple _ctc_loss(const at::Tensor & log_probs, const at::Tensor & targets, at::IntArrayRef input_lengths, at::IntArrayRef target_lengths, int64_t blank=0, bool zero_infinity=false); +TORCH_API ::std::tuple _ctc_loss(const at::Tensor & log_probs, const at::Tensor & targets, const at::Tensor & input_lengths, const at::Tensor & target_lengths, int64_t blank=0, bool zero_infinity=false); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_efficient_attention_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_efficient_attention_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..9b19823c80752bcddf1a45b1ed40e2b3f8253be2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_efficient_attention_backward_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 +#include +#include + +// 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 + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple _efficient_attention_backward(const at::Tensor & grad_out_, const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const c10::optional & bias, const at::Tensor & out, const c10::optional & cu_seqlens_q, const c10::optional & cu_seqlens_k, int64_t max_seqlen_q, int64_t max_seqlen_k, const at::Tensor & logsumexp, double dropout_p, const at::Tensor & philox_seed, const at::Tensor & philox_offset, int64_t custom_mask_type, bool bias_requires_grad, c10::optional scale=c10::nullopt, c10::optional num_splits_key=c10::nullopt); +TORCH_API ::std::tuple _efficient_attention_backward_symint(const at::Tensor & grad_out_, const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const c10::optional & bias, const at::Tensor & out, const c10::optional & cu_seqlens_q, const c10::optional & cu_seqlens_k, c10::SymInt max_seqlen_q, c10::SymInt max_seqlen_k, const at::Tensor & logsumexp, double dropout_p, const at::Tensor & philox_seed, const at::Tensor & philox_offset, int64_t custom_mask_type, bool bias_requires_grad, c10::optional scale=c10::nullopt, c10::optional num_splits_key=c10::nullopt); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_div_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_div_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..a97528c8505429e4d9ef7d040bbc9a55196397c6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_div_ops.h @@ -0,0 +1,149 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API _foreach_div_Scalar { + using schema = ::std::vector (at::TensorList, const at::Scalar &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div.Scalar(Tensor[] self, Scalar scalar) -> Tensor[]") + static ::std::vector call(at::TensorList self, const at::Scalar & scalar); + static ::std::vector redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, const at::Scalar & scalar); +}; + +struct TORCH_API _foreach_div__Scalar { + using schema = void (at::TensorList, const at::Scalar &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div_.Scalar(Tensor(a!)[] self, Scalar scalar) -> ()") + static void call(at::TensorList self, const at::Scalar & scalar); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, const at::Scalar & scalar); +}; + +struct TORCH_API _foreach_div_List { + using schema = ::std::vector (at::TensorList, at::TensorList); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "List") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div.List(Tensor[] self, Tensor[] other) -> Tensor[]") + static ::std::vector call(at::TensorList self, at::TensorList other); + static ::std::vector redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList other); +}; + +struct TORCH_API _foreach_div__List { + using schema = void (at::TensorList, at::TensorList); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "List") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div_.List(Tensor(a!)[] self, Tensor[] other) -> ()") + static void call(at::TensorList self, at::TensorList other); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList other); +}; + +struct TORCH_API _foreach_div_ScalarList { + using schema = ::std::vector (at::TensorList, at::ArrayRef); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "ScalarList") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div.ScalarList(Tensor[] self, Scalar[] scalars) -> Tensor[]") + static ::std::vector call(at::TensorList self, at::ArrayRef scalars); + static ::std::vector redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::ArrayRef scalars); +}; + +struct TORCH_API _foreach_div__ScalarList { + using schema = void (at::TensorList, at::ArrayRef); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "ScalarList") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div_.ScalarList(Tensor(a!)[] self, Scalar[] scalars) -> ()") + static void call(at::TensorList self, at::ArrayRef scalars); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::ArrayRef scalars); +}; + +struct TORCH_API _foreach_div_Tensor { + using schema = ::std::vector (at::TensorList, 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::_foreach_div") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div.Tensor(Tensor[] self, Tensor other) -> Tensor[]") + static ::std::vector call(at::TensorList self, const at::Tensor & other); + static ::std::vector redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, const at::Tensor & other); +}; + +struct TORCH_API _foreach_div__Tensor { + using schema = void (at::TensorList, 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::_foreach_div_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div_.Tensor(Tensor(a!)[] self, Tensor other) -> ()") + static void call(at::TensorList self, const at::Tensor & other); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, const at::Tensor & other); +}; + +struct TORCH_API _foreach_div_Scalar_out { + using schema = void (at::TensorList, const at::Scalar &, at::TensorList); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div.Scalar_out(Tensor[] self, Scalar scalar, *, Tensor(a!)[] out) -> ()") + static void call(at::TensorList self, const at::Scalar & scalar, at::TensorList out); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, const at::Scalar & scalar, at::TensorList out); +}; + +struct TORCH_API _foreach_div_List_out { + using schema = void (at::TensorList, at::TensorList, at::TensorList); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "List_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div.List_out(Tensor[] self, Tensor[] other, *, Tensor(a!)[] out) -> ()") + static void call(at::TensorList self, at::TensorList other, at::TensorList out); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList other, at::TensorList out); +}; + +struct TORCH_API _foreach_div_ScalarList_out { + using schema = void (at::TensorList, at::ArrayRef, at::TensorList); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "ScalarList_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div.ScalarList_out(Tensor[] self, Scalar[] scalars, *, Tensor(a!)[] out) -> ()") + static void call(at::TensorList self, at::ArrayRef scalars, at::TensorList out); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::ArrayRef scalars, at::TensorList out); +}; + +struct TORCH_API _foreach_div_Tensor_out { + using schema = void (at::TensorList, const at::Tensor &, at::TensorList); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_div") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_div.Tensor_out(Tensor[] self, Tensor other, *, Tensor(a!)[] out) -> ()") + static void call(at::TensorList self, const at::Tensor & other, at::TensorList out); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, const at::Tensor & other, at::TensorList out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_gather_sparse_backward.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_gather_sparse_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..ac0aed6fe52cdf841c36788a70d10fdcea657f96 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_gather_sparse_backward.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_gather_sparse_backward(Tensor self, int dim, Tensor index, Tensor grad) -> Tensor +inline at::Tensor _gather_sparse_backward(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & grad) { + return at::_ops::_gather_sparse_backward::call(self, dim, index, grad); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_has_same_storage_numel_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_has_same_storage_numel_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..c7044f641bcfc262b3395ee87e179b5137018955 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_has_same_storage_numel_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API _has_same_storage_numel { + using schema = bool (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::_has_same_storage_numel") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_has_same_storage_numel(Tensor self, Tensor other) -> bool") + static bool call(const at::Tensor & self, const at::Tensor & other); + static bool redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl_native.h new file mode 100644 index 0000000000000000000000000000000000000000..d81fa4f3094817cbe3e04d8635ef1a35979e759a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl_native.h @@ -0,0 +1,25 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor _index_put_impl(const at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false); +TORCH_API at::Tensor & _index_put_impl_out(const at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate, bool unsafe, at::Tensor & out); +TORCH_API at::Tensor & _index_put_impl_(at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false); +TORCH_API at::Tensor & _index_put_impl_quantized_cpu_(at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false); +TORCH_API at::Tensor & _index_put_impl_quantized_cuda_(at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_eigh_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_eigh_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..34ce1d188551928520447d8844b3fe99fe206c60 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_eigh_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 +#include +#include + +// 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 + +namespace at { + +namespace cpu { + +TORCH_API ::std::tuple _linalg_eigh(const at::Tensor & A, c10::string_view UPLO="L", bool compute_v=true); +TORCH_API ::std::tuple _linalg_eigh_out(at::Tensor & eigenvalues, at::Tensor & eigenvectors, const at::Tensor & A, c10::string_view UPLO="L", bool compute_v=true); +TORCH_API ::std::tuple _linalg_eigh_outf(const at::Tensor & A, c10::string_view UPLO, bool compute_v, at::Tensor & eigenvalues, at::Tensor & eigenvectors); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_storage_offsets_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_storage_offsets_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..bf57e1ff83a2b0962e21f817969cb131b545c139 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_storage_offsets_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API _nested_tensor_storage_offsets { + 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_storage_offsets") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_tensor_storage_offsets(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_storage_offsets_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_storage_offsets") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_tensor_storage_offsets.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 diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nnpack_spatial_convolution_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nnpack_spatial_convolution_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a68d9058703fa9c50d0f99138600a03ab5349f0d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nnpack_spatial_convolution_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor _nnpack_spatial_convolution(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef padding, at::IntArrayRef stride=1); +TORCH_API at::Tensor & _nnpack_spatial_convolution_out_symint(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_prelu_kernel_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_prelu_kernel_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..6c3a5d636a1c50a96b83be18355298606ee5d4ea --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_prelu_kernel_backward_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API _prelu_kernel_backward { + using schema = ::std::tuple (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::_prelu_kernel_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_prelu_kernel_backward(Tensor grad_output, Tensor self, Tensor weight) -> (Tensor, Tensor)") + static ::std::tuple call(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_saturate_weight_to_fp16.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_saturate_weight_to_fp16.h new file mode 100644 index 0000000000000000000000000000000000000000..6c26dc5f58698eb73935bcb2d36fef27363971d0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_saturate_weight_to_fp16.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_saturate_weight_to_fp16(Tensor weight) -> Tensor +inline at::Tensor _saturate_weight_to_fp16(const at::Tensor & weight) { + return at::_ops::_saturate_weight_to_fp16::call(weight); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_efficient_attention.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_efficient_attention.h new file mode 100644 index 0000000000000000000000000000000000000000..5437d826858922055b0c3524dd6ae172b06b584e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_efficient_attention.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_scaled_dot_product_efficient_attention(Tensor query, Tensor key, Tensor value, Tensor? attn_bias, bool compute_log_sumexp, float dropout_p=0.0, bool is_causal=False, *, float? scale=None) -> (Tensor output, Tensor log_sumexp, Tensor philox_seed, Tensor philox_offset) +inline ::std::tuple _scaled_dot_product_efficient_attention(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const c10::optional & attn_bias, bool compute_log_sumexp, double dropout_p=0.0, bool is_causal=false, c10::optional scale=c10::nullopt) { + return at::_ops::_scaled_dot_product_efficient_attention::call(query, key, value, attn_bias, compute_log_sumexp, dropout_p, is_causal, scale); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..90c9b73e02689ec145f196305bc143f58797362d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_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 +#include +#include + +// 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 + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor _softmax_backward_data(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype); +TORCH_API at::Tensor & _softmax_backward_data_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype); +TORCH_API at::Tensor & _softmax_backward_data_outf(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype, at::Tensor & grad_input); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_bsr_tensor_unsafe_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_bsr_tensor_unsafe_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..123c81fbaf9c09f38489931d87e9b2c8364037e4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_bsr_tensor_unsafe_compositeimplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor _sparse_bsr_tensor_unsafe(const at::Tensor & crow_indices, const at::Tensor & col_indices, const at::Tensor & values, at::IntArrayRef size, at::TensorOptions options={}); +TORCH_API at::Tensor _sparse_bsr_tensor_unsafe(const at::Tensor & crow_indices, const at::Tensor & col_indices, const at::Tensor & values, at::IntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_mm_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_mm_native.h new file mode 100644 index 0000000000000000000000000000000000000000..be526210d3321612d339630f5539f47dc7174916 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_mm_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor _sparse_mm(const at::Tensor & sparse, const at::Tensor & dense); +TORCH_API at::Tensor _sparse_mm(const at::Tensor & sparse, const at::Tensor & dense, c10::string_view reduce); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_standard_gamma_grad_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_standard_gamma_grad_native.h new file mode 100644 index 0000000000000000000000000000000000000000..1a5bd49bfad04b10d0badb985a7b673bdf50c847 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_standard_gamma_grad_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & _standard_gamma_grad_out(const at::Tensor & self, const at::Tensor & output, at::Tensor & out); +TORCH_API at::Tensor _standard_gamma_grad_cpu(const at::Tensor & self, const at::Tensor & output); +TORCH_API at::Tensor _standard_gamma_grad_cuda(const at::Tensor & self, const at::Tensor & output); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_autograd_multiple_dispatch_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_autograd_multiple_dispatch_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..af8f8ad9422e4402a164e41868b8d1717a4b1022 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_autograd_multiple_dispatch_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor _test_autograd_multiple_dispatch(const at::Tensor & self); +TORCH_API at::Tensor & _test_autograd_multiple_dispatch_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & _test_autograd_multiple_dispatch_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_fused_lstm_cell_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_fused_lstm_cell_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c64f6b554409de8fe5b7fa2772ac3d6e0b3c05c8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_fused_lstm_cell_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API ::std::tuple _thnn_fused_lstm_cell_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, const at::Tensor & input_gates, const at::Tensor & hidden_gates, const at::Tensor & cx, const c10::optional & input_bias={}, const c10::optional & hidden_bias={}); +TORCH_API ::std::tuple _thnn_fused_lstm_cell_outf(const at::Tensor & input_gates, const at::Tensor & hidden_gates, const at::Tensor & cx, const c10::optional & input_bias, const c10::optional & hidden_bias, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a555ded0dcf54265a44cccb0f528905237fd4caa --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique_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 +#include +#include + +// 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 + +namespace at { + +namespace cpu { + +TORCH_API ::std::tuple _unique(const at::Tensor & self, bool sorted=true, bool return_inverse=false); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bicubic2d_aa_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bicubic2d_aa_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..54740ced4327022e708e8ccedb7ac63c0f56fbaf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bicubic2d_aa_compositeimplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor _upsample_bicubic2d_aa(const at::Tensor & input, at::OptionalIntArrayRef output_size, bool align_corners, c10::optional> scale_factors); +TORCH_API at::Tensor _upsample_bicubic2d_aa_symint(const at::Tensor & input, at::OptionalSymIntArrayRef output_size, bool align_corners, c10::optional> scale_factors); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm.h new file mode 100644 index 0000000000000000000000000000000000000000..2fe87d080232741a72696bfa2eb5432be4309826 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_weight_norm(Tensor v, Tensor g, int dim=0) -> Tensor +inline at::Tensor _weight_norm(const at::Tensor & v, const at::Tensor & g, int64_t dim=0) { + return at::_ops::_weight_norm::call(v, g, dim); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool3d_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool3d_native.h new file mode 100644 index 0000000000000000000000000000000000000000..73bf982e94ddf290832d396d2a0c4c4d457a9004 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool3d_native.h @@ -0,0 +1,26 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_adaptive_max_pool3d_out_cpu : public at::meta::structured_adaptive_max_pool3d { +void impl(const at::Tensor & self, at::IntArrayRef output_size, const at::Tensor & out, const at::Tensor & indices); +}; +struct TORCH_API structured_adaptive_max_pool3d_out_cuda : public at::meta::structured_adaptive_max_pool3d { +void impl(const at::Tensor & self, at::IntArrayRef output_size, const at::Tensor & out, const at::Tensor & indices); +}; +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/add_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/add_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..935b184402a06f99844e2b8088dea574ec6ed0c1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/add_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// 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 + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor add(const at::Tensor & self, const at::Scalar & other, const at::Scalar & alpha=1); +TORCH_API at::Tensor & add_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other, const at::Scalar & alpha=1); +TORCH_API at::Tensor & add_outf(const at::Tensor & self, const at::Scalar & other, const at::Scalar & alpha, at::Tensor & out); +TORCH_API at::Tensor & add_(at::Tensor & self, const at::Scalar & other, const at::Scalar & alpha=1); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/add_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/add_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..76bdc28274e26c2993bfe226dcfbae9429e554c3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/add_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor add(const at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha=1); +TORCH_API at::Tensor & add_(at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha=1); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/alias_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/alias_native.h new file mode 100644 index 0000000000000000000000000000000000000000..2a20e02eebd39c50c13fdf01200488d1a686a7c8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/alias_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor alias(const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..74d08b99c172aae67a5f204e05a8d70b5597b47f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_backward_native.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_avg_pool3d_backward_out_cpu : public at::meta::structured_avg_pool3d_backward { +void impl(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override, const at::Tensor & grad_input); +}; +struct TORCH_API structured_avg_pool3d_backward_out_cuda : public at::meta::structured_avg_pool3d_backward { +void impl(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override, const at::Tensor & grad_input); +}; +TORCH_API at::Tensor mkldnn_avg_pool3d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override); +TORCH_API at::Tensor & mkldnn_avg_pool3d_backward_out(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override, at::Tensor & grad_input); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cauchy_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cauchy_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..6a89327379f4ed7b00354c2f08fd45ab5be5b33f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cauchy_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 +#include +#include + +// 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 + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor & cauchy_(at::Tensor & self, double median=0, double sigma=1, c10::optional generator=c10::nullopt); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_convolution_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_convolution_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..3351ab2d648e017c1ee90bf925f8f39d3d9f183b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_convolution_cuda_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// 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 + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor cudnn_convolution(const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef padding, at::IntArrayRef stride, at::IntArrayRef dilation, int64_t groups, bool benchmark, bool deterministic, bool allow_tf32); +TORCH_API at::Tensor cudnn_convolution_symint(const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, bool allow_tf32); +TORCH_API at::Tensor & cudnn_convolution_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef padding, at::IntArrayRef stride, at::IntArrayRef dilation, int64_t groups, bool benchmark, bool deterministic, bool allow_tf32); +TORCH_API at::Tensor & cudnn_convolution_outf(const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef padding, at::IntArrayRef stride, at::IntArrayRef dilation, int64_t groups, bool benchmark, bool deterministic, bool allow_tf32, at::Tensor & out); +TORCH_API at::Tensor & cudnn_convolution_symint_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, bool allow_tf32); +TORCH_API at::Tensor & cudnn_convolution_symint_outf(const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, bool allow_tf32, at::Tensor & out); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagflat_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagflat_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..23d79106c7f45b82c101d360dd31b3128894d40c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagflat_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor diagflat(const at::Tensor & self, int64_t offset=0); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fake_quantize_per_channel_affine_cachemask_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fake_quantize_per_channel_affine_cachemask_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ccec06658d996ab665d2ffbe4b1e6e146eca30a1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fake_quantize_per_channel_affine_cachemask_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 +#include +#include + +// 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 + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple fake_quantize_per_channel_affine_cachemask(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis, int64_t quant_min, int64_t quant_max); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_fftfreq_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_fftfreq_native.h new file mode 100644 index 0000000000000000000000000000000000000000..e5d7c16641a0066a9c914840b11db712735b6666 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_fftfreq_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor fft_fftfreq(int64_t n, double d=1.0, c10::optional dtype={}, c10::optional layout={}, c10::optional device={}, c10::optional pin_memory={}); +TORCH_API at::Tensor & fft_fftfreq_out(int64_t n, double d, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_ifft.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_ifft.h new file mode 100644 index 0000000000000000000000000000000000000000..02e9a00d7e15fbc3046fb0f9ff1957fbcbe9f785 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_ifft.h @@ -0,0 +1,91 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::fft_ifft(Tensor self, SymInt? n=None, int dim=-1, str? norm=None) -> Tensor +inline at::Tensor fft_ifft(const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt) { + return at::_ops::fft_ifft::call(self, n.has_value() ? c10::make_optional(c10::SymInt(*n)) : c10::nullopt, dim, norm); +} +namespace symint { + template ::value>> + at::Tensor fft_ifft(const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt) { + return at::_ops::fft_ifft::call(self, n.has_value() ? c10::make_optional(c10::SymInt(*n)) : c10::nullopt, dim, norm); + } +} + +// aten::fft_ifft(Tensor self, SymInt? n=None, int dim=-1, str? norm=None) -> Tensor +inline at::Tensor fft_ifft_symint(const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt) { + return at::_ops::fft_ifft::call(self, n, dim, norm); +} +namespace symint { + template ::value>> + at::Tensor fft_ifft(const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt) { + return at::_ops::fft_ifft::call(self, n, dim, norm); + } +} + +// aten::fft_ifft.out(Tensor self, SymInt? n=None, int dim=-1, str? norm=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & fft_ifft_out(at::Tensor & out, const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt) { + return at::_ops::fft_ifft_out::call(self, n.has_value() ? c10::make_optional(c10::SymInt(*n)) : c10::nullopt, dim, norm, out); +} +namespace symint { + template ::value>> + at::Tensor & fft_ifft_out(at::Tensor & out, const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt) { + return at::_ops::fft_ifft_out::call(self, n.has_value() ? c10::make_optional(c10::SymInt(*n)) : c10::nullopt, dim, norm, out); + } +} + +// aten::fft_ifft.out(Tensor self, SymInt? n=None, int dim=-1, str? norm=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & fft_ifft_outf(const at::Tensor & self, c10::optional n, int64_t dim, c10::optional norm, at::Tensor & out) { + return at::_ops::fft_ifft_out::call(self, n.has_value() ? c10::make_optional(c10::SymInt(*n)) : c10::nullopt, dim, norm, out); +} +namespace symint { + template ::value>> + at::Tensor & fft_ifft_outf(const at::Tensor & self, c10::optional n, int64_t dim, c10::optional norm, at::Tensor & out) { + return at::_ops::fft_ifft_out::call(self, n.has_value() ? c10::make_optional(c10::SymInt(*n)) : c10::nullopt, dim, norm, out); + } +} + +// aten::fft_ifft.out(Tensor self, SymInt? n=None, int dim=-1, str? norm=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & fft_ifft_symint_out(at::Tensor & out, const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt) { + return at::_ops::fft_ifft_out::call(self, n, dim, norm, out); +} +namespace symint { + template ::value>> + at::Tensor & fft_ifft_out(at::Tensor & out, const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt) { + return at::_ops::fft_ifft_out::call(self, n, dim, norm, out); + } +} + +// aten::fft_ifft.out(Tensor self, SymInt? n=None, int dim=-1, str? norm=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & fft_ifft_symint_outf(const at::Tensor & self, c10::optional n, int64_t dim, c10::optional norm, at::Tensor & out) { + return at::_ops::fft_ifft_out::call(self, n, dim, norm, out); +} +namespace symint { + template ::value>> + at::Tensor & fft_ifft_outf(const at::Tensor & self, c10::optional n, int64_t dim, c10::optional norm, at::Tensor & out) { + return at::_ops::fft_ifft_out::call(self, n, dim, norm, out); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/full_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/full_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..2ad293e3f7cf146d08d1800bb614a59a96433f3c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/full_compositeexplicitautograd_dispatch.h @@ -0,0 +1,34 @@ +#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 +#include +#include + +// 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 + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor full(at::IntArrayRef size, const at::Scalar & fill_value, c10::optional names, at::TensorOptions options={}); +TORCH_API at::Tensor full(at::IntArrayRef size, const at::Scalar & fill_value, c10::optional names, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor & full_out(at::Tensor & out, at::IntArrayRef size, const at::Scalar & fill_value, c10::optional names); +TORCH_API at::Tensor & full_outf(at::IntArrayRef size, const at::Scalar & fill_value, c10::optional names, at::Tensor & out); +TORCH_API at::Tensor full(at::IntArrayRef size, const at::Scalar & fill_value, at::TensorOptions options={}); +TORCH_API at::Tensor full(at::IntArrayRef size, const at::Scalar & fill_value, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor full_symint(c10::SymIntArrayRef size, const at::Scalar & fill_value, at::TensorOptions options={}); +TORCH_API at::Tensor full_symint(c10::SymIntArrayRef size, const at::Scalar & fill_value, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor & full_out(at::Tensor & out, at::IntArrayRef size, const at::Scalar & fill_value); +TORCH_API at::Tensor & full_outf(at::IntArrayRef size, const at::Scalar & fill_value, at::Tensor & out); +TORCH_API at::Tensor & full_symint_out(at::Tensor & out, c10::SymIntArrayRef size, const at::Scalar & fill_value); +TORCH_API at::Tensor & full_symint_outf(c10::SymIntArrayRef size, const at::Scalar & fill_value, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ge_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ge_native.h new file mode 100644 index 0000000000000000000000000000000000000000..10a606d243640058f3fc2c553c190eb4a38fef42 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ge_native.h @@ -0,0 +1,31 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_ge_Scalar_out : public at::meta::structured_ge_Scalar { +void impl(const at::Tensor & self, const at::Scalar & other, const at::Tensor & out); +}; +TORCH_API at::Tensor ge_scalar_nested(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor ge_quantized_cpu(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & ge_out_quantized_cpu(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +struct TORCH_API structured_ge_Tensor_out : public at::meta::structured_ge_Tensor { +void impl(const at::Tensor & self, const at::Tensor & other, const at::Tensor & out); +}; +TORCH_API at::Tensor ge_quantized_cpu(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & ge_out_quantized_cpu(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/group_norm_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/group_norm_native.h new file mode 100644 index 0000000000000000000000000000000000000000..f22ad3ca9a71bcd67ed978299118ad34d87daa4b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/group_norm_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor group_norm(const at::Tensor & input, int64_t num_groups, const c10::optional & weight={}, const c10::optional & bias={}, double eps=1e-05, bool cudnn_enabled=true); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gt_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gt_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..359ba036e171128210e518de814c615484bca24d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gt_meta_dispatch.h @@ -0,0 +1,30 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// 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 + +namespace at { + +namespace meta { + +TORCH_API at::Tensor gt(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & gt_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & gt_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +TORCH_API at::Tensor & gt_(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor gt(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & gt_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & gt_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & gt_(at::Tensor & self, const at::Tensor & other); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..34819b9c9b3b79bc52463c381a563e722046dca8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_hardshrink : public TensorIteratorBase { + + + void meta(const at::Tensor & self, const at::Scalar & lambd); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hinge_embedding_loss_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hinge_embedding_loss_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..e5b29bc9b9f7e3b8ef594bc0295912bb24f88935 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hinge_embedding_loss_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor hinge_embedding_loss(const at::Tensor & self, const at::Tensor & target, double margin=1.0, int64_t reduction=at::Reduction::Mean); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/i0_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/i0_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..f21daeb1bfe50734f5646f253d62d86a33406c04 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/i0_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 +#include +#include + +// 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 + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor i0(const at::Tensor & self); +TORCH_API at::Tensor & i0_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & i0_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & i0_(at::Tensor & self); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_coalesced_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_coalesced_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..34c78ad9f09d6207cc6d03eb39f10534cd9076d2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_coalesced_compositeexplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// 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 + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API bool is_coalesced(const at::Tensor & self); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isin_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isin_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..bd02933656f4ecfdb3a19f02cc78303158f93ef1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isin_cpu_dispatch.h @@ -0,0 +1,31 @@ +#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 +#include +#include + +// 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 + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor isin(const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false); +TORCH_API at::Tensor & isin_out(at::Tensor & out, const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false); +TORCH_API at::Tensor & isin_outf(const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out); +TORCH_API at::Tensor isin(const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique=false, bool invert=false); +TORCH_API at::Tensor & isin_out(at::Tensor & out, const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique=false, bool invert=false); +TORCH_API at::Tensor & isin_outf(const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique, bool invert, at::Tensor & out); +TORCH_API at::Tensor isin(const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false); +TORCH_API at::Tensor & isin_out(at::Tensor & out, const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false); +TORCH_API at::Tensor & isin_outf(const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/le_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/le_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..1db66e031270da765232ce8d5baa2158bf73e19c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/le_meta_dispatch.h @@ -0,0 +1,30 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// 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 + +namespace at { + +namespace meta { + +TORCH_API at::Tensor le(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & le_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & le_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +TORCH_API at::Tensor & le_(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor le(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & le_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & le_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & le_(at::Tensor & self, const at::Tensor & other); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_cholesky_ex_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_cholesky_ex_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..120f835c20223366de5637d7f85fc4db1acd264d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_cholesky_ex_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API linalg_cholesky_ex { + using schema = ::std::tuple (const at::Tensor &, bool, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_cholesky_ex") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_cholesky_ex(Tensor self, *, bool upper=False, bool check_errors=False) -> (Tensor L, Tensor info)") + static ::std::tuple call(const at::Tensor & self, bool upper, bool check_errors); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, bool upper, bool check_errors); +}; + +struct TORCH_API linalg_cholesky_ex_L { + using schema = ::std::tuple (const at::Tensor &, bool, bool, at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_cholesky_ex") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "L") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_cholesky_ex.L(Tensor self, *, bool upper=False, bool check_errors=False, Tensor(a!) L, Tensor(b!) info) -> (Tensor(a!) L, Tensor(b!) info)") + static ::std::tuple call(const at::Tensor & self, bool upper, bool check_errors, at::Tensor & L, at::Tensor & info); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, bool upper, bool check_errors, at::Tensor & L, at::Tensor & info); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_det_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_det_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..b7ec419f0b79876a4d1e51bf319baf120fc3ce64 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_det_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor linalg_det(const at::Tensor & A); +TORCH_API at::Tensor & linalg_det_out(at::Tensor & out, const at::Tensor & A); +TORCH_API at::Tensor & linalg_det_outf(const at::Tensor & A, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_matrix_exp_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_matrix_exp_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..a66562f96947f829325e5d5ba44e438a48a5ab5e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_matrix_exp_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API linalg_matrix_exp { + 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::linalg_matrix_exp") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_matrix_exp(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 linalg_matrix_exp_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::linalg_matrix_exp") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_matrix_exp.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 diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_not.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_not.h new file mode 100644 index 0000000000000000000000000000000000000000..a9d4232ebc373b7802bb6c25c7a62f3e6d7e99a2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_not.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::logical_not(Tensor self) -> Tensor +inline at::Tensor logical_not(const at::Tensor & self) { + return at::_ops::logical_not::call(self); +} + +// aten::logical_not.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & logical_not_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::logical_not_out::call(self, out); +} +// aten::logical_not.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & logical_not_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::logical_not_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lt_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lt_native.h new file mode 100644 index 0000000000000000000000000000000000000000..1d5ec0f4d5623723fe8639fd7e4ba3140d3c522e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lt_native.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_lt_Scalar_out : public at::meta::structured_lt_Scalar { +void impl(const at::Tensor & self, const at::Scalar & other, const at::Tensor & out); +}; +TORCH_API at::Tensor lt_quantized_cpu(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & lt_out_quantized_cpu(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +struct TORCH_API structured_lt_Tensor_out : public at::meta::structured_lt_Tensor { +void impl(const at::Tensor & self, const at::Tensor & other, const at::Tensor & out); +}; +TORCH_API at::Tensor lt_quantized_cpu(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & lt_out_quantized_cpu(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mT_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mT_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..4ee2bcf3e1d2c5545de4a7af15da37007d9bea76 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mT_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor mT(const at::Tensor & self); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matmul.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matmul.h new file mode 100644 index 0000000000000000000000000000000000000000..102a2e27175e681c10816e5ebf229d9b086bdd49 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matmul.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::matmul(Tensor self, Tensor other) -> Tensor +inline at::Tensor matmul(const at::Tensor & self, const at::Tensor & other) { + return at::_ops::matmul::call(self, other); +} + +// aten::matmul.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & matmul_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) { + return at::_ops::matmul_out::call(self, other, out); +} +// aten::matmul.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & matmul_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) { + return at::_ops::matmul_out::call(self, other, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_with_indices_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_with_indices_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..9bd4d715527d294ee7182d85a264379ef27f72d9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_with_indices_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 +#include +#include + +// 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 + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple max_pool2d_with_indices(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, at::IntArrayRef dilation=1, bool ceil_mode=false); +TORCH_API ::std::tuple max_pool2d_with_indices_out(at::Tensor & out, at::Tensor & indices, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, at::IntArrayRef dilation=1, bool ceil_mode=false); +TORCH_API ::std::tuple max_pool2d_with_indices_outf(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); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_adaptive_avg_pool2d_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_adaptive_avg_pool2d_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..d1dc8e9589101c2e7f0a7224f872fe508b1023b1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_adaptive_avg_pool2d_backward_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API mkldnn_adaptive_avg_pool2d_backward { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::mkldnn_adaptive_avg_pool2d_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mkldnn_adaptive_avg_pool2d_backward(Tensor grad_output, Tensor self) -> Tensor") + static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self); +}; + +struct TORCH_API mkldnn_adaptive_avg_pool2d_backward_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::mkldnn_adaptive_avg_pool2d_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mkldnn_adaptive_avg_pool2d_backward.out(Tensor grad_output, Tensor self, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_rnn_layer_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_rnn_layer_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8eee96e8c6150b72f9cad8299d0b182612559880 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_rnn_layer_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API ::std::tuple mkldnn_rnn_layer_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3, const at::Tensor & input, const at::Tensor & weight0, const at::Tensor & weight1, const at::Tensor & weight2, const at::Tensor & weight3, const at::Tensor & hx_, const at::Tensor & cx_, bool reverse, at::IntArrayRef batch_sizes, int64_t mode, int64_t hidden_size, int64_t num_layers, bool has_biases, bool bidirectional, bool batch_first, bool train); +TORCH_API ::std::tuple mkldnn_rnn_layer_outf(const at::Tensor & input, const at::Tensor & weight0, const at::Tensor & weight1, const at::Tensor & weight2, const at::Tensor & weight3, const at::Tensor & hx_, const at::Tensor & cx_, bool reverse, at::IntArrayRef batch_sizes, int64_t mode, int64_t hidden_size, int64_t num_layers, bool has_biases, bool bidirectional, bool batch_first, bool train, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/negative_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/negative_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..2cbf842bbd0825a9f0ec4574ad7f0d066f6b0881 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/negative_ops.h @@ -0,0 +1,50 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API negative { + 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::negative") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "negative(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 negative_ { + using schema = at::Tensor & (at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::negative_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "negative_(Tensor(a!) self) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self); +}; + +struct TORCH_API negative_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::negative") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "negative.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 diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nonzero_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nonzero_native.h new file mode 100644 index 0000000000000000000000000000000000000000..40a5a211dd714ce948228ab42a576a50e939c7b9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nonzero_native.h @@ -0,0 +1,24 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor nonzero_cpu(const at::Tensor & self); +TORCH_API at::Tensor & nonzero_out_cpu(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor nonzero_cuda(const at::Tensor & self); +TORCH_API at::Tensor & nonzero_out_cuda(const at::Tensor & self, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_max_pool2d.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_max_pool2d.h new file mode 100644 index 0000000000000000000000000000000000000000..13cce5d369e0491b0864e92855701fff2cbdc348 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_max_pool2d.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::quantized_max_pool2d(Tensor self, int[2] kernel_size, int[2] stride=[], int[2] padding=0, int[2] dilation=1, bool ceil_mode=False) -> Tensor +inline at::Tensor quantized_max_pool2d(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, at::IntArrayRef dilation=1, bool ceil_mode=false) { + return at::_ops::quantized_max_pool2d::call(self, kernel_size, stride, padding, dilation, ceil_mode); +} + +// aten::quantized_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!) +inline at::Tensor & quantized_max_pool2d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, at::IntArrayRef dilation=1, bool ceil_mode=false) { + return at::_ops::quantized_max_pool2d_out::call(self, kernel_size, stride, padding, dilation, ceil_mode, out); +} +// aten::quantized_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!) +inline at::Tensor & quantized_max_pool2d_outf(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, at::Tensor & out) { + return at::_ops::quantized_max_pool2d_out::call(self, kernel_size, stride, padding, dilation, ceil_mode, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/repeat_interleave_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/repeat_interleave_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..b594cefe7f6fc3fbd79f3b5663df75e0125f20e0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/repeat_interleave_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// 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 + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & repeat_interleave_out(at::Tensor & out, const at::Tensor & repeats, c10::optional output_size=c10::nullopt); +TORCH_API at::Tensor & repeat_interleave_outf(const at::Tensor & repeats, c10::optional output_size, at::Tensor & out); +TORCH_API at::Tensor & repeat_interleave_symint_out(at::Tensor & out, const at::Tensor & repeats, c10::optional output_size=c10::nullopt); +TORCH_API at::Tensor & repeat_interleave_symint_outf(const at::Tensor & repeats, c10::optional output_size, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rrelu_with_noise_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rrelu_with_noise_native.h new file mode 100644 index 0000000000000000000000000000000000000000..013e9cbc07e3dcfa0c2ca6acd8f6b3cbf34b6150 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rrelu_with_noise_native.h @@ -0,0 +1,26 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor rrelu_with_noise_cpu(const at::Tensor & self, const at::Tensor & noise, const at::Scalar & lower=0.125, const at::Scalar & upper=0.3333333333333333, bool training=false, c10::optional generator=c10::nullopt); +TORCH_API at::Tensor & rrelu_with_noise_out_cpu(const at::Tensor & self, const at::Tensor & noise, const at::Scalar & lower, const at::Scalar & upper, bool training, c10::optional generator, at::Tensor & out); +TORCH_API at::Tensor & rrelu_with_noise_cpu_(at::Tensor & self, const at::Tensor & noise, const at::Scalar & lower=0.125, const at::Scalar & upper=0.3333333333333333, bool training=false, c10::optional generator=c10::nullopt); +TORCH_API at::Tensor rrelu_with_noise_cuda(const at::Tensor & self, const at::Tensor & noise, const at::Scalar & lower=0.125, const at::Scalar & upper=0.3333333333333333, bool training=false, c10::optional generator=c10::nullopt); +TORCH_API at::Tensor & rrelu_with_noise_out_cuda(const at::Tensor & self, const at::Tensor & noise, const at::Scalar & lower, const at::Scalar & upper, bool training, c10::optional generator, at::Tensor & out); +TORCH_API at::Tensor & rrelu_with_noise_cuda_(at::Tensor & self, const at::Tensor & noise, const at::Scalar & lower=0.125, const at::Scalar & upper=0.3333333333333333, bool training=false, c10::optional generator=c10::nullopt); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/slice_copy_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/slice_copy_native.h new file mode 100644 index 0000000000000000000000000000000000000000..9103fb16eac2fa4474e21cb843bc646dc87b536b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/slice_copy_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & slice_copy_Tensor_out_symint(const at::Tensor & self, int64_t dim, c10::optional start, c10::optional end, c10::SymInt step, at::Tensor & out); +TORCH_API at::Tensor slice_copy_Tensor_symint(const at::Tensor & self, int64_t dim=0, c10::optional start=c10::nullopt, c10::optional end=c10::nullopt, c10::SymInt step=1); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_erfcx_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_erfcx_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..6081da479a861a371eb8e4590dae7f12f96cc295 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_erfcx_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor special_erfcx(const at::Tensor & self); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_erfinv_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_erfinv_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..f013747c084c3db2399f9e117554b5319bebca87 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_erfinv_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor special_erfinv(const at::Tensor & self); +TORCH_API at::Tensor & special_erfinv_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & special_erfinv_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_expit_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_expit_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a942097b4f90295027b67145b06d4207cc292794 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_expit_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor special_expit(const at::Tensor & self); +TORCH_API at::Tensor & special_expit_out(const at::Tensor & self, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_i0e_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_i0e_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..6de4593cf0007f4aec805b0ad9b9539f829899b3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_i0e_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_special_i0e : public TensorIteratorBase { + + + void meta(const at::Tensor & self); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_log_ndtr_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_log_ndtr_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..37eb5d53d7c05ed26a683676d6fe6aa4be741937 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_log_ndtr_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 +#include +#include + +// 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 + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor special_log_ndtr(const at::Tensor & self); +TORCH_API at::Tensor & special_log_ndtr_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & special_log_ndtr_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_log_softmax_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_log_softmax_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8b64eac2e58168dcc25814fa7eb617b4ffcc007f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_log_softmax_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor special_log_softmax(const at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_log_softmax_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_log_softmax_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..cd9e2631a9abd488377c53455833605c44eb26b1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_log_softmax_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API special_log_softmax { + using schema = at::Tensor (const at::Tensor &, int64_t, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_log_softmax") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_log_softmax(Tensor self, int dim, *, ScalarType? dtype=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t dim, c10::optional dtype); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, c10::optional dtype); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/squeeze.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/squeeze.h new file mode 100644 index 0000000000000000000000000000000000000000..158498ff957cbd42c55901ca6a02a4a087c50d12 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/squeeze.h @@ -0,0 +1,45 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::squeeze(Tensor(a) self) -> Tensor(a) +inline at::Tensor squeeze(const at::Tensor & self) { + return at::_ops::squeeze::call(self); +} + +// aten::squeeze.dim(Tensor(a) self, int dim) -> Tensor(a) +inline at::Tensor squeeze(const at::Tensor & self, int64_t dim) { + return at::_ops::squeeze_dim::call(self, dim); +} + +// aten::squeeze.dimname(Tensor(a) self, Dimname dim) -> Tensor(a) +inline at::Tensor squeeze(const at::Tensor & self, at::Dimname dim) { + return at::_ops::squeeze_dimname::call(self, dim); +} + +// aten::squeeze.dims(Tensor(a) self, int[] dim) -> Tensor(a) +inline at::Tensor squeeze(const at::Tensor & self, at::IntArrayRef dim) { + return at::_ops::squeeze_dims::call(self, dim); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/svd_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/svd_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..7ded1579c82c630e6e7c4ce60e063f4e32795844 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/svd_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API ::std::tuple svd(const at::Tensor & self, bool some=true, bool compute_uv=true); +TORCH_API ::std::tuple svd_out(at::Tensor & U, at::Tensor & S, at::Tensor & V, const at::Tensor & self, bool some=true, bool compute_uv=true); +TORCH_API ::std::tuple svd_outf(const at::Tensor & self, bool some, bool compute_uv, at::Tensor & U, at::Tensor & S, at::Tensor & V); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_sparse_bsr.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_sparse_bsr.h new file mode 100644 index 0000000000000000000000000000000000000000..2b33f665f5496a570e6431d238c4e70dc17b0b95 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_sparse_bsr.h @@ -0,0 +1,26 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/type_as_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/type_as_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..7c3f58b82ca0ad2b82bb4589277ab1d4711729c7 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/type_as_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 +#include +#include + +// 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 + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor type_as(const at::Tensor & self, const at::Tensor & other); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bicubic2d_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bicubic2d_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..221872a001b6832700f49be9d4ecc7f6a684fcb4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bicubic2d_ops.h @@ -0,0 +1,50 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// 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 + +namespace at { +namespace _ops { + + +struct TORCH_API upsample_bicubic2d_vec { + using schema = at::Tensor (const at::Tensor &, at::OptionalSymIntArrayRef, bool, c10::optional>); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::upsample_bicubic2d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "vec") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "upsample_bicubic2d.vec(Tensor input, SymInt[]? output_size, bool align_corners, float[]? scale_factors) -> Tensor") + static at::Tensor call(const at::Tensor & input, at::OptionalSymIntArrayRef output_size, bool align_corners, c10::optional> scale_factors); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, at::OptionalSymIntArrayRef output_size, bool align_corners, c10::optional> scale_factors); +}; + +struct TORCH_API upsample_bicubic2d_out { + using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, bool, c10::optional, c10::optional, 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") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "upsample_bicubic2d.out(Tensor self, SymInt[2] output_size, bool align_corners, float? scales_h=None, float? scales_w=None, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional scales_h, c10::optional scales_w, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional scales_h, c10::optional scales_w, at::Tensor & out); +}; + +struct TORCH_API upsample_bicubic2d { + using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef, bool, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::upsample_bicubic2d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "upsample_bicubic2d(Tensor self, SymInt[2] output_size, bool align_corners, float? scales_h=None, float? scales_w=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional scales_h, c10::optional scales_w); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional scales_h, c10::optional scales_w); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_linear1d_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_linear1d_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..9610b6b562cb275c46b5993099a1c27af1a78100 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_linear1d_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_upsample_linear1d : public at::impl::MetaBase { + + + void meta(const at::Tensor & self, at::ArrayRef output_size, bool align_corners, c10::optional scales); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest2d_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest2d_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..601466059a93230790b0f68959d6c4458719b891 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest2d_backward_cuda_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// 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 + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor upsample_nearest2d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor upsample_nearest2d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_nearest2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_nearest2d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); +TORCH_API at::Tensor & upsample_nearest2d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_nearest2d_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest2d_backward_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest2d_backward_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..5cc58b1d564821edb0ebc56699d97ae907edaa1c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest2d_backward_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_upsample_nearest2d_backward : public at::impl::MetaBase { + + + void meta(const at::Tensor & grad_output, at::ArrayRef output_size, at::ArrayRef input_size, c10::optional scales_h, c10::optional scales_w); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest3d_backward_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest3d_backward_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..e8f6bcb9bc213c4041229dab1f52265425214907 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest3d_backward_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_upsample_nearest3d_backward : public at::impl::MetaBase { + + + void meta(const at::Tensor & grad_output, at::ArrayRef output_size, at::ArrayRef input_size, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w); +}; + +} // namespace native +} // namespace at