diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_adaptive_avg_pool2d_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_adaptive_avg_pool2d_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..e1d4a4ef7e55ed850f5f768e7469f32c91038019 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_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 _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::_adaptive_avg_pool2d_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_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 _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::_adaptive_avg_pool2d_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_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/_cast_Byte_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cast_Byte_native.h new file mode 100644 index 0000000000000000000000000000000000000000..387b1266e845d0d55de89caae7c516384b5f379b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cast_Byte_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 _cast_Byte(const at::Tensor & self, bool non_blocking=false); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_coalesced_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_coalesced_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..eddef49ceb921a59f26806a645dbc09e12237dc3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_coalesced_meta_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 meta { + +TORCH_API at::Tensor & _coalesced_(at::Tensor & self, bool coalesced); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fake_quantize_learnable_per_channel_affine_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fake_quantize_learnable_per_channel_affine_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c3b7af9484fd7f48a4cb60df71b3f8aa6a3e3994 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fake_quantize_learnable_per_channel_affine_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 _fake_quantize_learnable_per_channel_affine(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis, int64_t quant_min, int64_t quant_max, double grad_factor=1.0); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foobar_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foobar_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..25dc56eef7aae040d6f74657b80c4394a95c478f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foobar_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 at::Tensor & _foobar_out(at::Tensor & out, const at::Tensor & self, bool arg1=true, bool arg2=true, bool arg3=true); +TORCH_API at::Tensor & _foobar_outf(const at::Tensor & self, bool arg1, bool arg2, bool arg3, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_copy_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_copy_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..d3c2a7cd7f718f9295f8421f7c0aba5904f49bdb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_copy_compositeexplicitautograd_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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::vector _foreach_copy(at::TensorList self, at::TensorList src, bool non_blocking=false); +TORCH_API void _foreach_copy_out(at::TensorList out, at::TensorList self, at::TensorList src, bool non_blocking=false); +TORCH_API void _foreach_copy_outf(at::TensorList self, at::TensorList src, bool non_blocking, at::TensorList out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_lerp_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_lerp_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..c51274a7ae94d01774f3997df525d8c9a5399f41 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_lerp_ops.h @@ -0,0 +1,83 @@ +#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_lerp_List { + using schema = ::std::vector (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_lerp") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "List") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_lerp.List(Tensor[] self, Tensor[] tensors1, Tensor[] weights) -> Tensor[]") + static ::std::vector call(at::TensorList self, at::TensorList tensors1, at::TensorList weights); + static ::std::vector redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList tensors1, at::TensorList weights); +}; + +struct TORCH_API _foreach_lerp__List { + 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_lerp_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "List") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_lerp_.List(Tensor(a!)[] self, Tensor[] tensors1, Tensor[] weights) -> ()") + static void call(at::TensorList self, at::TensorList tensors1, at::TensorList weights); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList tensors1, at::TensorList weights); +}; + +struct TORCH_API _foreach_lerp_Scalar { + using schema = ::std::vector (at::TensorList, 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_lerp") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_lerp.Scalar(Tensor[] self, Tensor[] tensors1, Scalar weight) -> Tensor[]") + static ::std::vector call(at::TensorList self, at::TensorList tensors1, const at::Scalar & weight); + static ::std::vector redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList tensors1, const at::Scalar & weight); +}; + +struct TORCH_API _foreach_lerp__Scalar { + using schema = void (at::TensorList, 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_lerp_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_lerp_.Scalar(Tensor(a!)[] self, Tensor[] tensors1, Scalar weight) -> ()") + static void call(at::TensorList self, at::TensorList tensors1, const at::Scalar & weight); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList tensors1, const at::Scalar & weight); +}; + +struct TORCH_API _foreach_lerp_List_out { + using schema = void (at::TensorList, 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_lerp") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "List_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_lerp.List_out(Tensor[] self, Tensor[] tensors1, Tensor[] weights, *, Tensor(a!)[] out) -> ()") + static void call(at::TensorList self, at::TensorList tensors1, at::TensorList weights, at::TensorList out); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList tensors1, at::TensorList weights, at::TensorList out); +}; + +struct TORCH_API _foreach_lerp_Scalar_out { + using schema = void (at::TensorList, 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_lerp") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_lerp.Scalar_out(Tensor[] self, Tensor[] tensors1, Scalar weight, *, Tensor(a!)[] out) -> ()") + static void call(at::TensorList self, at::TensorList tensors1, const at::Scalar & weight, at::TensorList out); + static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList tensors1, const at::Scalar & weight, at::TensorList out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_round_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_round_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..1489ff82f1350f141f1e37da34e71bad9e3cd27f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_round_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 void _foreach_round_out(at::TensorList out, at::TensorList self); +TORCH_API void _foreach_round_outf(at::TensorList self, at::TensorList out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sigmoid_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sigmoid_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..587b84510696096c8b2bb50fb4619ca59736ab2f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sigmoid_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::vector _foreach_sigmoid(at::TensorList self); +TORCH_API void _foreach_sigmoid_(at::TensorList self); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_softmax_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_softmax_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..e125441d32ff5a72d7f5d03b52587c050520dd41 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_softmax_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 _masked_softmax(const at::Tensor & self, const at::Tensor & mask, c10::optional dim=c10::nullopt, c10::optional mask_type=c10::nullopt); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_softmax_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_softmax_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..c149d0d391aadd24de6bc30683a07c83d3d7e368 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_softmax_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 _masked_softmax { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, 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::_masked_softmax") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_masked_softmax(Tensor self, Tensor mask, int? dim=None, int? mask_type=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & mask, c10::optional dim, c10::optional mask_type); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & mask, c10::optional dim, c10::optional mask_type); +}; + +struct TORCH_API _masked_softmax_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, 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::_masked_softmax") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_masked_softmax.out(Tensor self, Tensor mask, int? dim=None, int? mask_type=None, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & mask, c10::optional dim, c10::optional mask_type, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & mask, c10::optional dim, c10::optional mask_type, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_saturate_weight_to_fp16_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_saturate_weight_to_fp16_native.h new file mode 100644 index 0000000000000000000000000000000000000000..04f029c77f8af446a192e6ca0da2cd652fcfff6b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_saturate_weight_to_fp16_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 _saturate_weight_to_fp16(const at::Tensor & weight); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_flash_attention_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_flash_attention_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..d28883466a43fbb11b015bcee519acb310ca5cfb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_flash_attention_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 _scaled_dot_product_flash_attention { + using schema = ::std::tuple (const at::Tensor &, const at::Tensor &, const at::Tensor &, double, bool, 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::_scaled_dot_product_flash_attention") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_scaled_dot_product_flash_attention(Tensor query, Tensor key, Tensor value, float dropout_p=0.0, bool is_causal=False, bool return_debug_mask=False, *, float? scale=None) -> (Tensor output, Tensor logsumexp, Tensor cum_seq_q, Tensor cum_seq_k, SymInt max_q, SymInt max_k, Tensor philox_seed, Tensor philox_offset, Tensor debug_attn_mask)") + static ::std::tuple call(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, double dropout_p, bool is_causal, bool return_debug_mask, c10::optional scale); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, double dropout_p, bool is_causal, bool return_debug_mask, c10::optional scale); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data.h new file mode 100644 index 0000000000000000000000000000000000000000..2d749a1878b16c88e5251f633f122f65c4ff84ac --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data.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::_softmax_backward_data(Tensor grad_output, Tensor output, int dim, ScalarType input_dtype) -> Tensor +inline at::Tensor _softmax_backward_data(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype) { + return at::_ops::_softmax_backward_data::call(grad_output, output, dim, input_dtype); +} + +// aten::_softmax_backward_data.out(Tensor grad_output, Tensor output, int dim, ScalarType input_dtype, *, Tensor(a!) grad_input) -> Tensor(a!) +inline 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) { + return at::_ops::_softmax_backward_data_out::call(grad_output, output, dim, input_dtype, grad_input); +} +// aten::_softmax_backward_data.out(Tensor grad_output, Tensor output, int dim, ScalarType input_dtype, *, Tensor(a!) grad_input) -> Tensor(a!) +inline 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) { + return at::_ops::_softmax_backward_data_out::call(grad_output, output, dim, input_dtype, grad_input); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..7981ce6d643a79f94ec909015e57133deaa66514 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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(const at::Tensor & self, int64_t dim, bool half_to_float); +TORCH_API at::Tensor & _softmax_out(at::Tensor & out, const at::Tensor & self, int64_t dim, bool half_to_float); +TORCH_API at::Tensor & _softmax_outf(const at::Tensor & self, int64_t dim, bool half_to_float, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_bsc_tensor_unsafe.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_bsc_tensor_unsafe.h new file mode 100644 index 0000000000000000000000000000000000000000..471e746a396be957541eedd2ce0984115a32750a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_bsc_tensor_unsafe.h @@ -0,0 +1,34 @@ +#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::_sparse_bsc_tensor_unsafe(Tensor ccol_indices, Tensor row_indices, Tensor values, int[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor _sparse_bsc_tensor_unsafe(const at::Tensor & ccol_indices, const at::Tensor & row_indices, const at::Tensor & values, at::IntArrayRef size, at::TensorOptions options={}) { + return at::_ops::_sparse_bsc_tensor_unsafe::call(ccol_indices, row_indices, values, size, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt()); +} +// aten::_sparse_bsc_tensor_unsafe(Tensor ccol_indices, Tensor row_indices, Tensor values, int[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor _sparse_bsc_tensor_unsafe(const at::Tensor & ccol_indices, const at::Tensor & row_indices, const at::Tensor & values, at::IntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory) { + return at::_ops::_sparse_bsc_tensor_unsafe::call(ccol_indices, row_indices, values, size, dtype, layout, device, pin_memory); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_autograd_multiple_dispatch_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_autograd_multiple_dispatch_native.h new file mode 100644 index 0000000000000000000000000000000000000000..c6104a5ca54fc296939305b4cdc9349b94b48005 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_autograd_multiple_dispatch_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 _test_autograd_multiple_dispatch_fullcoverage(const at::Tensor & self); +TORCH_API at::Tensor & _test_autograd_multiple_dispatch_fullcoverage_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor _test_autograd_multiple_dispatch_ntonly(const at::Tensor & self, bool b); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bilinear2d_aa_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bilinear2d_aa_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..e139801693bb40534468e6913afea195aa40b8fd --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bilinear2d_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_bilinear2d_aa(const at::Tensor & input, at::OptionalIntArrayRef output_size, bool align_corners, c10::optional> scale_factors); +TORCH_API at::Tensor _upsample_bilinear2d_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/_upsample_nearest_exact3d_backward_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact3d_backward_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a86678db6f48a426b884e9cbea8381a81f635d00 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact3d_backward_cpu_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 _upsample_nearest_exact3d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor _upsample_nearest_exact3d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & _upsample_nearest_exact3d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & _upsample_nearest_exact3d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); +TORCH_API at::Tensor & _upsample_nearest_exact3d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & _upsample_nearest_exact3d_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_validate_sparse_compressed_tensor_args_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_validate_sparse_compressed_tensor_args_native.h new file mode 100644 index 0000000000000000000000000000000000000000..2f9d0441ea8602dd59eb24b5f49a9220a6aad07d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_validate_sparse_compressed_tensor_args_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 void _validate_sparse_compressed_tensor_args(const at::Tensor & compressed_indices, const at::Tensor & plain_indices, const at::Tensor & values, at::IntArrayRef size, at::Layout layout); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_version_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_version_native.h new file mode 100644 index 0000000000000000000000000000000000000000..c6daa50be4658317a28f6cde9ed32804da3f762f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_version_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 int64_t _version(const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/abs_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/abs_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0257e11bd0ce2b1d4712d6cd0c4b018110457583 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/abs_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 at::Tensor & abs_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & abs_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/acos_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/acos_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..69b6f73e8566b7b889ba56626dbb7bb41fa415ef --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/acos_cuda_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 acos(const at::Tensor & self); +TORCH_API at::Tensor & acos_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & acos_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & acos_(at::Tensor & self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addcmul.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addcmul.h new file mode 100644 index 0000000000000000000000000000000000000000..cdc12f1b3cc11f833f5461bd30046424eec80c55 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addcmul.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::addcmul.out(Tensor self, Tensor tensor1, Tensor tensor2, *, Scalar value=1, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & addcmul_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & tensor1, const at::Tensor & tensor2, const at::Scalar & value=1) { + return at::_ops::addcmul_out::call(self, tensor1, tensor2, value, out); +} +// aten::addcmul.out(Tensor self, Tensor tensor1, Tensor tensor2, *, Scalar value=1, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & addcmul_outf(const at::Tensor & self, const at::Tensor & tensor1, const at::Tensor & tensor2, const at::Scalar & value, at::Tensor & out) { + return at::_ops::addcmul_out::call(self, tensor1, tensor2, value, out); +} + +// aten::addcmul(Tensor self, Tensor tensor1, Tensor tensor2, *, Scalar value=1) -> Tensor +inline at::Tensor addcmul(const at::Tensor & self, const at::Tensor & tensor1, const at::Tensor & tensor2, const at::Scalar & value=1) { + return at::_ops::addcmul::call(self, tensor1, tensor2, value); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/amax_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/amax_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..15333c3c84b3b1c4ecade8802c529033ae9a7697 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/amax_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 amax(const at::Tensor & self, at::IntArrayRef dim={}, bool keepdim=false); +TORCH_API at::Tensor & amax_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef dim={}, bool keepdim=false); +TORCH_API at::Tensor & amax_outf(const at::Tensor & self, at::IntArrayRef dim, bool keepdim, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/angle_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/angle_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..26d6316b282cb69626ea9aba5b4a35034cca41b0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/angle_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 angle { + 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::angle") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "angle(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 angle_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::angle") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "angle.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/arctan_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/arctan_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..0d764bbb313fcd0b36a5e66ba9da3b2db77d13d0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/arctan_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 arctan { + 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::arctan") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "arctan(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 arctan_ { + 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::arctan_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "arctan_(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 arctan_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::arctan") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "arctan.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/ccol_indices.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ccol_indices.h new file mode 100644 index 0000000000000000000000000000000000000000..c6e736c0685b4f97158da6202defae8f44472963 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ccol_indices.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/clamp_min_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/clamp_min_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..4cdf87ce9bafb2f679216cbe14dbe050ad2a0db7 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/clamp_min_meta.h @@ -0,0 +1,32 @@ +#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_clamp_min : public TensorIteratorBase { + + + void meta(const at::Tensor & self, const at::Scalar & min); +}; +struct TORCH_API structured_clamp_min_Tensor : public TensorIteratorBase { + + + void meta(const at::Tensor & self, const at::Tensor & min); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/clip.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/clip.h new file mode 100644 index 0000000000000000000000000000000000000000..6188cf90ae364d429be56d8648316140196cbdbf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/clip.h @@ -0,0 +1,63 @@ +#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::clip(Tensor self, Scalar? min=None, Scalar? max=None) -> Tensor +inline at::Tensor clip(const at::Tensor & self, const c10::optional & min, const c10::optional & max=c10::nullopt) { + return at::_ops::clip::call(self, min, max); +} + +// aten::clip.Tensor(Tensor self, Tensor? min=None, Tensor? max=None) -> Tensor +inline at::Tensor clip(const at::Tensor & self, const c10::optional & min={}, const c10::optional & max={}) { + return at::_ops::clip_Tensor::call(self, min, max); +} + +// aten::clip_(Tensor(a!) self, Scalar? min=None, Scalar? max=None) -> Tensor(a!) +inline at::Tensor & clip_(at::Tensor & self, const c10::optional & min, const c10::optional & max=c10::nullopt) { + return at::_ops::clip_::call(self, min, max); +} + +// aten::clip_.Tensor(Tensor(a!) self, Tensor? min=None, Tensor? max=None) -> Tensor(a!) +inline at::Tensor & clip_(at::Tensor & self, const c10::optional & min={}, const c10::optional & max={}) { + return at::_ops::clip__Tensor::call(self, min, max); +} + +// aten::clip.out(Tensor self, Scalar? min=None, Scalar? max=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & clip_out(at::Tensor & out, const at::Tensor & self, const c10::optional & min, const c10::optional & max=c10::nullopt) { + return at::_ops::clip_out::call(self, min, max, out); +} +// aten::clip.out(Tensor self, Scalar? min=None, Scalar? max=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & clip_outf(const at::Tensor & self, const c10::optional & min, const c10::optional & max, at::Tensor & out) { + return at::_ops::clip_out::call(self, min, max, out); +} + +// aten::clip.Tensor_out(Tensor self, Tensor? min=None, Tensor? max=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & clip_out(at::Tensor & out, const at::Tensor & self, const c10::optional & min={}, const c10::optional & max={}) { + return at::_ops::clip_Tensor_out::call(self, min, max, out); +} +// aten::clip.Tensor_out(Tensor self, Tensor? min=None, Tensor? max=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & clip_outf(const at::Tensor & self, const c10::optional & min, const c10::optional & max, at::Tensor & out) { + return at::_ops::clip_Tensor_out::call(self, min, max, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_backward.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..55c538a66fbd3a91f03cd2f66d7597f4c8345327 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_backward.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::cudnn_batch_norm_backward(Tensor input, Tensor grad_output, Tensor weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_var, float epsilon, Tensor reserveSpace) -> (Tensor, Tensor, Tensor) +inline ::std::tuple cudnn_batch_norm_backward(const at::Tensor & input, const at::Tensor & grad_output, const at::Tensor & weight, const c10::optional & running_mean, const c10::optional & running_var, const c10::optional & save_mean, const c10::optional & save_var, double epsilon, const at::Tensor & reserveSpace) { + return at::_ops::cudnn_batch_norm_backward::call(input, grad_output, weight, running_mean, running_var, save_mean, save_var, epsilon, reserveSpace); +} + +// aten::cudnn_batch_norm_backward.out(Tensor input, Tensor grad_output, Tensor weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_var, float epsilon, Tensor reserveSpace, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!)) +inline ::std::tuple cudnn_batch_norm_backward_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, const at::Tensor & input, const at::Tensor & grad_output, const at::Tensor & weight, const c10::optional & running_mean, const c10::optional & running_var, const c10::optional & save_mean, const c10::optional & save_var, double epsilon, const at::Tensor & reserveSpace) { + return at::_ops::cudnn_batch_norm_backward_out::call(input, grad_output, weight, running_mean, running_var, save_mean, save_var, epsilon, reserveSpace, out0, out1, out2); +} +// aten::cudnn_batch_norm_backward.out(Tensor input, Tensor grad_output, Tensor weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_var, float epsilon, Tensor reserveSpace, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!)) +inline ::std::tuple cudnn_batch_norm_backward_outf(const at::Tensor & input, const at::Tensor & grad_output, const at::Tensor & weight, const c10::optional & running_mean, const c10::optional & running_var, const c10::optional & save_mean, const c10::optional & save_var, double epsilon, const at::Tensor & reserveSpace, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2) { + return at::_ops::cudnn_batch_norm_backward_out::call(input, grad_output, weight, running_mean, running_var, save_mean, save_var, epsilon, reserveSpace, out0, out1, out2); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..5e018c22f13f5629534c8c13405fc971d02d3358 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_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 cudnn_batch_norm { + using schema = ::std::tuple (const at::Tensor &, const at::Tensor &, const c10::optional &, const c10::optional &, const c10::optional &, bool, double, double); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cudnn_batch_norm") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cudnn_batch_norm(Tensor input, Tensor weight, Tensor? bias, Tensor? running_mean, Tensor? running_var, bool training, float exponential_average_factor, float epsilon) -> (Tensor, Tensor, Tensor, Tensor)") + static ::std::tuple call(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, const c10::optional & running_mean, const c10::optional & running_var, bool training, double exponential_average_factor, double epsilon); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, const c10::optional & running_mean, const c10::optional & running_var, bool training, double exponential_average_factor, double epsilon); +}; + +struct TORCH_API cudnn_batch_norm_out { + using schema = ::std::tuple (const at::Tensor &, const at::Tensor &, const c10::optional &, const c10::optional &, const c10::optional &, bool, double, double, at::Tensor &, at::Tensor &, at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cudnn_batch_norm") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cudnn_batch_norm.out(Tensor input, Tensor weight, Tensor? bias, Tensor? running_mean, Tensor? running_var, bool training, float exponential_average_factor, float epsilon, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2, Tensor(d!) out3) -> (Tensor(a!), Tensor(b!), Tensor(c!), Tensor(d!))") + static ::std::tuple call(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, const c10::optional & running_mean, const c10::optional & running_var, bool training, double exponential_average_factor, double epsilon, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, const c10::optional & running_mean, const c10::optional & running_var, bool training, double exponential_average_factor, double epsilon, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diag_embed.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diag_embed.h new file mode 100644 index 0000000000000000000000000000000000000000..c98363e9a18cc9fc16448af85e2790af3b1c3933 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diag_embed.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::diag_embed(Tensor self, int offset=0, int dim1=-2, int dim2=-1) -> Tensor +inline at::Tensor diag_embed(const at::Tensor & self, int64_t offset=0, int64_t dim1=-2, int64_t dim2=-1) { + return at::_ops::diag_embed::call(self, offset, dim1, dim2); +} + +// aten::diag_embed.out(Tensor self, int offset=0, int dim1=-2, int dim2=-1, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & diag_embed_out(at::Tensor & out, const at::Tensor & self, int64_t offset=0, int64_t dim1=-2, int64_t dim2=-1) { + return at::_ops::diag_embed_out::call(self, offset, dim1, dim2, out); +} +// aten::diag_embed.out(Tensor self, int offset=0, int dim1=-2, int dim2=-1, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & diag_embed_outf(const at::Tensor & self, int64_t offset, int64_t dim1, int64_t dim2, at::Tensor & out) { + return at::_ops::diag_embed_out::call(self, offset, dim1, dim2, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/expm1_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/expm1_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8976a19ebc551f29df02a57471452849be779f17 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/expm1_cuda_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 expm1(const at::Tensor & self); +TORCH_API at::Tensor & expm1_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & expm1_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & expm1_(at::Tensor & self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_linear_quantize_weight_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_linear_quantize_weight_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..e6429268828a20df855cb21b4782128dffd181de --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_linear_quantize_weight_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 fbgemm_linear_quantize_weight { + using schema = ::std::tuple (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::fbgemm_linear_quantize_weight") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fbgemm_linear_quantize_weight(Tensor input) -> (Tensor, Tensor, float, int)") + static ::std::tuple call(const at::Tensor & input); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_pack_gemm_matrix_fp16_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_pack_gemm_matrix_fp16_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..cd9ab325220f69f555ad9807b15208e62ebf16a0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_pack_gemm_matrix_fp16_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 fbgemm_pack_gemm_matrix_fp16(const at::Tensor & input); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/floor_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/floor_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..531c2bb8583d2be35c35b2334e2aace8b3487008 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/floor_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 floor { + 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::floor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "floor(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 floor_ { + 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::floor_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "floor_(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 floor_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::floor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "floor.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/fmin_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fmin_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..b7c1a54fe39fcfd062131299d3fc28a37372cd6a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fmin_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 fmin(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & fmin_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & fmin_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fmod_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fmod_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a93af795ba6130deb5c016110ec8d823fe028b72 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fmod_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 fmod(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & fmod_(at::Tensor & self, const at::Tensor & other); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool2d_backward_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool2d_backward_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..64561c87b0da88a0458f9b0a07963a82b9da4b23 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool2d_backward_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 fractional_max_pool2d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices); +TORCH_API at::Tensor & fractional_max_pool2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices); +TORCH_API at::Tensor & fractional_max_pool2d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices, at::Tensor & grad_input); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ca4f68c6950cc5a8a594522ba278599177839b30 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_backward_cuda_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 hardshrink_backward(const at::Tensor & grad_out, const at::Tensor & self, const at::Scalar & lambd); +TORCH_API at::Tensor & hardshrink_backward_out(at::Tensor & grad_input, const at::Tensor & grad_out, const at::Tensor & self, const at::Scalar & lambd); +TORCH_API at::Tensor & hardshrink_backward_outf(const at::Tensor & grad_out, const at::Tensor & self, const at::Scalar & lambd, at::Tensor & grad_input); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/heaviside.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/heaviside.h new file mode 100644 index 0000000000000000000000000000000000000000..d276c070bc059328346f2a6257a0df810f58dafd --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/heaviside.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::heaviside.out(Tensor self, Tensor values, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & heaviside_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & values) { + return at::_ops::heaviside_out::call(self, values, out); +} +// aten::heaviside.out(Tensor self, Tensor values, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & heaviside_outf(const at::Tensor & self, const at::Tensor & values, at::Tensor & out) { + return at::_ops::heaviside_out::call(self, values, out); +} + +// aten::heaviside(Tensor self, Tensor values) -> Tensor +inline at::Tensor heaviside(const at::Tensor & self, const at::Tensor & values) { + return at::_ops::heaviside::call(self, values); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_select_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_select_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ac1dcb68bfe88a267b5014a6268e988d2234b421 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_select_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 index_select(const at::Tensor & self, at::Dimname dim, const at::Tensor & index); +TORCH_API at::Tensor & index_select_out(at::Tensor & out, const at::Tensor & self, at::Dimname dim, const at::Tensor & index); +TORCH_API at::Tensor & index_select_outf(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_neg_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_neg_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..bbdbdc6ce135ae86576da68c1e2512245cf717c8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_neg_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 is_neg { + using schema = bool (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::is_neg") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "is_neg(Tensor self) -> bool") + static bool call(const at::Tensor & self); + static bool redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isfinite_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isfinite_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..b84a8eacdea28f4ebb03b86fc87489cc6a04fa20 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isfinite_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 isfinite { + 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::isfinite") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "isfinite(Tensor self) -> Tensor") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/layer_norm.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/layer_norm.h new file mode 100644 index 0000000000000000000000000000000000000000..835ad3010a4379ad0e404521e53ef90567e983a5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/layer_norm.h @@ -0,0 +1,47 @@ +#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::layer_norm(Tensor input, SymInt[] normalized_shape, Tensor? weight=None, Tensor? bias=None, float eps=1e-05, bool cudnn_enable=True) -> Tensor +inline at::Tensor layer_norm(const at::Tensor & input, at::IntArrayRef normalized_shape, const c10::optional & weight={}, const c10::optional & bias={}, double eps=1e-05, bool cudnn_enable=true) { + return at::_ops::layer_norm::call(input, c10::fromIntArrayRefSlow(normalized_shape), weight, bias, eps, cudnn_enable); +} +namespace symint { + template ::value>> + at::Tensor layer_norm(const at::Tensor & input, at::IntArrayRef normalized_shape, const c10::optional & weight={}, const c10::optional & bias={}, double eps=1e-05, bool cudnn_enable=true) { + return at::_ops::layer_norm::call(input, c10::fromIntArrayRefSlow(normalized_shape), weight, bias, eps, cudnn_enable); + } +} + +// aten::layer_norm(Tensor input, SymInt[] normalized_shape, Tensor? weight=None, Tensor? bias=None, float eps=1e-05, bool cudnn_enable=True) -> Tensor +inline at::Tensor layer_norm_symint(const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const c10::optional & weight={}, const c10::optional & bias={}, double eps=1e-05, bool cudnn_enable=true) { + return at::_ops::layer_norm::call(input, normalized_shape, weight, bias, eps, cudnn_enable); +} +namespace symint { + template ::value>> + at::Tensor layer_norm(const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const c10::optional & weight={}, const c10::optional & bias={}, double eps=1e-05, bool cudnn_enable=true) { + return at::_ops::layer_norm::call(input, normalized_shape, weight, bias, eps, cudnn_enable); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/layer_norm_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/layer_norm_native.h new file mode 100644 index 0000000000000000000000000000000000000000..60fcbf89c529384210c020bdab3d6416f2ac3d3f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/layer_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 layer_norm_symint(const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const c10::optional & weight={}, const c10::optional & bias={}, double eps=1e-05, bool cudnn_enable=true); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logaddexp2.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logaddexp2.h new file mode 100644 index 0000000000000000000000000000000000000000..5279724c29b5f06c058cf7c25ba40f9ebdcee599 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logaddexp2.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::logaddexp2.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & logaddexp2_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) { + return at::_ops::logaddexp2_out::call(self, other, out); +} +// aten::logaddexp2.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & logaddexp2_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) { + return at::_ops::logaddexp2_out::call(self, other, out); +} + +// aten::logaddexp2(Tensor self, Tensor other) -> Tensor +inline at::Tensor logaddexp2(const at::Tensor & self, const at::Tensor & other) { + return at::_ops::logaddexp2::call(self, other); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logaddexp_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logaddexp_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..389d86be5c7256139462484651ad9811c8af20d5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logaddexp_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 logaddexp(const at::Tensor & self, const at::Tensor & other); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logcumsumexp_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logcumsumexp_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..6986d398400cfa66a2b6a647e0ce2b32926f65bd --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logcumsumexp_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 logcumsumexp(const at::Tensor & self, int64_t dim); +TORCH_API at::Tensor & logcumsumexp_out(at::Tensor & out, const at::Tensor & self, int64_t dim); +TORCH_API at::Tensor & logcumsumexp_outf(const at::Tensor & self, int64_t dim, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_scatter_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_scatter_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..52225f1698ab5f577980a103947dc9577c883cc0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_scatter_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 masked_scatter(const at::Tensor & self, const at::Tensor & mask, const at::Tensor & source); +TORCH_API at::Tensor & masked_scatter_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & mask, const at::Tensor & source); +TORCH_API at::Tensor & masked_scatter_outf(const at::Tensor & self, const at::Tensor & mask, const at::Tensor & source, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/min_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/min_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a7d972ba66aab4338481312c15d58c47860c6a28 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/min_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 ::std::tuple min(const at::Tensor & self, int64_t dim, bool keepdim=false); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mish_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mish_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..c13221a4d2c513ce31a0bdf35d2a49184175845d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mish_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 mish { + 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::mish") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mish(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 mish_ { + 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::mish_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mish_(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 mish_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::mish") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mish.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/ne_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ne_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..81257f73d74f27bbe76f7367de7fc05fdc848742 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ne_cpu_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 cpu { + +TORCH_API at::Tensor ne(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & ne_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & ne_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +TORCH_API at::Tensor & ne_(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor ne(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & ne_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & ne_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & ne_(at::Tensor & self, const at::Tensor & other); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/new_empty_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/new_empty_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..a14fb2b106bb7c4e60dc3a832ea72601937b142d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/new_empty_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 new_empty { + using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef, c10::optional, c10::optional, 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::new_empty") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "new_empty(Tensor self, SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +}; + +struct TORCH_API new_empty_out { + using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::new_empty") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "new_empty.out(Tensor self, SymInt[] size, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, c10::SymIntArrayRef size, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef size, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/q_per_channel_scales_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/q_per_channel_scales_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..f460708a4d7ca77f65855ad4d665f27c4ea15167 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/q_per_channel_scales_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 at::Tensor & q_per_channel_scales_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & q_per_channel_scales_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/quantized_rnn_relu_cell_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_rnn_relu_cell_native.h new file mode 100644 index 0000000000000000000000000000000000000000..37c0e2272be6abd25ff16aa432fbdbcf9eb052f0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_rnn_relu_cell_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 quantized_rnn_relu_cell(const at::Tensor & input, const at::Tensor & hx, const at::Tensor & w_ih, const at::Tensor & w_hh, const at::Tensor & b_ih, const at::Tensor & b_hh, const at::Tensor & packed_ih, const at::Tensor & packed_hh, const at::Tensor & col_offsets_ih, const at::Tensor & col_offsets_hh, const at::Scalar & scale_ih, const at::Scalar & scale_hh, const at::Scalar & zero_point_ih, const at::Scalar & zero_point_hh); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rename_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rename_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..9df3d9d7205b065ac37dadd80c1fa2788beb2290 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rename_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 & rename_(at::Tensor & self, c10::optional names); +TORCH_API at::Tensor rename(const at::Tensor & self, c10::optional names); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/scatter_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/scatter_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..9294ba854b56891a7c8e1361eca34bdc238896ee --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/scatter_meta_dispatch.h @@ -0,0 +1,38 @@ +#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 scatter(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src); +TORCH_API at::Tensor & scatter_out(at::Tensor & out, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src); +TORCH_API at::Tensor & scatter_outf(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, at::Tensor & out); +TORCH_API at::Tensor & scatter_(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src); +TORCH_API at::Tensor scatter(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value); +TORCH_API at::Tensor & scatter_out(at::Tensor & out, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value); +TORCH_API at::Tensor & scatter_outf(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, at::Tensor & out); +TORCH_API at::Tensor & scatter_(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value); +TORCH_API at::Tensor scatter(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce); +TORCH_API at::Tensor & scatter_out(at::Tensor & out, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce); +TORCH_API at::Tensor & scatter_outf(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce, at::Tensor & out); +TORCH_API at::Tensor & scatter_(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & src, c10::string_view reduce); +TORCH_API at::Tensor scatter(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce); +TORCH_API at::Tensor & scatter_out(at::Tensor & out, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce); +TORCH_API at::Tensor & scatter_outf(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce, at::Tensor & out); +TORCH_API at::Tensor & scatter_(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, c10::string_view reduce); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softplus_backward_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softplus_backward_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..364e38c4040f1051292ddebd5dff11dc2348a2bf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softplus_backward_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 softplus_backward(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & beta, const at::Scalar & threshold); +TORCH_API at::Tensor & softplus_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & beta, const at::Scalar & threshold); +TORCH_API at::Tensor & softplus_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & beta, const at::Scalar & threshold, at::Tensor & grad_input); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_csr_tensor_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_csr_tensor_native.h new file mode 100644 index 0000000000000000000000000000000000000000..2534d71a06bb4aa788d85edb5e0c5abbcf923adb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_csr_tensor_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_csr_tensor(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={}); +TORCH_API at::Tensor sparse_csr_tensor(const at::Tensor & crow_indices, const at::Tensor & col_indices, const at::Tensor & values, c10::optional dtype={}, c10::optional layout={}, c10::optional device={}, c10::optional pin_memory={}); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_bessel_j0_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_bessel_j0_native.h new file mode 100644 index 0000000000000000000000000000000000000000..bc8e07f3ff5e81f9e6c5a4b829152630170d2b8c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_bessel_j0_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 +#include + +namespace at { +namespace native { +struct TORCH_API structured_special_bessel_j0_out : public at::meta::structured_special_bessel_j0 { +void impl(const at::Tensor & self, const at::Tensor & out); +}; +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_i0.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_i0.h new file mode 100644 index 0000000000000000000000000000000000000000..16fc251ce90401ffaf36d7a9128c99d896f52e1b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_i0.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::special_i0(Tensor self) -> Tensor +inline at::Tensor special_i0(const at::Tensor & self) { + return at::_ops::special_i0::call(self); +} + +// aten::special_i0.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & special_i0_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::special_i0_out::call(self, out); +} +// aten::special_i0.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & special_i0_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::special_i0_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_logit_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_logit_native.h new file mode 100644 index 0000000000000000000000000000000000000000..d0870996dcb34312e8ca882f368f6d804da25523 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_logit_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_logit(const at::Tensor & self, c10::optional eps=c10::nullopt); +TORCH_API at::Tensor & special_logit_out(const at::Tensor & self, c10::optional eps, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/svd_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/svd_native.h new file mode 100644 index 0000000000000000000000000000000000000000..14de68f13909fa878efc7ef949dac533db036668 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/svd_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 ::std::tuple svd(const at::Tensor & self, bool some=true, bool compute_uv=true); +TORCH_API ::std::tuple svd_out(const at::Tensor & self, bool some, bool compute_uv, at::Tensor & U, at::Tensor & S, at::Tensor & V); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/tanh_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/tanh_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..3f5ab73bf80ea349e9b73f09fe9b85866340797e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/tanh_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 tanh_backward_grad_input { + 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::tanh_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "tanh_backward.grad_input(Tensor grad_output, Tensor output, *, Tensor(a!) grad_input) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & output, at::Tensor & grad_input); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & output, at::Tensor & grad_input); +}; + +struct TORCH_API tanh_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::tanh_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "tanh_backward(Tensor grad_output, Tensor output) -> Tensor") + static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & output); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & output); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/tensor_split.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/tensor_split.h new file mode 100644 index 0000000000000000000000000000000000000000..dbe39dda0b935f51311a9aaa270fdcfaf6c1465d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/tensor_split.h @@ -0,0 +1,74 @@ +#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::tensor_split.sections(Tensor(a -> *) self, SymInt sections, int dim=0) -> Tensor(a)[] +inline ::std::vector tensor_split(const at::Tensor & self, int64_t sections, int64_t dim=0) { + return at::_ops::tensor_split_sections::call(self, sections, dim); +} +namespace symint { + template ::value>> + ::std::vector tensor_split(const at::Tensor & self, int64_t sections, int64_t dim=0) { + return at::_ops::tensor_split_sections::call(self, sections, dim); + } +} + +// aten::tensor_split.sections(Tensor(a -> *) self, SymInt sections, int dim=0) -> Tensor(a)[] +inline ::std::vector tensor_split_symint(const at::Tensor & self, c10::SymInt sections, int64_t dim=0) { + return at::_ops::tensor_split_sections::call(self, sections, dim); +} +namespace symint { + template ::value>> + ::std::vector tensor_split(const at::Tensor & self, c10::SymInt sections, int64_t dim=0) { + return at::_ops::tensor_split_sections::call(self, sections, dim); + } +} + +// aten::tensor_split.indices(Tensor(a -> *) self, SymInt[] indices, int dim=0) -> Tensor(a)[] +inline ::std::vector tensor_split(const at::Tensor & self, at::IntArrayRef indices, int64_t dim=0) { + return at::_ops::tensor_split_indices::call(self, c10::fromIntArrayRefSlow(indices), dim); +} +namespace symint { + template ::value>> + ::std::vector tensor_split(const at::Tensor & self, at::IntArrayRef indices, int64_t dim=0) { + return at::_ops::tensor_split_indices::call(self, c10::fromIntArrayRefSlow(indices), dim); + } +} + +// aten::tensor_split.indices(Tensor(a -> *) self, SymInt[] indices, int dim=0) -> Tensor(a)[] +inline ::std::vector tensor_split_symint(const at::Tensor & self, c10::SymIntArrayRef indices, int64_t dim=0) { + return at::_ops::tensor_split_indices::call(self, indices, dim); +} +namespace symint { + template ::value>> + ::std::vector tensor_split(const at::Tensor & self, c10::SymIntArrayRef indices, int64_t dim=0) { + return at::_ops::tensor_split_indices::call(self, indices, dim); + } +} + +// aten::tensor_split.tensor_indices_or_sections(Tensor(a -> *) self, Tensor tensor_indices_or_sections, int dim=0) -> Tensor(a)[] +inline ::std::vector tensor_split(const at::Tensor & self, const at::Tensor & tensor_indices_or_sections, int64_t dim=0) { + return at::_ops::tensor_split_tensor_indices_or_sections::call(self, tensor_indices_or_sections, dim); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/trace_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/trace_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..108adcdabf633c78fa8191faadf2edea6a4fd1b2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/trace_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 at::Tensor & trace_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & trace_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/trace_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/trace_native.h new file mode 100644 index 0000000000000000000000000000000000000000..ebf39b21a19df482822ba0ac03675b386f5fc082 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/trace_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 & trace_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor trace_cpu(const at::Tensor & self); +TORCH_API at::Tensor trace_cuda(const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/triangular_solve_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/triangular_solve_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..615866269f1ff66aae727d7e4c392bf1d2c18a05 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/triangular_solve_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 ::std::tuple triangular_solve(const at::Tensor & self, const at::Tensor & A, bool upper=true, bool transpose=false, bool unitriangular=false); +TORCH_API ::std::tuple triangular_solve_out(at::Tensor & X, at::Tensor & M, const at::Tensor & self, const at::Tensor & A, bool upper=true, bool transpose=false, bool unitriangular=false); +TORCH_API ::std::tuple triangular_solve_outf(const at::Tensor & self, const at::Tensor & A, bool upper, bool transpose, bool unitriangular, at::Tensor & X, at::Tensor & M); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unfold_backward.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unfold_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..38be86e2559e8a0409d6994f52f96f42333bbaae --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unfold_backward.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::unfold_backward(Tensor grad_in, SymInt[] input_sizes, int dim, int size, int step) -> Tensor +inline at::Tensor unfold_backward(const at::Tensor & grad_in, at::IntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step) { + return at::_ops::unfold_backward::call(grad_in, c10::fromIntArrayRefSlow(input_sizes), dim, size, step); +} +namespace symint { + template ::value>> + at::Tensor unfold_backward(const at::Tensor & grad_in, at::IntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step) { + return at::_ops::unfold_backward::call(grad_in, c10::fromIntArrayRefSlow(input_sizes), dim, size, step); + } +} + +// aten::unfold_backward(Tensor grad_in, SymInt[] input_sizes, int dim, int size, int step) -> Tensor +inline at::Tensor unfold_backward_symint(const at::Tensor & grad_in, c10::SymIntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step) { + return at::_ops::unfold_backward::call(grad_in, input_sizes, dim, size, step); +} +namespace symint { + template ::value>> + at::Tensor unfold_backward(const at::Tensor & grad_in, c10::SymIntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step) { + return at::_ops::unfold_backward::call(grad_in, input_sizes, dim, size, step); + } +} + +// aten::unfold_backward.out(Tensor grad_in, SymInt[] input_sizes, int dim, int size, int step, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & unfold_backward_out(at::Tensor & out, const at::Tensor & grad_in, at::IntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step) { + return at::_ops::unfold_backward_out::call(grad_in, c10::fromIntArrayRefSlow(input_sizes), dim, size, step, out); +} +namespace symint { + template ::value>> + at::Tensor & unfold_backward_out(at::Tensor & out, const at::Tensor & grad_in, at::IntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step) { + return at::_ops::unfold_backward_out::call(grad_in, c10::fromIntArrayRefSlow(input_sizes), dim, size, step, out); + } +} + +// aten::unfold_backward.out(Tensor grad_in, SymInt[] input_sizes, int dim, int size, int step, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & unfold_backward_outf(const at::Tensor & grad_in, at::IntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step, at::Tensor & out) { + return at::_ops::unfold_backward_out::call(grad_in, c10::fromIntArrayRefSlow(input_sizes), dim, size, step, out); +} +namespace symint { + template ::value>> + at::Tensor & unfold_backward_outf(const at::Tensor & grad_in, at::IntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step, at::Tensor & out) { + return at::_ops::unfold_backward_out::call(grad_in, c10::fromIntArrayRefSlow(input_sizes), dim, size, step, out); + } +} + +// aten::unfold_backward.out(Tensor grad_in, SymInt[] input_sizes, int dim, int size, int step, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & unfold_backward_symint_out(at::Tensor & out, const at::Tensor & grad_in, c10::SymIntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step) { + return at::_ops::unfold_backward_out::call(grad_in, input_sizes, dim, size, step, out); +} +namespace symint { + template ::value>> + at::Tensor & unfold_backward_out(at::Tensor & out, const at::Tensor & grad_in, c10::SymIntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step) { + return at::_ops::unfold_backward_out::call(grad_in, input_sizes, dim, size, step, out); + } +} + +// aten::unfold_backward.out(Tensor grad_in, SymInt[] input_sizes, int dim, int size, int step, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & unfold_backward_symint_outf(const at::Tensor & grad_in, c10::SymIntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step, at::Tensor & out) { + return at::_ops::unfold_backward_out::call(grad_in, input_sizes, dim, size, step, out); +} +namespace symint { + template ::value>> + at::Tensor & unfold_backward_outf(const at::Tensor & grad_in, c10::SymIntArrayRef input_sizes, int64_t dim, int64_t size, int64_t step, at::Tensor & out) { + return at::_ops::unfold_backward_out::call(grad_in, input_sizes, dim, size, step, out); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_consecutive.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_consecutive.h new file mode 100644 index 0000000000000000000000000000000000000000..b92a6628e36d527a6ed60b5546c9eebdeeadbc18 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_consecutive.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::unique_consecutive(Tensor self, bool return_inverse=False, bool return_counts=False, int? dim=None) -> (Tensor, Tensor, Tensor) +inline ::std::tuple unique_consecutive(const at::Tensor & self, bool return_inverse=false, bool return_counts=false, c10::optional dim=c10::nullopt) { + return at::_ops::unique_consecutive::call(self, return_inverse, return_counts, dim); +} + +// aten::unique_consecutive.out(Tensor self, bool return_inverse=False, bool return_counts=False, int? dim=None, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!)) +inline ::std::tuple unique_consecutive_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, const at::Tensor & self, bool return_inverse=false, bool return_counts=false, c10::optional dim=c10::nullopt) { + return at::_ops::unique_consecutive_out::call(self, return_inverse, return_counts, dim, out0, out1, out2); +} +// aten::unique_consecutive.out(Tensor self, bool return_inverse=False, bool return_counts=False, int? dim=None, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!)) +inline ::std::tuple unique_consecutive_outf(const at::Tensor & self, bool return_inverse, bool return_counts, c10::optional dim, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2) { + return at::_ops::unique_consecutive_out::call(self, return_inverse, return_counts, dim, out0, out1, out2); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_dim_consecutive_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_dim_consecutive_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..d2b8343b2e7680ae60a861812715f8a291f2a85c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_dim_consecutive_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 unique_dim_consecutive { + using schema = ::std::tuple (const at::Tensor &, int64_t, bool, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::unique_dim_consecutive") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "unique_dim_consecutive(Tensor self, int dim, bool return_inverse=False, bool return_counts=False) -> (Tensor, Tensor, Tensor)") + static ::std::tuple call(const at::Tensor & self, int64_t dim, bool return_inverse, bool return_counts); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, bool return_inverse, bool return_counts); +}; + +struct TORCH_API unique_dim_consecutive_out { + using schema = ::std::tuple (const at::Tensor &, int64_t, bool, bool, at::Tensor &, at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::unique_dim_consecutive") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "unique_dim_consecutive.out(Tensor self, int dim, bool return_inverse=False, bool return_counts=False, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!))") + static ::std::tuple call(const at::Tensor & self, int64_t dim, bool return_inverse, bool return_counts, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, bool return_inverse, bool return_counts, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bicubic2d_backward_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bicubic2d_backward_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a6a9ce211b1ac204977768d6efd8606643a02f99 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bicubic2d_backward_cpu_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 upsample_bicubic2d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor upsample_bicubic2d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_bicubic2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_bicubic2d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); +TORCH_API at::Tensor & upsample_bicubic2d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_bicubic2d_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/zero_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/zero_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..99c49dbe1ae1cb8a9d8d8274ca2c1497784d2fa9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/zero_meta_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 meta { + +TORCH_API at::Tensor & zero_(at::Tensor & self); + +} // namespace meta +} // namespace at