diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_flash_attention_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_flash_attention_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..d30703ae48b2ca53de43771d3183d6bf922eda16 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_flash_attention_backward_cuda_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple _flash_attention_backward(const at::Tensor & grad_out, const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const at::Tensor & out, const at::Tensor & logsumexp, const at::Tensor & cum_seq_q, const at::Tensor & cum_seq_k, int64_t max_q, int64_t max_k, double dropout_p, bool is_causal, const at::Tensor & philox_seed, const at::Tensor & philox_offset, c10::optional scale=c10::nullopt); +TORCH_API ::std::tuple _flash_attention_backward_symint(const at::Tensor & grad_out, const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const at::Tensor & out, const at::Tensor & logsumexp, const at::Tensor & cum_seq_q, const at::Tensor & cum_seq_k, c10::SymInt max_q, c10::SymInt max_k, double dropout_p, bool is_causal, const at::Tensor & philox_seed, const at::Tensor & philox_offset, c10::optional scale=c10::nullopt); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_div_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_div_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..9ceb296f2bb7b878d768780460d43e5f31bbf2a6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_div_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 ::std::vector _foreach_div(at::TensorList self, const at::Scalar & scalar); +TORCH_API void _foreach_div_(at::TensorList self, const at::Scalar & scalar); +TORCH_API ::std::vector _foreach_div(at::TensorList self, at::TensorList other); +TORCH_API void _foreach_div_(at::TensorList self, at::TensorList other); +TORCH_API ::std::vector _foreach_div(at::TensorList self, at::ArrayRef scalars); +TORCH_API void _foreach_div_(at::TensorList self, at::ArrayRef scalars); +TORCH_API ::std::vector _foreach_div(at::TensorList self, const at::Tensor & other); +TORCH_API void _foreach_div_(at::TensorList 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/_foreach_lgamma_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_lgamma_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..064f3feb66c1b880818414998e84e2a38a192ad3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_lgamma_cuda_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::vector _foreach_lgamma(at::TensorList self); +TORCH_API void _foreach_lgamma_(at::TensorList self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sin.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sin.h new file mode 100644 index 0000000000000000000000000000000000000000..db381272f0f99330d78e715179989a10b3eb443b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sin.h @@ -0,0 +1,44 @@ +#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::_foreach_sin(Tensor[] self) -> Tensor[] +inline ::std::vector _foreach_sin(at::TensorList self) { + return at::_ops::_foreach_sin::call(self); +} + +// aten::_foreach_sin_(Tensor(a!)[] self) -> () +inline void _foreach_sin_(at::TensorList self) { + return at::_ops::_foreach_sin_::call(self); +} + +// aten::_foreach_sin.out(Tensor[] self, *, Tensor(a!)[] out) -> () +inline void _foreach_sin_out(at::TensorList out, at::TensorList self) { + return at::_ops::_foreach_sin_out::call(self, out); +} +// aten::_foreach_sin.out(Tensor[] self, *, Tensor(a!)[] out) -> () +inline void _foreach_sin_outf(at::TensorList self, at::TensorList out) { + return at::_ops::_foreach_sin_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sqrt_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sqrt_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..63a1e65400988f524f074cdd9feb05414d5aeb6d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sqrt_cuda_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::vector _foreach_sqrt(at::TensorList self); +TORCH_API void _foreach_sqrt_(at::TensorList self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sub.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sub.h new file mode 100644 index 0000000000000000000000000000000000000000..a231c416add126a7b00c2b136fd48c340259234d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sub.h @@ -0,0 +1,82 @@ +#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::_foreach_sub.Scalar(Tensor[] self, Scalar scalar) -> Tensor[] +inline ::std::vector _foreach_sub(at::TensorList self, const at::Scalar & scalar) { + return at::_ops::_foreach_sub_Scalar::call(self, scalar); +} + +// aten::_foreach_sub_.Scalar(Tensor(a!)[] self, Scalar scalar) -> () +inline void _foreach_sub_(at::TensorList self, const at::Scalar & scalar) { + return at::_ops::_foreach_sub__Scalar::call(self, scalar); +} + +// aten::_foreach_sub.List(Tensor[] self, Tensor[] other, *, Scalar alpha=1) -> Tensor[] +inline ::std::vector _foreach_sub(at::TensorList self, at::TensorList other, const at::Scalar & alpha=1) { + return at::_ops::_foreach_sub_List::call(self, other, alpha); +} + +// aten::_foreach_sub_.List(Tensor(a!)[] self, Tensor[] other, *, Scalar alpha=1) -> () +inline void _foreach_sub_(at::TensorList self, at::TensorList other, const at::Scalar & alpha=1) { + return at::_ops::_foreach_sub__List::call(self, other, alpha); +} + +// aten::_foreach_sub.ScalarList(Tensor[] self, Scalar[] scalars) -> Tensor[] +inline ::std::vector _foreach_sub(at::TensorList self, at::ArrayRef scalars) { + return at::_ops::_foreach_sub_ScalarList::call(self, scalars); +} + +// aten::_foreach_sub_.ScalarList(Tensor(a!)[] self, Scalar[] scalars) -> () +inline void _foreach_sub_(at::TensorList self, at::ArrayRef scalars) { + return at::_ops::_foreach_sub__ScalarList::call(self, scalars); +} + +// aten::_foreach_sub.Scalar_out(Tensor[] self, Scalar scalar, *, Tensor(a!)[] out) -> () +inline void _foreach_sub_out(at::TensorList out, at::TensorList self, const at::Scalar & scalar) { + return at::_ops::_foreach_sub_Scalar_out::call(self, scalar, out); +} +// aten::_foreach_sub.Scalar_out(Tensor[] self, Scalar scalar, *, Tensor(a!)[] out) -> () +inline void _foreach_sub_outf(at::TensorList self, const at::Scalar & scalar, at::TensorList out) { + return at::_ops::_foreach_sub_Scalar_out::call(self, scalar, out); +} + +// aten::_foreach_sub.List_out(Tensor[] self, Tensor[] other, *, Scalar alpha=1, Tensor(a!)[] out) -> () +inline void _foreach_sub_out(at::TensorList out, at::TensorList self, at::TensorList other, const at::Scalar & alpha=1) { + return at::_ops::_foreach_sub_List_out::call(self, other, alpha, out); +} +// aten::_foreach_sub.List_out(Tensor[] self, Tensor[] other, *, Scalar alpha=1, Tensor(a!)[] out) -> () +inline void _foreach_sub_outf(at::TensorList self, at::TensorList other, const at::Scalar & alpha, at::TensorList out) { + return at::_ops::_foreach_sub_List_out::call(self, other, alpha, out); +} + +// aten::_foreach_sub.ScalarList_out(Tensor[] self, Scalar[] scalars, *, Tensor(a!)[] out) -> () +inline void _foreach_sub_out(at::TensorList out, at::TensorList self, at::ArrayRef scalars) { + return at::_ops::_foreach_sub_ScalarList_out::call(self, scalars, out); +} +// aten::_foreach_sub.ScalarList_out(Tensor[] self, Scalar[] scalars, *, Tensor(a!)[] out) -> () +inline void _foreach_sub_outf(at::TensorList self, at::ArrayRef scalars, at::TensorList out) { + return at::_ops::_foreach_sub_ScalarList_out::call(self, scalars, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_histogramdd_from_bin_cts_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_histogramdd_from_bin_cts_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..5a48c737ffb349b6d1c2a64cb7082336eb49d580 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_histogramdd_from_bin_cts_cpu_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor _histogramdd_from_bin_cts(const at::Tensor & self, at::IntArrayRef bins, c10::optional> range=c10::nullopt, const c10::optional & weight={}, bool density=false); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl.h new file mode 100644 index 0000000000000000000000000000000000000000..b5565b6f744900f94fee0b8196d114e3145a7a3e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl.h @@ -0,0 +1,44 @@ +#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::_index_put_impl_(Tensor(a!) self, Tensor?[] indices, Tensor values, bool accumulate=False, bool unsafe=False) -> Tensor(a!) +inline at::Tensor & _index_put_impl_(at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false) { + return at::_ops::_index_put_impl_::call(self, indices, values, accumulate, unsafe); +} + +// aten::_index_put_impl.out(Tensor self, Tensor?[] indices, Tensor values, bool accumulate=False, bool unsafe=False, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _index_put_impl_out(at::Tensor & out, const at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false) { + return at::_ops::_index_put_impl_out::call(self, indices, values, accumulate, unsafe, out); +} +// aten::_index_put_impl.out(Tensor self, Tensor?[] indices, Tensor values, bool accumulate=False, bool unsafe=False, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _index_put_impl_outf(const at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate, bool unsafe, at::Tensor & out) { + return at::_ops::_index_put_impl_out::call(self, indices, values, accumulate, unsafe, out); +} + +// aten::_index_put_impl(Tensor self, Tensor?[] indices, Tensor values, bool accumulate=False, bool unsafe=False) -> Tensor +inline at::Tensor _index_put_impl(const at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false) { + return at::_ops::_index_put_impl::call(self, indices, values, accumulate, unsafe); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_backward_data_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_backward_data_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..15d142633a32b3072b29d5f68e84b36d077b4b14 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_backward_data_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 _log_softmax_backward_data(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype); +TORCH_API at::Tensor & _log_softmax_backward_data_out(at::Tensor & out, const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype); +TORCH_API at::Tensor & _log_softmax_backward_data_outf(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype, at::Tensor & out); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_view_from_jagged_copy_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_view_from_jagged_copy_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..7db3a22ebe3da96ce714320016469daba3f55ecf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_view_from_jagged_copy_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API _nested_view_from_jagged_copy { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &, const c10::optional &, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_nested_view_from_jagged_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_view_from_jagged_copy(Tensor self, Tensor offsets, Tensor dummy, Tensor? lengths=None, int ragged_idx=1) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & offsets, const at::Tensor & dummy, const c10::optional & lengths, int64_t ragged_idx); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & offsets, const at::Tensor & dummy, const c10::optional & lengths, int64_t ragged_idx); +}; + +struct TORCH_API _nested_view_from_jagged_copy_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const at::Tensor &, const c10::optional &, int64_t, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_nested_view_from_jagged_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_view_from_jagged_copy.out(Tensor self, Tensor offsets, Tensor dummy, Tensor? lengths=None, int ragged_idx=1, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & offsets, const at::Tensor & dummy, const c10::optional & lengths, int64_t ragged_idx, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & offsets, const at::Tensor & dummy, const c10::optional & lengths, int64_t ragged_idx, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..f6b14c247d1f8b9c363ddeec0d762141876d8611 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured__softmax : public at::impl::MetaBase { + + + void meta(const at::Tensor & self, int64_t dim, bool half_to_float); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_broadcast_to_copy_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_broadcast_to_copy_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..74ebc67068920e2aea0a097392d9a143cb1a29ab --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_broadcast_to_copy_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 _sparse_broadcast_to_copy { + using schema = at::Tensor (const at::Tensor &, at::IntArrayRef); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_sparse_broadcast_to_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_sparse_broadcast_to_copy(Tensor self, int[] size) -> Tensor") + static at::Tensor call(const at::Tensor & self, at::IntArrayRef size); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::IntArrayRef size); +}; + +struct TORCH_API _sparse_broadcast_to_copy_out { + using schema = at::Tensor & (const at::Tensor &, at::IntArrayRef, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_sparse_broadcast_to_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_sparse_broadcast_to_copy.out(Tensor self, int[] size, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, at::IntArrayRef size, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::IntArrayRef size, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_transform_bias_rescale_qkv_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_transform_bias_rescale_qkv_native.h new file mode 100644 index 0000000000000000000000000000000000000000..17f78afef759a9eb8ce73a462d929fb8c8f30796 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_transform_bias_rescale_qkv_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 ::std::tuple _transform_bias_rescale_qkv_out(const at::Tensor & qkv, const at::Tensor & qkv_bias, int64_t num_heads, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); +TORCH_API ::std::tuple transform_bias_rescale_qkv_cpu(const at::Tensor & qkv, const at::Tensor & qkv_bias, int64_t num_heads); +TORCH_API ::std::tuple transform_bias_rescale_qkv_cuda(const at::Tensor & qkv, const at::Tensor & qkv_bias, int64_t num_heads); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_triton_scaled_dot_attention_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_triton_scaled_dot_attention_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..fec042e1effaa81f6c65e0b212f372e3a42a45e8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_triton_scaled_dot_attention_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 _triton_scaled_dot_attention { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &, double); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_triton_scaled_dot_attention") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_triton_scaled_dot_attention(Tensor q, Tensor k, Tensor v, float dropout_p=0.0) -> Tensor") + static at::Tensor call(const at::Tensor & q, const at::Tensor & k, const at::Tensor & v, double dropout_p); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & q, const at::Tensor & k, const at::Tensor & v, double dropout_p); +}; + +struct TORCH_API _triton_scaled_dot_attention_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const at::Tensor &, double, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_triton_scaled_dot_attention") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_triton_scaled_dot_attention.out(Tensor q, Tensor k, Tensor v, float dropout_p=0.0, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & q, const at::Tensor & k, const at::Tensor & v, double dropout_p, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & q, const at::Tensor & k, const at::Tensor & v, double dropout_p, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique2_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique2_native.h new file mode 100644 index 0000000000000000000000000000000000000000..e0d1d575c6003855bb1f1fe10f790b1384a13d3f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique2_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 ::std::tuple _unique2_out(const at::Tensor & self, bool sorted, bool return_inverse, bool return_counts, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); +TORCH_API ::std::tuple _unique2_cpu(const at::Tensor & self, bool sorted=true, bool return_inverse=false, bool return_counts=false); +TORCH_API ::std::tuple _unique2_cuda(const at::Tensor & self, bool sorted=true, bool return_inverse=false, bool return_counts=false); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm_interface_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm_interface_native.h new file mode 100644 index 0000000000000000000000000000000000000000..704afa7feae422312edd657f28d1c72bc4b838d9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm_interface_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 ::std::tuple _weight_norm_interface_out(const at::Tensor & v, const at::Tensor & g, int64_t dim, at::Tensor & out0, at::Tensor & out1); +TORCH_API ::std::tuple weight_norm_cpu(const at::Tensor & v, const at::Tensor & g, int64_t dim=0); +TORCH_API ::std::tuple weight_norm_cuda(const at::Tensor & v, const at::Tensor & g, int64_t dim=0); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..cf7247b97d6139d12b3fede11e78572df9abb290 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm_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 _weight_norm { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_weight_norm") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_weight_norm(Tensor v, Tensor g, int dim=0) -> Tensor") + static at::Tensor call(const at::Tensor & v, const at::Tensor & g, int64_t dim); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & v, const at::Tensor & g, int64_t dim); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_native.h new file mode 100644 index 0000000000000000000000000000000000000000..ac37410439c051dc81c231c8ed2688fc552ceef4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_native.h @@ -0,0 +1,24 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor adaptive_avg_pool2d_symint(const at::Tensor & self, c10::SymIntArrayRef output_size); +TORCH_API at::Tensor & adaptive_avg_pool2d_out_cpu(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out); +TORCH_API at::Tensor & adaptive_avg_pool2d_out_cuda(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out); +TORCH_API at::Tensor & mkldnn_adaptive_avg_pool2d_out_stub(const at::Tensor & self, at::IntArrayRef output_size, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/affine_grid_generator.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/affine_grid_generator.h new file mode 100644 index 0000000000000000000000000000000000000000..3c06cb2ca0849804337b848ed07ce882341dba3b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/affine_grid_generator.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::affine_grid_generator(Tensor theta, SymInt[] size, bool align_corners) -> Tensor +inline at::Tensor affine_grid_generator(const at::Tensor & theta, at::IntArrayRef size, bool align_corners) { + return at::_ops::affine_grid_generator::call(theta, c10::fromIntArrayRefSlow(size), align_corners); +} +namespace symint { + template ::value>> + at::Tensor affine_grid_generator(const at::Tensor & theta, at::IntArrayRef size, bool align_corners) { + return at::_ops::affine_grid_generator::call(theta, c10::fromIntArrayRefSlow(size), align_corners); + } +} + +// aten::affine_grid_generator(Tensor theta, SymInt[] size, bool align_corners) -> Tensor +inline at::Tensor affine_grid_generator_symint(const at::Tensor & theta, c10::SymIntArrayRef size, bool align_corners) { + return at::_ops::affine_grid_generator::call(theta, size, align_corners); +} +namespace symint { + template ::value>> + at::Tensor affine_grid_generator(const at::Tensor & theta, c10::SymIntArrayRef size, bool align_corners) { + return at::_ops::affine_grid_generator::call(theta, size, align_corners); + } +} + +// aten::affine_grid_generator.out(Tensor theta, SymInt[] size, bool align_corners, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & affine_grid_generator_out(at::Tensor & out, const at::Tensor & theta, at::IntArrayRef size, bool align_corners) { + return at::_ops::affine_grid_generator_out::call(theta, c10::fromIntArrayRefSlow(size), align_corners, out); +} +namespace symint { + template ::value>> + at::Tensor & affine_grid_generator_out(at::Tensor & out, const at::Tensor & theta, at::IntArrayRef size, bool align_corners) { + return at::_ops::affine_grid_generator_out::call(theta, c10::fromIntArrayRefSlow(size), align_corners, out); + } +} + +// aten::affine_grid_generator.out(Tensor theta, SymInt[] size, bool align_corners, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & affine_grid_generator_outf(const at::Tensor & theta, at::IntArrayRef size, bool align_corners, at::Tensor & out) { + return at::_ops::affine_grid_generator_out::call(theta, c10::fromIntArrayRefSlow(size), align_corners, out); +} +namespace symint { + template ::value>> + at::Tensor & affine_grid_generator_outf(const at::Tensor & theta, at::IntArrayRef size, bool align_corners, at::Tensor & out) { + return at::_ops::affine_grid_generator_out::call(theta, c10::fromIntArrayRefSlow(size), align_corners, out); + } +} + +// aten::affine_grid_generator.out(Tensor theta, SymInt[] size, bool align_corners, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & affine_grid_generator_symint_out(at::Tensor & out, const at::Tensor & theta, c10::SymIntArrayRef size, bool align_corners) { + return at::_ops::affine_grid_generator_out::call(theta, size, align_corners, out); +} +namespace symint { + template ::value>> + at::Tensor & affine_grid_generator_out(at::Tensor & out, const at::Tensor & theta, c10::SymIntArrayRef size, bool align_corners) { + return at::_ops::affine_grid_generator_out::call(theta, size, align_corners, out); + } +} + +// aten::affine_grid_generator.out(Tensor theta, SymInt[] size, bool align_corners, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & affine_grid_generator_symint_outf(const at::Tensor & theta, c10::SymIntArrayRef size, bool align_corners, at::Tensor & out) { + return at::_ops::affine_grid_generator_out::call(theta, size, align_corners, out); +} +namespace symint { + template ::value>> + at::Tensor & affine_grid_generator_outf(const at::Tensor & theta, c10::SymIntArrayRef size, bool align_corners, at::Tensor & out) { + return at::_ops::affine_grid_generator_out::call(theta, size, align_corners, out); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/and_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/and_native.h new file mode 100644 index 0000000000000000000000000000000000000000..b388a3d663227234abdbbdc0a9db6416c3986c55 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/and_native.h @@ -0,0 +1,24 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor __and__(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & __iand__(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor __and__(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & __iand__(at::Tensor & self, const at::Tensor & other); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/atan_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/atan_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..13e338cf621111171f64c08759e120d63299a392 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/atan_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 atan(const at::Tensor & self); +TORCH_API at::Tensor & atan_(at::Tensor & self); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bernoulli_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bernoulli_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..4460ac549c03a9cc1cba0fa7dddb8d8a38dd0863 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bernoulli_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 bernoulli(const at::Tensor & self, double p, c10::optional generator=c10::nullopt); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_and_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_and_native.h new file mode 100644 index 0000000000000000000000000000000000000000..bf50218d537b580470c5f5cd6530b3010f6dc276 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_and_native.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_bitwise_and_out : public at::meta::structured_bitwise_and_Tensor { +void impl(const at::Tensor & self, const at::Tensor & other, const at::Tensor & out); +}; +TORCH_API at::Tensor bitwise_and(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & bitwise_and_out(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +TORCH_API at::Tensor & bitwise_and_(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor bitwise_and(const at::Scalar & self, const at::Tensor & other); +TORCH_API at::Tensor & bitwise_and_Scalar_Tensor_out(const at::Scalar & self, const at::Tensor & other, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/col_indices_copy.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/col_indices_copy.h new file mode 100644 index 0000000000000000000000000000000000000000..fc919ed41bb7e376f36373f6b8371c03a162a0e6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/col_indices_copy.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::col_indices_copy(Tensor self) -> Tensor +inline at::Tensor col_indices_copy(const at::Tensor & self) { + return at::_ops::col_indices_copy::call(self); +} + +// aten::col_indices_copy.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & col_indices_copy_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::col_indices_copy_out::call(self, out); +} +// aten::col_indices_copy.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & col_indices_copy_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::col_indices_copy_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cosine_embedding_loss_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cosine_embedding_loss_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..3cbee1cc4ce9f61c043e85c988a655b2b6ac15b1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cosine_embedding_loss_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 cosine_embedding_loss { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &, double, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cosine_embedding_loss") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cosine_embedding_loss(Tensor input1, Tensor input2, Tensor target, float margin=0.0, int reduction=Mean) -> Tensor") + static at::Tensor call(const at::Tensor & input1, const at::Tensor & input2, const at::Tensor & target, double margin, int64_t reduction); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input1, const at::Tensor & input2, const at::Tensor & target, double margin, int64_t reduction); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_affine_grid_generator_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_affine_grid_generator_native.h new file mode 100644 index 0000000000000000000000000000000000000000..b30bbd728621cf7c5247294c0b368f9fed50802b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_affine_grid_generator_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 & cudnn_affine_grid_generator_out(const at::Tensor & theta, int64_t N, int64_t C, int64_t H, int64_t W, at::Tensor & out); +TORCH_API at::Tensor cudnn_affine_grid_generator_forward(const at::Tensor & theta, int64_t N, int64_t C, int64_t H, int64_t W); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..07937ca03e7cd598b3656f3ac57f2ffb51340fb2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_backward_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 cudnn_batch_norm_backward_out(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); +TORCH_API ::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); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_grid_sampler_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_grid_sampler_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a6d41c754e38b54ff0cc9275db35eb7fecdf69d0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_grid_sampler_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 & cudnn_grid_sampler_out(const at::Tensor & self, const at::Tensor & grid, at::Tensor & out); +TORCH_API at::Tensor cudnn_grid_sampler_forward(const at::Tensor & self, const at::Tensor & grid); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cummax.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cummax.h new file mode 100644 index 0000000000000000000000000000000000000000..f9bd56b6cd9d4523f32b0cdb8687a66f1718f925 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cummax.h @@ -0,0 +1,53 @@ +#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::cummax(Tensor self, int dim) -> (Tensor values, Tensor indices) +inline ::std::tuple cummax(const at::Tensor & self, int64_t dim) { + return at::_ops::cummax::call(self, dim); +} + +// aten::cummax.out(Tensor self, int dim, *, Tensor(a!) values, Tensor(b!) indices) -> (Tensor(a!) values, Tensor(b!) indices) +inline ::std::tuple cummax_out(at::Tensor & values, at::Tensor & indices, const at::Tensor & self, int64_t dim) { + return at::_ops::cummax_out::call(self, dim, values, indices); +} +// aten::cummax.out(Tensor self, int dim, *, Tensor(a!) values, Tensor(b!) indices) -> (Tensor(a!) values, Tensor(b!) indices) +inline ::std::tuple cummax_outf(const at::Tensor & self, int64_t dim, at::Tensor & values, at::Tensor & indices) { + return at::_ops::cummax_out::call(self, dim, values, indices); +} + +// aten::cummax.dimname(Tensor self, Dimname dim) -> (Tensor values, Tensor indices) +inline ::std::tuple cummax(const at::Tensor & self, at::Dimname dim) { + return at::_ops::cummax_dimname::call(self, dim); +} + +// aten::cummax.dimname_out(Tensor self, Dimname dim, *, Tensor(a!) values, Tensor(b!) indices) -> (Tensor(a!) values, Tensor(b!) indices) +inline ::std::tuple cummax_out(at::Tensor & values, at::Tensor & indices, const at::Tensor & self, at::Dimname dim) { + return at::_ops::cummax_dimname_out::call(self, dim, values, indices); +} +// aten::cummax.dimname_out(Tensor self, Dimname dim, *, Tensor(a!) values, Tensor(b!) indices) -> (Tensor(a!) values, Tensor(b!) indices) +inline ::std::tuple cummax_outf(const at::Tensor & self, at::Dimname dim, at::Tensor & values, at::Tensor & indices) { + return at::_ops::cummax_dimname_out::call(self, dim, values, indices); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cummax_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cummax_native.h new file mode 100644 index 0000000000000000000000000000000000000000..d0b5434db7e0a645061841e3775d545f25ca173d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cummax_native.h @@ -0,0 +1,24 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API ::std::tuple cummax(const at::Tensor & self, int64_t dim); +TORCH_API ::std::tuple cummax_out(const at::Tensor & self, int64_t dim, at::Tensor & values, at::Tensor & indices); +TORCH_API ::std::tuple cummax(const at::Tensor & self, at::Dimname dim); +TORCH_API ::std::tuple cummax_out(const at::Tensor & self, at::Dimname dim, at::Tensor & values, at::Tensor & indices); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..04b171804ab4414037a201d8d45d6b2d2d9947ac --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_meta_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 cumsum(const at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); +TORCH_API at::Tensor & cumsum_out(at::Tensor & out, const at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); +TORCH_API at::Tensor & cumsum_outf(const at::Tensor & self, int64_t dim, c10::optional dtype, at::Tensor & out); +TORCH_API at::Tensor & cumsum_(at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumulative_trapezoid_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumulative_trapezoid_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..b5b05288aef27f70e8f136858a1b721251da8b47 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumulative_trapezoid_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 cumulative_trapezoid_x { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cumulative_trapezoid") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "x") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cumulative_trapezoid.x(Tensor y, Tensor x, *, int dim=-1) -> Tensor") + static at::Tensor call(const at::Tensor & y, const at::Tensor & x, int64_t dim); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & y, const at::Tensor & x, int64_t dim); +}; + +struct TORCH_API cumulative_trapezoid_dx { + using schema = at::Tensor (const at::Tensor &, const at::Scalar &, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::cumulative_trapezoid") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dx") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cumulative_trapezoid.dx(Tensor y, *, Scalar dx=1, int dim=-1) -> Tensor") + static at::Tensor call(const at::Tensor & y, const at::Scalar & dx, int64_t dim); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & y, const at::Scalar & dx, int64_t dim); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagonal_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagonal_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..9d150fba82758fe5756c3e635548805b05ddd6ea --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagonal_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 diagonal { + using schema = at::Tensor (const at::Tensor &, int64_t, int64_t, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::diagonal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "diagonal(Tensor(a) self, int offset=0, int dim1=0, int dim2=1) -> Tensor(a)") + static at::Tensor call(const at::Tensor & self, int64_t offset, int64_t dim1, int64_t dim2); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t offset, int64_t dim1, int64_t dim2); +}; + +struct TORCH_API diagonal_Dimname { + using schema = at::Tensor (const at::Tensor &, at::Dimname, at::Dimname, at::Dimname, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::diagonal") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Dimname") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "diagonal.Dimname(Tensor(a) self, *, Dimname outdim, Dimname dim1, Dimname dim2, int offset=0) -> Tensor(a)") + static at::Tensor call(const at::Tensor & self, at::Dimname outdim, at::Dimname dim1, at::Dimname dim2, int64_t offset); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname outdim, at::Dimname dim1, at::Dimname dim2, int64_t offset); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diff_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diff_native.h new file mode 100644 index 0000000000000000000000000000000000000000..9413a8f7f7477e7c6f7b95135f06f35ca8ceb7e8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diff_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 diff(const at::Tensor & self, int64_t n=1, int64_t dim=-1, const c10::optional & prepend={}, const c10::optional & append={}); +TORCH_API at::Tensor & diff_out(const at::Tensor & self, int64_t n, int64_t dim, const c10::optional & prepend, const c10::optional & append, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/einsum_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/einsum_native.h new file mode 100644 index 0000000000000000000000000000000000000000..2c6a7d117ec1849812f03e4912a4f05f2d84b757 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/einsum_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 einsum(c10::string_view equation, at::TensorList tensors, at::OptionalIntArrayRef path=c10::nullopt); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fill_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fill_native.h new file mode 100644 index 0000000000000000000000000000000000000000..f178ffe51a2a47b6d248b0932803285b62c46eb6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fill_native.h @@ -0,0 +1,33 @@ +#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 fill(const at::Tensor & self, const at::Scalar & value); +TORCH_API at::Tensor & fill_Scalar_out(const at::Tensor & self, const at::Scalar & value, at::Tensor & out); +TORCH_API at::Tensor & fill_(at::Tensor & self, const at::Scalar & value); +TORCH_API at::Tensor & fill_nested_(at::Tensor & self, const at::Scalar & value); +TORCH_API at::Tensor & fill_sparse_csr_(at::Tensor & self, const at::Scalar & value); +TORCH_API at::Tensor & fill_meta_(at::Tensor & self, const at::Scalar & value); +TORCH_API at::Tensor & fill_quantized_(at::Tensor & self, const at::Scalar & value); +TORCH_API at::Tensor fill(const at::Tensor & self, const at::Tensor & value); +TORCH_API at::Tensor & fill_Tensor_out(const at::Tensor & self, const at::Tensor & value, at::Tensor & out); +TORCH_API at::Tensor & fill_(at::Tensor & self, const at::Tensor & value); +TORCH_API at::Tensor & fill_nested_(at::Tensor & self, const at::Tensor & value); +TORCH_API at::Tensor & fill_meta_(at::Tensor & self, const at::Tensor & value); +TORCH_API at::Tensor & fill_quantized_(at::Tensor & self, const at::Tensor & value); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fmod_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fmod_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..e910d990f42b6cbcf29f9505cd8b2e8bc06a8463 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fmod_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 fmod(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & fmod_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & fmod_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & fmod_(at::Tensor & self, const at::Tensor & other); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gelu_backward_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gelu_backward_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..6d18a50d910b76d5f6e7a7864a6908d35c6c8d2e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gelu_backward_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_gelu_backward : public TensorIteratorBase { + + + void meta(const at::Tensor & grad_output, const at::Tensor & self, c10::string_view approximate); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu.h new file mode 100644 index 0000000000000000000000000000000000000000..7d355f67c512a4596a0fdb80c9dfa390db36e3d4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu.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::glu.out(Tensor self, int dim=-1, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & glu_out(at::Tensor & out, const at::Tensor & self, int64_t dim=-1) { + return at::_ops::glu_out::call(self, dim, out); +} +// aten::glu.out(Tensor self, int dim=-1, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & glu_outf(const at::Tensor & self, int64_t dim, at::Tensor & out) { + return at::_ops::glu_out::call(self, dim, out); +} + +// aten::glu(Tensor self, int dim=-1) -> Tensor +inline at::Tensor glu(const at::Tensor & self, int64_t dim=-1) { + return at::_ops::glu::call(self, dim); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/group_norm_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/group_norm_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..2ca9137024842060accfd8f1e64d35928c92ac19 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/group_norm_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 group_norm { + using schema = at::Tensor (const at::Tensor &, int64_t, const c10::optional &, const c10::optional &, double, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::group_norm") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "group_norm(Tensor input, int num_groups, Tensor? weight=None, Tensor? bias=None, float eps=1e-05, bool cudnn_enabled=True) -> Tensor") + static at::Tensor call(const at::Tensor & input, int64_t num_groups, const c10::optional & weight, const c10::optional & bias, double eps, bool cudnn_enabled); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, int64_t num_groups, const c10::optional & weight, const c10::optional & bias, double eps, bool cudnn_enabled); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..9ea2ee18b4f10495bd44d2f6e581133d190aa61e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_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 hardshrink_out { + using schema = at::Tensor & (const at::Tensor &, const at::Scalar &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::hardshrink") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "hardshrink.out(Tensor self, Scalar lambd=0.5, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Scalar & lambd, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & lambd, at::Tensor & out); +}; + +struct TORCH_API hardshrink { + using schema = at::Tensor (const at::Tensor &, 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::hardshrink") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "hardshrink(Tensor self, Scalar lambd=0.5) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Scalar & lambd); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & lambd); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardtanh_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardtanh_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..38b0c66ba237f53279c6efc9faeb71ae75a4afd0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardtanh_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 hardtanh(const at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1); +TORCH_API at::Tensor & hardtanh_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1); +TORCH_API at::Tensor & hardtanh_outf(const at::Tensor & self, const at::Scalar & min_val, const at::Scalar & max_val, at::Tensor & out); +TORCH_API at::Tensor & hardtanh_(at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_add_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_add_native.h new file mode 100644 index 0000000000000000000000000000000000000000..6f2218d28f5fe31a028ba6a829478f1cb2588da8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_add_native.h @@ -0,0 +1,27 @@ +#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_index_add_cpu_out : public at::meta::structured_index_add { +void impl(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha, const at::Tensor & out); +}; +struct TORCH_API structured_index_add_cuda_out : public at::meta::structured_index_add { +void impl(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha, const at::Tensor & out); +}; +TORCH_API at::Tensor index_add(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha=1); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/inner_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/inner_native.h new file mode 100644 index 0000000000000000000000000000000000000000..c3dabf7ad9273918075bf7175f1d5767a8f4eebe --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/inner_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 inner(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & inner_out(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isinf.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isinf.h new file mode 100644 index 0000000000000000000000000000000000000000..9773e41211bbcd2dc44ea696e69c56463f90cb6f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isinf.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::isinf(Tensor self) -> Tensor +inline at::Tensor isinf(const at::Tensor & self) { + return at::_ops::isinf::call(self); +} + +// aten::isinf.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & isinf_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::isinf_out::call(self, out); +} +// aten::isinf.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & isinf_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::isinf_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isneginf_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isneginf_native.h new file mode 100644 index 0000000000000000000000000000000000000000..223aad6f7b37d034add411d904a49ebca57b8e43 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isneginf_native.h @@ -0,0 +1,27 @@ +#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_isneginf_out : public at::meta::structured_isneginf { +void impl(const at::Tensor & self, const at::Tensor & out); +}; +TORCH_API at::Tensor isneginf_sparse(const at::Tensor & self); +TORCH_API at::Tensor & isneginf_sparse_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor isneginf_sparse_csr(const at::Tensor & self); +TORCH_API at::Tensor & isneginf_sparse_csr_out(const at::Tensor & self, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/leaky_relu_backward_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/leaky_relu_backward_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..14e3fac8110b6794b296607b16382d98a2d80e87 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/leaky_relu_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 leaky_relu_backward(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result); +TORCH_API at::Tensor & leaky_relu_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result); +TORCH_API at::Tensor & leaky_relu_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result, at::Tensor & grad_input); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/leaky_relu_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/leaky_relu_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..ad2cb96cfb236417ae26f851cbac7143aba3c709 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/leaky_relu_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 leaky_relu_backward_grad_input { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const at::Scalar &, bool, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::leaky_relu_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "leaky_relu_backward.grad_input(Tensor grad_output, Tensor self, Scalar negative_slope, bool self_is_result, *, Tensor(a!) grad_input) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result, at::Tensor & grad_input); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result, at::Tensor & grad_input); +}; + +struct TORCH_API leaky_relu_backward { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Scalar &, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::leaky_relu_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "leaky_relu_backward(Tensor grad_output, Tensor self, Scalar negative_slope, bool self_is_result) -> Tensor") + static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Scalar & negative_slope, bool self_is_result); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/less_equal.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/less_equal.h new file mode 100644 index 0000000000000000000000000000000000000000..8c5623b6bd6de9ddccb0874338b3b4bd6bdbe4a1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/less_equal.h @@ -0,0 +1,53 @@ +#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::less_equal.Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & less_equal_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other) { + return at::_ops::less_equal_Scalar_out::call(self, other, out); +} +// aten::less_equal.Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & less_equal_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out) { + return at::_ops::less_equal_Scalar_out::call(self, other, out); +} + +// aten::less_equal.Scalar(Tensor self, Scalar other) -> Tensor +inline at::Tensor less_equal(const at::Tensor & self, const at::Scalar & other) { + return at::_ops::less_equal_Scalar::call(self, other); +} + +// aten::less_equal.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & less_equal_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) { + return at::_ops::less_equal_Tensor_out::call(self, other, out); +} +// aten::less_equal.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & less_equal_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) { + return at::_ops::less_equal_Tensor_out::call(self, other, out); +} + +// aten::less_equal.Tensor(Tensor self, Tensor other) -> Tensor +inline at::Tensor less_equal(const at::Tensor & self, const at::Tensor & other) { + return at::_ops::less_equal_Tensor::call(self, other); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_eigh.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_eigh.h new file mode 100644 index 0000000000000000000000000000000000000000..8a51c95e605ba492d6632e8d203bc49a32f430fb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_eigh.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::linalg_eigh(Tensor self, str UPLO="L") -> (Tensor eigenvalues, Tensor eigenvectors) +inline ::std::tuple linalg_eigh(const at::Tensor & self, c10::string_view UPLO="L") { + return at::_ops::linalg_eigh::call(self, UPLO); +} + +// aten::linalg_eigh.eigvals(Tensor self, str UPLO="L", *, Tensor(a!) eigvals, Tensor(b!) eigvecs) -> (Tensor(a!) eigenvalues, Tensor(b!) eigenvectors) +inline ::std::tuple linalg_eigh_out(at::Tensor & eigvals, at::Tensor & eigvecs, const at::Tensor & self, c10::string_view UPLO="L") { + return at::_ops::linalg_eigh_eigvals::call(self, UPLO, eigvals, eigvecs); +} +// aten::linalg_eigh.eigvals(Tensor self, str UPLO="L", *, Tensor(a!) eigvals, Tensor(b!) eigvecs) -> (Tensor(a!) eigenvalues, Tensor(b!) eigenvectors) +inline ::std::tuple linalg_eigh_outf(const at::Tensor & self, c10::string_view UPLO, at::Tensor & eigvals, at::Tensor & eigvecs) { + return at::_ops::linalg_eigh_eigvals::call(self, UPLO, eigvals, eigvecs); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_matrix_rank_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_matrix_rank_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a7d581b0f8f987a2a155999a73b5c0439d0b0fdb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_matrix_rank_compositeimplicitautograd_dispatch.h @@ -0,0 +1,34 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor linalg_matrix_rank(const at::Tensor & input, const c10::optional & atol={}, const c10::optional & rtol={}, bool hermitian=false); +TORCH_API at::Tensor & linalg_matrix_rank_out(at::Tensor & out, const at::Tensor & input, const c10::optional & atol={}, const c10::optional & rtol={}, bool hermitian=false); +TORCH_API at::Tensor & linalg_matrix_rank_outf(const at::Tensor & input, const c10::optional & atol, const c10::optional & rtol, bool hermitian, at::Tensor & out); +TORCH_API at::Tensor linalg_matrix_rank(const at::Tensor & self, c10::optional atol, c10::optional rtol, bool hermitian=false); +TORCH_API at::Tensor & linalg_matrix_rank_out(at::Tensor & out, const at::Tensor & self, c10::optional atol, c10::optional rtol, bool hermitian=false); +TORCH_API at::Tensor & linalg_matrix_rank_outf(const at::Tensor & self, c10::optional atol, c10::optional rtol, bool hermitian, at::Tensor & out); +TORCH_API at::Tensor linalg_matrix_rank(const at::Tensor & self, double tol, bool hermitian=false); +TORCH_API at::Tensor & linalg_matrix_rank_out(at::Tensor & out, const at::Tensor & self, double tol, bool hermitian=false); +TORCH_API at::Tensor & linalg_matrix_rank_outf(const at::Tensor & self, double tol, bool hermitian, at::Tensor & out); +TORCH_API at::Tensor linalg_matrix_rank(const at::Tensor & input, const at::Tensor & tol, bool hermitian=false); +TORCH_API at::Tensor & linalg_matrix_rank_out(at::Tensor & out, const at::Tensor & input, const at::Tensor & tol, bool hermitian=false); +TORCH_API at::Tensor & linalg_matrix_rank_outf(const at::Tensor & input, const at::Tensor & tol, bool hermitian, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_not_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_not_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..69344c6b589728d39ab5d4448e9cc27e80d1af19 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_not_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 & logical_not_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & logical_not_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/logical_not_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_not_native.h new file mode 100644 index 0000000000000000000000000000000000000000..5a40d9d029688d2c05f67c8730d96ec42b348203 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_not_native.h @@ -0,0 +1,25 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor logical_not(const at::Tensor & self); +TORCH_API at::Tensor & logical_not_(at::Tensor & self); +TORCH_API at::Tensor & logical_not_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor NestedTensor_logical_not(const at::Tensor & self); +TORCH_API at::Tensor & NestedTensor_logical_not_(at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..aee685b18fdc45b216a774bb09d44d2519bc713b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_backward_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 & max_pool2d_backward_out(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_batch_norm_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_batch_norm_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..00f567354f39da503a3ac3bf0780df4eaff2afe7 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_batch_norm_backward_cuda_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple miopen_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); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_rnn_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_rnn_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..81d81a29cc5352091e529087dd15f9253debbdaf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_rnn_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 miopen_rnn_backward { + using schema = ::std::tuple> (const at::Tensor &, at::TensorList, int64_t, const at::Tensor &, const at::Tensor &, const c10::optional &, const at::Tensor &, const c10::optional &, const c10::optional &, const c10::optional &, int64_t, int64_t, int64_t, bool, double, bool, bool, at::IntArrayRef, const c10::optional &, const at::Tensor &, ::std::array); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::miopen_rnn_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "miopen_rnn_backward(Tensor input, Tensor[] weight, int weight_stride0, Tensor weight_buf, Tensor hx, Tensor? cx, Tensor output, Tensor? grad_output, Tensor? grad_hy, Tensor? grad_cy, int mode, int hidden_size, int num_layers, bool batch_first, float dropout, bool train, bool bidirectional, int[] batch_sizes, Tensor? dropout_state, Tensor reserve, bool[4] output_mask) -> (Tensor, Tensor, Tensor, Tensor[])") + static ::std::tuple> call(const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const c10::optional & cx, const at::Tensor & output, const c10::optional & grad_output, const c10::optional & grad_hy, const c10::optional & grad_cy, int64_t mode, int64_t hidden_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, at::IntArrayRef batch_sizes, const c10::optional & dropout_state, const at::Tensor & reserve, ::std::array output_mask); + static ::std::tuple> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const c10::optional & cx, const at::Tensor & output, const c10::optional & grad_output, const c10::optional & grad_hy, const c10::optional & grad_cy, int64_t mode, int64_t hidden_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, at::IntArrayRef batch_sizes, const c10::optional & dropout_state, const at::Tensor & reserve, ::std::array output_mask); +}; + +struct TORCH_API miopen_rnn_backward_out { + using schema = void (const at::Tensor &, at::TensorList, int64_t, const at::Tensor &, const at::Tensor &, const c10::optional &, const at::Tensor &, const c10::optional &, const c10::optional &, const c10::optional &, int64_t, int64_t, int64_t, bool, double, bool, bool, at::IntArrayRef, const c10::optional &, const at::Tensor &, ::std::array, at::Tensor &, at::Tensor &, at::Tensor &, at::TensorList); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::miopen_rnn_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "miopen_rnn_backward.out(Tensor input, Tensor[] weight, int weight_stride0, Tensor weight_buf, Tensor hx, Tensor? cx, Tensor output, Tensor? grad_output, Tensor? grad_hy, Tensor? grad_cy, int mode, int hidden_size, int num_layers, bool batch_first, float dropout, bool train, bool bidirectional, int[] batch_sizes, Tensor? dropout_state, Tensor reserve, bool[4] output_mask, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2, Tensor(d!)[] out3) -> ()") + static void call(const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const c10::optional & cx, const at::Tensor & output, const c10::optional & grad_output, const c10::optional & grad_hy, const c10::optional & grad_cy, int64_t mode, int64_t hidden_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, at::IntArrayRef batch_sizes, const c10::optional & dropout_state, const at::Tensor & reserve, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::TensorList out3); + static void redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const c10::optional & cx, const at::Tensor & output, const c10::optional & grad_output, const c10::optional & grad_hy, const c10::optional & grad_cy, int64_t mode, int64_t hidden_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, at::IntArrayRef batch_sizes, const c10::optional & dropout_state, const at::Tensor & reserve, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::TensorList out3); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/multilabel_margin_loss_forward_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/multilabel_margin_loss_forward_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..5f3d70eb8a6c764c92c9900885901da5efca41eb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/multilabel_margin_loss_forward_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API ::std::tuple multilabel_margin_loss_forward(const at::Tensor & self, const at::Tensor & target, int64_t reduction); +TORCH_API ::std::tuple multilabel_margin_loss_forward_out(at::Tensor & output, at::Tensor & is_target, const at::Tensor & self, const at::Tensor & target, int64_t reduction); +TORCH_API ::std::tuple multilabel_margin_loss_forward_outf(const at::Tensor & self, const at::Tensor & target, int64_t reduction, at::Tensor & output, at::Tensor & is_target); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/multilabel_margin_loss_forward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/multilabel_margin_loss_forward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..792a59b34ff28a468c404afee2d7e5182b9e8921 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/multilabel_margin_loss_forward_native.h @@ -0,0 +1,24 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API ::std::tuple multilabel_margin_loss_forward_cpu(const at::Tensor & self, const at::Tensor & target, int64_t reduction); +TORCH_API ::std::tuple multilabel_margin_loss_forward_out_cpu(const at::Tensor & self, const at::Tensor & target, int64_t reduction, at::Tensor & output, at::Tensor & is_target); +TORCH_API ::std::tuple multilabel_margin_loss_forward_cuda(const at::Tensor & self, const at::Tensor & target, int64_t reduction); +TORCH_API ::std::tuple multilabel_margin_loss_forward_out_cuda(const at::Tensor & self, const at::Tensor & target, int64_t reduction, at::Tensor & output, at::Tensor & is_target); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_norm_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_norm_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..a3fa2f83516ec4c058797dfc493b04084541757e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_norm_ops.h @@ -0,0 +1,61 @@ +#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 native_norm { + using schema = at::Tensor (const at::Tensor &, 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::native_norm") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "native_norm(Tensor self, Scalar p=2) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Scalar & p); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & p); +}; + +struct TORCH_API native_norm_ScalarOpt_dim_dtype { + using schema = at::Tensor (const at::Tensor &, const c10::optional &, at::IntArrayRef, 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::native_norm") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "ScalarOpt_dim_dtype") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "native_norm.ScalarOpt_dim_dtype(Tensor self, Scalar? p, int[1] dim, bool keepdim, ScalarType? dtype) -> Tensor") + static at::Tensor call(const at::Tensor & self, const c10::optional & p, at::IntArrayRef dim, bool keepdim, c10::optional dtype); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const c10::optional & p, at::IntArrayRef dim, bool keepdim, c10::optional dtype); +}; + +struct TORCH_API native_norm_out { + using schema = at::Tensor & (const at::Tensor &, const at::Scalar &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::native_norm") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "native_norm.out(Tensor self, Scalar p=2, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Scalar & p, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & p, at::Tensor & out); +}; + +struct TORCH_API native_norm_ScalarOpt_dim_dtype_out { + using schema = at::Tensor & (const at::Tensor &, const c10::optional &, at::IntArrayRef, bool, 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::native_norm") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "ScalarOpt_dim_dtype_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "native_norm.ScalarOpt_dim_dtype_out(Tensor self, Scalar? p, int[1] dim, bool keepdim, ScalarType? dtype, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const c10::optional & p, at::IntArrayRef dim, bool keepdim, c10::optional dtype, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const c10::optional & p, at::IntArrayRef dim, bool keepdim, c10::optional dtype, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nll_loss2d_forward_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nll_loss2d_forward_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..56329698b51c57b5f04f48a27bf861d5fa313b61 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nll_loss2d_forward_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 ::std::tuple nll_loss2d_forward(const at::Tensor & self, const at::Tensor & target, const c10::optional & weight, int64_t reduction, int64_t ignore_index); +TORCH_API ::std::tuple nll_loss2d_forward_symint(const at::Tensor & self, const at::Tensor & target, const c10::optional & weight, int64_t reduction, c10::SymInt ignore_index); +TORCH_API ::std::tuple nll_loss2d_forward_out(at::Tensor & output, at::Tensor & total_weight, const at::Tensor & self, const at::Tensor & target, const c10::optional & weight, int64_t reduction, int64_t ignore_index); +TORCH_API ::std::tuple nll_loss2d_forward_outf(const at::Tensor & self, const at::Tensor & target, const c10::optional & weight, int64_t reduction, int64_t ignore_index, at::Tensor & output, at::Tensor & total_weight); +TORCH_API ::std::tuple nll_loss2d_forward_symint_out(at::Tensor & output, at::Tensor & total_weight, const at::Tensor & self, const at::Tensor & target, const c10::optional & weight, int64_t reduction, c10::SymInt ignore_index); +TORCH_API ::std::tuple nll_loss2d_forward_symint_outf(const at::Tensor & self, const at::Tensor & target, const c10::optional & weight, int64_t reduction, c10::SymInt ignore_index, at::Tensor & output, at::Tensor & total_weight); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pad_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pad_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..9deb2d792d4fdcec11390527b29227befde71100 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pad_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 pad(const at::Tensor & self, at::IntArrayRef pad, c10::string_view mode="constant", c10::optional value=c10::nullopt); +TORCH_API at::Tensor pad_symint(const at::Tensor & self, c10::SymIntArrayRef pad, c10::string_view mode="constant", c10::optional value=c10::nullopt); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pixel_shuffle_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pixel_shuffle_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..71456b97c5c3c871e009f80cf4250402efd76dfa --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pixel_shuffle_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 pixel_shuffle(const at::Tensor & self, int64_t upscale_factor); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantile_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantile_native.h new file mode 100644 index 0000000000000000000000000000000000000000..34a8035754af910fb86b5230f83965066a1b027c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantile_native.h @@ -0,0 +1,24 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor quantile(const at::Tensor & self, const at::Tensor & q, c10::optional dim=c10::nullopt, bool keepdim=false, c10::string_view interpolation="linear"); +TORCH_API at::Tensor & quantile_out(const at::Tensor & self, const at::Tensor & q, c10::optional dim, bool keepdim, c10::string_view interpolation, at::Tensor & out); +TORCH_API at::Tensor quantile(const at::Tensor & self, double q, c10::optional dim=c10::nullopt, bool keepdim=false, c10::string_view interpolation="linear"); +TORCH_API at::Tensor & quantile_out(const at::Tensor & self, double q, c10::optional dim, bool keepdim, c10::string_view interpolation, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rshift_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rshift_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..4b8b42e2c7676e6b3579e6143cef4f0feb3871d2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rshift_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 __rshift__(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & __irshift__(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor __rshift__(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & __irshift__(at::Tensor & self, const at::Tensor & other); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/scaled_dot_product_attention_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/scaled_dot_product_attention_native.h new file mode 100644 index 0000000000000000000000000000000000000000..0bf4f5174df6757e00c8091a904d931b5d7b56d7 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/scaled_dot_product_attention_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 scaled_dot_product_attention(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const c10::optional & attn_mask={}, double dropout_p=0.0, bool is_causal=false, c10::optional scale=c10::nullopt); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/slogdet_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/slogdet_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..4569985bdfd9f15f89cf61640e5bc414de0187a8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/slogdet_compositeimplicitautograd_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API ::std::tuple slogdet(const at::Tensor & self); +TORCH_API ::std::tuple slogdet_out(at::Tensor & sign, at::Tensor & logabsdet, const at::Tensor & self); +TORCH_API ::std::tuple slogdet_outf(const at::Tensor & self, at::Tensor & sign, at::Tensor & logabsdet); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sort_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sort_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..1ff1cb8a160ade0070e82bddb33e1cfe2673276d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sort_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_sort_stable : public at::impl::MetaBase { + + + void meta(const at::Tensor & self, c10::optional stable, int64_t dim, bool descending); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_i1e.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_i1e.h new file mode 100644 index 0000000000000000000000000000000000000000..af2d15cfce01caad3f03aa64d584f0baaf4b49a2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_i1e.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_i1e(Tensor self) -> Tensor +inline at::Tensor special_i1e(const at::Tensor & self) { + return at::_ops::special_i1e::call(self); +} + +// aten::special_i1e.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & special_i1e_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::special_i1e_out::call(self, out); +} +// aten::special_i1e.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & special_i1e_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::special_i1e_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sym_constrain_range_for_size_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sym_constrain_range_for_size_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..9be6584303dc243d3b8be4ccc1acb654e587fae9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sym_constrain_range_for_size_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 sym_constrain_range_for_size { + using schema = void (const at::Scalar &, 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::sym_constrain_range_for_size") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "sym_constrain_range_for_size(Scalar size, *, int? min=None, int? max=None) -> ()") + static void call(const at::Scalar & size, c10::optional min, c10::optional max); + static void redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & size, c10::optional min, c10::optional max); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_sparse_bsr_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_sparse_bsr_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..13ded2354cb79a0566a694216bd41cdc000f9259 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_sparse_bsr_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 to_sparse_bsr { + using schema = at::Tensor (const at::Tensor &, at::IntArrayRef, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::to_sparse_bsr") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "to_sparse_bsr(Tensor self, int[2] blocksize, int? dense_dim=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, at::IntArrayRef blocksize, c10::optional dense_dim); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::IntArrayRef blocksize, c10::optional dense_dim); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/true_divide.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/true_divide.h new file mode 100644 index 0000000000000000000000000000000000000000..70e3eab39562e983c073bbe74c6359fe8617f367 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/true_divide.h @@ -0,0 +1,44 @@ +#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::true_divide.Tensor(Tensor self, Tensor other) -> Tensor +inline at::Tensor true_divide(const at::Tensor & self, const at::Tensor & other) { + return at::_ops::true_divide_Tensor::call(self, other); +} + +// aten::true_divide.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & true_divide_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) { + return at::_ops::true_divide_out::call(self, other, out); +} +// aten::true_divide.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & true_divide_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) { + return at::_ops::true_divide_out::call(self, other, out); +} + +// aten::true_divide.Scalar(Tensor self, Scalar other) -> Tensor +inline at::Tensor true_divide(const at::Tensor & self, const at::Scalar & other) { + return at::_ops::true_divide_Scalar::call(self, other); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bicubic2d_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bicubic2d_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..48c7575bb9b8010ceb77d7334a9233ff38b20d29 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bicubic2d_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_upsample_bicubic2d : public at::impl::MetaBase { + + + void meta(const at::Tensor & self, at::ArrayRef output_size, bool align_corners, c10::optional scales_h, c10::optional scales_w); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest3d_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest3d_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c9555de88cafa615cd5099ae684a1932fac2961d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest3d_cuda_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor upsample_nearest3d(const at::Tensor & self, at::IntArrayRef output_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor upsample_nearest3d_symint(const at::Tensor & self, c10::SymIntArrayRef output_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_nearest3d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_nearest3d_outf(const at::Tensor & self, at::IntArrayRef output_size, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w, at::Tensor & out); +TORCH_API at::Tensor & upsample_nearest3d_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_nearest3d_symint_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w, at::Tensor & out); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/view_as_complex_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/view_as_complex_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..f1e571fcf2c497f02a4942aea90d55dfcf34bf74 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/view_as_complex_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 view_as_complex(const at::Tensor & self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/view_as_complex_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/view_as_complex_native.h new file mode 100644 index 0000000000000000000000000000000000000000..389748fc411118879791a56d906eb5f18e6492d5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/view_as_complex_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 view_as_complex(const at::Tensor & self); +} // namespace native +} // namespace at