diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_assert_scalar.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_assert_scalar.h new file mode 100644 index 0000000000000000000000000000000000000000..f4f60b0f792836537e5241d6cbf7abeb32c99fad --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_assert_scalar.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_assert_scalar(Scalar self, str assert_msg) -> () +inline void _assert_scalar(const at::Scalar & self, c10::string_view assert_msg) { + return at::_ops::_assert_scalar::call(self, assert_msg); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foobar_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foobar_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..cc90a22df20302e07b540d5a0561b4a5aa3c8c6a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foobar_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 _foobar(const at::Tensor & self, bool arg1=true, bool arg2=true, bool arg3=true); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_erfc.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_erfc.h new file mode 100644 index 0000000000000000000000000000000000000000..9e95a2cc999c32bcb75e14d9b0da21285993944e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_erfc.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_erfc(Tensor[] self) -> Tensor[] +inline ::std::vector _foreach_erfc(at::TensorList self) { + return at::_ops::_foreach_erfc::call(self); +} + +// aten::_foreach_erfc_(Tensor(a!)[] self) -> () +inline void _foreach_erfc_(at::TensorList self) { + return at::_ops::_foreach_erfc_::call(self); +} + +// aten::_foreach_erfc.out(Tensor[] self, *, Tensor(a!)[] out) -> () +inline void _foreach_erfc_out(at::TensorList out, at::TensorList self) { + return at::_ops::_foreach_erfc_out::call(self, out); +} +// aten::_foreach_erfc.out(Tensor[] self, *, Tensor(a!)[] out) -> () +inline void _foreach_erfc_outf(at::TensorList self, at::TensorList out) { + return at::_ops::_foreach_erfc_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_lerp.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_lerp.h new file mode 100644 index 0000000000000000000000000000000000000000..fbf89b778ae5e303d04a59112e7790a63e9d8c99 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_lerp.h @@ -0,0 +1,63 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_foreach_lerp.List(Tensor[] self, Tensor[] tensors1, Tensor[] weights) -> Tensor[] +inline ::std::vector _foreach_lerp(at::TensorList self, at::TensorList tensors1, at::TensorList weights) { + return at::_ops::_foreach_lerp_List::call(self, tensors1, weights); +} + +// aten::_foreach_lerp_.List(Tensor(a!)[] self, Tensor[] tensors1, Tensor[] weights) -> () +inline void _foreach_lerp_(at::TensorList self, at::TensorList tensors1, at::TensorList weights) { + return at::_ops::_foreach_lerp__List::call(self, tensors1, weights); +} + +// aten::_foreach_lerp.Scalar(Tensor[] self, Tensor[] tensors1, Scalar weight) -> Tensor[] +inline ::std::vector _foreach_lerp(at::TensorList self, at::TensorList tensors1, const at::Scalar & weight) { + return at::_ops::_foreach_lerp_Scalar::call(self, tensors1, weight); +} + +// aten::_foreach_lerp_.Scalar(Tensor(a!)[] self, Tensor[] tensors1, Scalar weight) -> () +inline void _foreach_lerp_(at::TensorList self, at::TensorList tensors1, const at::Scalar & weight) { + return at::_ops::_foreach_lerp__Scalar::call(self, tensors1, weight); +} + +// aten::_foreach_lerp.List_out(Tensor[] self, Tensor[] tensors1, Tensor[] weights, *, Tensor(a!)[] out) -> () +inline void _foreach_lerp_out(at::TensorList out, at::TensorList self, at::TensorList tensors1, at::TensorList weights) { + return at::_ops::_foreach_lerp_List_out::call(self, tensors1, weights, out); +} +// aten::_foreach_lerp.List_out(Tensor[] self, Tensor[] tensors1, Tensor[] weights, *, Tensor(a!)[] out) -> () +inline void _foreach_lerp_outf(at::TensorList self, at::TensorList tensors1, at::TensorList weights, at::TensorList out) { + return at::_ops::_foreach_lerp_List_out::call(self, tensors1, weights, out); +} + +// aten::_foreach_lerp.Scalar_out(Tensor[] self, Tensor[] tensors1, Scalar weight, *, Tensor(a!)[] out) -> () +inline void _foreach_lerp_out(at::TensorList out, at::TensorList self, at::TensorList tensors1, const at::Scalar & weight) { + return at::_ops::_foreach_lerp_Scalar_out::call(self, tensors1, weight, out); +} +// aten::_foreach_lerp.Scalar_out(Tensor[] self, Tensor[] tensors1, Scalar weight, *, Tensor(a!)[] out) -> () +inline void _foreach_lerp_outf(at::TensorList self, at::TensorList tensors1, const at::Scalar & weight, at::TensorList out) { + return at::_ops::_foreach_lerp_Scalar_out::call(self, tensors1, weight, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sinh_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sinh_native.h new file mode 100644 index 0000000000000000000000000000000000000000..2ef10396f6c8b3efc8e12440bd4e856bf4772615 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sinh_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 void _foreach_sinh_out(at::TensorList self, at::TensorList out); +TORCH_API ::std::vector foreach_tensor_sinh_slow(at::TensorList self); +TORCH_API void foreach_tensor_sinh_slow_(at::TensorList self); +TORCH_API ::std::vector foreach_tensor_sinh_cuda(at::TensorList self); +TORCH_API void foreach_tensor_sinh_cuda_(at::TensorList self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_sgd.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_sgd.h new file mode 100644 index 0000000000000000000000000000000000000000..9f16f621b345113a4e265abcd5de19e8095bc5e8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_sgd.h @@ -0,0 +1,63 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_fused_sgd_(Tensor(a!)[] self, Tensor(b!)[] grads, Tensor(c!)[] momentum_buffer_list, *, float weight_decay, float momentum, float lr, float dampening, bool nesterov, bool maximize, bool is_first_step, Tensor? grad_scale=None, Tensor? found_inf=None) -> () +inline void _fused_sgd_(at::TensorList self, at::TensorList grads, at::TensorList momentum_buffer_list, double weight_decay, double momentum, double lr, double dampening, bool nesterov, bool maximize, bool is_first_step, const c10::optional & grad_scale={}, const c10::optional & found_inf={}) { + return at::_ops::_fused_sgd_::call(self, grads, momentum_buffer_list, weight_decay, momentum, lr, dampening, nesterov, maximize, is_first_step, grad_scale, found_inf); +} + +// aten::_fused_sgd_.tensor_lr(Tensor(a!)[] self, Tensor(b!)[] grads, Tensor(c!)[] momentum_buffer_list, *, float weight_decay, float momentum, Tensor lr, float dampening, bool nesterov, bool maximize, bool is_first_step, Tensor? grad_scale=None, Tensor? found_inf=None) -> () +inline void _fused_sgd_(at::TensorList self, at::TensorList grads, at::TensorList momentum_buffer_list, double weight_decay, double momentum, const at::Tensor & lr, double dampening, bool nesterov, bool maximize, bool is_first_step, const c10::optional & grad_scale={}, const c10::optional & found_inf={}) { + return at::_ops::_fused_sgd__tensor_lr::call(self, grads, momentum_buffer_list, weight_decay, momentum, lr, dampening, nesterov, maximize, is_first_step, grad_scale, found_inf); +} + +// aten::_fused_sgd.out(Tensor[] self, Tensor(b!)[] grads, Tensor(c!)[] momentum_buffer_list, *, float weight_decay, float momentum, float lr, float dampening, bool nesterov, bool maximize, bool is_first_step, Tensor? grad_scale=None, Tensor? found_inf=None, Tensor(a!)[] out) -> () +inline void _fused_sgd_out(at::TensorList out, at::TensorList self, at::TensorList grads, at::TensorList momentum_buffer_list, double weight_decay, double momentum, double lr, double dampening, bool nesterov, bool maximize, bool is_first_step, const c10::optional & grad_scale={}, const c10::optional & found_inf={}) { + return at::_ops::_fused_sgd_out::call(self, grads, momentum_buffer_list, weight_decay, momentum, lr, dampening, nesterov, maximize, is_first_step, grad_scale, found_inf, out); +} +// aten::_fused_sgd.out(Tensor[] self, Tensor(b!)[] grads, Tensor(c!)[] momentum_buffer_list, *, float weight_decay, float momentum, float lr, float dampening, bool nesterov, bool maximize, bool is_first_step, Tensor? grad_scale=None, Tensor? found_inf=None, Tensor(a!)[] out) -> () +inline void _fused_sgd_outf(at::TensorList self, at::TensorList grads, at::TensorList momentum_buffer_list, double weight_decay, double momentum, double lr, double dampening, bool nesterov, bool maximize, bool is_first_step, const c10::optional & grad_scale, const c10::optional & found_inf, at::TensorList out) { + return at::_ops::_fused_sgd_out::call(self, grads, momentum_buffer_list, weight_decay, momentum, lr, dampening, nesterov, maximize, is_first_step, grad_scale, found_inf, out); +} + +// aten::_fused_sgd(Tensor[] self, Tensor[] grads, Tensor[] momentum_buffer_list, *, float weight_decay, float momentum, float lr, float dampening, bool nesterov, bool maximize, bool is_first_step, Tensor? grad_scale=None, Tensor? found_inf=None) -> (Tensor[] self_out, Tensor[] grads_out, Tensor[] momentum_buffer_list_out) +inline ::std::tuple<::std::vector,::std::vector,::std::vector> _fused_sgd(at::TensorList self, at::TensorList grads, at::TensorList momentum_buffer_list, double weight_decay, double momentum, double lr, double dampening, bool nesterov, bool maximize, bool is_first_step, const c10::optional & grad_scale={}, const c10::optional & found_inf={}) { + return at::_ops::_fused_sgd::call(self, grads, momentum_buffer_list, weight_decay, momentum, lr, dampening, nesterov, maximize, is_first_step, grad_scale, found_inf); +} + +// aten::_fused_sgd.tensor_lr_out(Tensor[] self, Tensor(b!)[] grads, Tensor(c!)[] momentum_buffer_list, *, float weight_decay, float momentum, Tensor lr, float dampening, bool nesterov, bool maximize, bool is_first_step, Tensor? grad_scale=None, Tensor? found_inf=None, Tensor(a!)[] out) -> () +inline void _fused_sgd_out(at::TensorList out, at::TensorList self, at::TensorList grads, at::TensorList momentum_buffer_list, double weight_decay, double momentum, const at::Tensor & lr, double dampening, bool nesterov, bool maximize, bool is_first_step, const c10::optional & grad_scale={}, const c10::optional & found_inf={}) { + return at::_ops::_fused_sgd_tensor_lr_out::call(self, grads, momentum_buffer_list, weight_decay, momentum, lr, dampening, nesterov, maximize, is_first_step, grad_scale, found_inf, out); +} +// aten::_fused_sgd.tensor_lr_out(Tensor[] self, Tensor(b!)[] grads, Tensor(c!)[] momentum_buffer_list, *, float weight_decay, float momentum, Tensor lr, float dampening, bool nesterov, bool maximize, bool is_first_step, Tensor? grad_scale=None, Tensor? found_inf=None, Tensor(a!)[] out) -> () +inline void _fused_sgd_outf(at::TensorList self, at::TensorList grads, at::TensorList momentum_buffer_list, double weight_decay, double momentum, const at::Tensor & lr, double dampening, bool nesterov, bool maximize, bool is_first_step, const c10::optional & grad_scale, const c10::optional & found_inf, at::TensorList out) { + return at::_ops::_fused_sgd_tensor_lr_out::call(self, grads, momentum_buffer_list, weight_decay, momentum, lr, dampening, nesterov, maximize, is_first_step, grad_scale, found_inf, out); +} + +// aten::_fused_sgd.tensor_lr(Tensor[] self, Tensor[] grads, Tensor[] momentum_buffer_list, *, float weight_decay, float momentum, Tensor lr, float dampening, bool nesterov, bool maximize, bool is_first_step, Tensor? grad_scale=None, Tensor? found_inf=None) -> (Tensor[] self_out, Tensor[] grads_out, Tensor[] momentum_buffer_list_out) +inline ::std::tuple<::std::vector,::std::vector,::std::vector> _fused_sgd(at::TensorList self, at::TensorList grads, at::TensorList momentum_buffer_list, double weight_decay, double momentum, const at::Tensor & lr, double dampening, bool nesterov, bool maximize, bool is_first_step, const c10::optional & grad_scale={}, const c10::optional & found_inf={}) { + return at::_ops::_fused_sgd_tensor_lr::call(self, grads, momentum_buffer_list, weight_decay, momentum, lr, dampening, nesterov, maximize, is_first_step, grad_scale, found_inf); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fw_primal_copy_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fw_primal_copy_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..686b3fb354a076593d2c0acdbd6824551d9b5c9e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fw_primal_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 _fw_primal_copy { + using schema = 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::_fw_primal_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_fw_primal_copy(Tensor self, int level) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t level); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t level); +}; + +struct TORCH_API _fw_primal_copy_out { + using schema = at::Tensor & (const at::Tensor &, 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::_fw_primal_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_fw_primal_copy.out(Tensor self, int level, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, int64_t level, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t level, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..9b85a16eb42b94b22f3e37bcf95dc62317b676a4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl_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 & _index_put_impl_(at::Tensor & self, const c10::List> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_native.h new file mode 100644 index 0000000000000000000000000000000000000000..f61b20fee12646cf63753db0de9384562d6e04a4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_native.h @@ -0,0 +1,26 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_log_softmax_cpu_out : public at::meta::structured__log_softmax { +void impl(const at::Tensor & self, int64_t dim, bool half_to_float, const at::Tensor & out); +}; +struct TORCH_API structured_log_softmax_cuda_out : public at::meta::structured__log_softmax { +void impl(const at::Tensor & self, int64_t dim, bool half_to_float, const at::Tensor & out); +}; +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_make_per_channel_quantized_tensor.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_make_per_channel_quantized_tensor.h new file mode 100644 index 0000000000000000000000000000000000000000..d7b8ed2c70c3f3c074cca3b7c59d2b70758f4ce5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_make_per_channel_quantized_tensor.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::_make_per_channel_quantized_tensor(Tensor self, Tensor scale, Tensor zero_point, int axis) -> Tensor +inline at::Tensor _make_per_channel_quantized_tensor(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis) { + return at::_ops::_make_per_channel_quantized_tensor::call(self, scale, zero_point, axis); +} + +// aten::_make_per_channel_quantized_tensor.out(Tensor self, Tensor scale, Tensor zero_point, int axis, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _make_per_channel_quantized_tensor_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis) { + return at::_ops::_make_per_channel_quantized_tensor_out::call(self, scale, zero_point, axis, out); +} +// aten::_make_per_channel_quantized_tensor.out(Tensor self, Tensor scale, Tensor zero_point, int axis, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _make_per_channel_quantized_tensor_outf(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis, at::Tensor & out) { + return at::_ops::_make_per_channel_quantized_tensor_out::call(self, scale, zero_point, axis, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_neg_view_copy_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_neg_view_copy_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a9c0d7b840e402186378c89844a5871fe695d28d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_neg_view_copy_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & _neg_view_copy_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & _neg_view_copy_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_softmax_with_shape_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_softmax_with_shape_native.h new file mode 100644 index 0000000000000000000000000000000000000000..1b40d1a0b7cf70e4983fc8526cad31004ae25b21 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_softmax_with_shape_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 NestedTensor_softmax_dropout(const at::Tensor & self, const at::Tensor & query); +TORCH_API at::Tensor NestedTensor_softmax_dropout_cuda(const at::Tensor & self, const at::Tensor & query); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_strides_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_strides_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..406098009966baf37efea23e78407119aacdd936 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_strides_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & _nested_tensor_strides_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & _nested_tensor_strides_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_reshape_from_tensor.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_reshape_from_tensor.h new file mode 100644 index 0000000000000000000000000000000000000000..2d21aa9f503f093b4df876410540bc04748ca51a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_reshape_from_tensor.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_reshape_from_tensor(Tensor self, Tensor shape) -> Tensor +inline at::Tensor _reshape_from_tensor(const at::Tensor & self, const at::Tensor & shape) { + return at::_ops::_reshape_from_tensor::call(self, shape); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_resize_output_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_resize_output_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..3de95359cfdc4ab81eac2643ba64e8db36400f7b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_resize_output_meta_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 const at::Tensor & _resize_output_(const at::Tensor & self, at::IntArrayRef size, at::Device device); +TORCH_API const at::Tensor & _resize_output__symint(const at::Tensor & self, c10::SymIntArrayRef size, at::Device device); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_slow_conv2d_backward_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_slow_conv2d_backward_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..6844c06089055c7a9d828963dd67cf7c01cee983 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_slow_conv2d_backward_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API ::std::tuple _slow_conv2d_backward_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, ::std::array output_mask); +TORCH_API ::std::tuple _slow_conv2d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); +TORCH_API ::std::tuple _slow_conv2d_backward_symint_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, ::std::array output_mask); +TORCH_API ::std::tuple _slow_conv2d_backward_symint_outf(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_broadcast_to_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_broadcast_to_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..97bbe89c41f4fd672f410cd792ee0dc121411940 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_broadcast_to_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 _sparse_broadcast_to { + 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") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_sparse_broadcast_to(Tensor(a) self, int[] size) -> Tensor(a)") + 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); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_log_softmax_backward_data_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_log_softmax_backward_data_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..01aab769ee9e990baa7720b039e7373a362177fa --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_log_softmax_backward_data_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & _sparse_log_softmax_backward_data_out(at::Tensor & out, const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, const at::Tensor & self); +TORCH_API at::Tensor & _sparse_log_softmax_backward_data_outf(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, const at::Tensor & self, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_log_softmax_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_log_softmax_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..65943d90ab306062ff4e0fd16809c283a931b4f8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_log_softmax_compositeimplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor _sparse_log_softmax(const at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); +TORCH_API at::Tensor _sparse_log_softmax(const at::Tensor & self, at::Dimname dim, c10::optional dtype=c10::nullopt); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_stack.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_stack.h new file mode 100644 index 0000000000000000000000000000000000000000..170f641870fd178512f587f5da602618b1bb2fc0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_stack.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::_stack(Tensor[] tensors, int dim=0) -> Tensor +inline at::Tensor _stack(at::TensorList tensors, int64_t dim=0) { + return at::_ops::_stack::call(tensors, dim); +} + +// aten::_stack.out(Tensor[] tensors, int dim=0, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _stack_out(at::Tensor & out, at::TensorList tensors, int64_t dim=0) { + return at::_ops::_stack_out::call(tensors, dim, out); +} +// aten::_stack.out(Tensor[] tensors, int dim=0, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _stack_outf(at::TensorList tensors, int64_t dim, at::Tensor & out) { + return at::_ops::_stack_out::call(tensors, dim, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_optional_intlist_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_optional_intlist_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..b32cd22a77a6511f6994fa19fc647414e9cf32d1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_optional_intlist_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 _test_optional_intlist(const at::Tensor & values, at::OptionalIntArrayRef addends); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_parallel_materialize_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_parallel_materialize_native.h new file mode 100644 index 0000000000000000000000000000000000000000..72641db0aa6993de3a062c2dfd84a7bd7b82af22 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_parallel_materialize_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 _test_parallel_materialize(const at::Tensor & self, int64_t num_parallel, bool skip_first=false); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_fused_lstm_cell_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_fused_lstm_cell_native.h new file mode 100644 index 0000000000000000000000000000000000000000..9d435ecbcb6eb99d549dbe708b4767a06508b732 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_fused_lstm_cell_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 _thnn_fused_lstm_cell_out(const at::Tensor & input_gates, const at::Tensor & hidden_gates, const at::Tensor & cx, const c10::optional & input_bias, const c10::optional & hidden_bias, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); +TORCH_API ::std::tuple _thnn_fused_lstm_cell_cuda(const at::Tensor & input_gates, const at::Tensor & hidden_gates, const at::Tensor & cx, const c10::optional & input_bias={}, const c10::optional & hidden_bias={}); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_sparse_bsc.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_sparse_bsc.h new file mode 100644 index 0000000000000000000000000000000000000000..9b6ccffa14234c37ffff63fa082e53648f65179f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_sparse_bsc.h @@ -0,0 +1,34 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_to_sparse_bsc.out(Tensor self, int[2] blocksize, int? dense_dim=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _to_sparse_bsc_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef blocksize, c10::optional dense_dim=c10::nullopt) { + return at::_ops::_to_sparse_bsc_out::call(self, blocksize, dense_dim, out); +} +// aten::_to_sparse_bsc.out(Tensor self, int[2] blocksize, int? dense_dim=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _to_sparse_bsc_outf(const at::Tensor & self, at::IntArrayRef blocksize, c10::optional dense_dim, at::Tensor & out) { + return at::_ops::_to_sparse_bsc_out::call(self, blocksize, dense_dim, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact2d_backward.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact2d_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..412bd147a4845a6cddfb794bc8254b8e6c5fa694 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact2d_backward.h @@ -0,0 +1,91 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_upsample_nearest_exact2d_backward.grad_input(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & _upsample_nearest_exact2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt) { + return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & _upsample_nearest_exact2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt) { + return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w, grad_input); + } +} + +// aten::_upsample_nearest_exact2d_backward.grad_input(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & _upsample_nearest_exact2d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input) { + return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & _upsample_nearest_exact2d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input) { + return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w, grad_input); + } +} + +// aten::_upsample_nearest_exact2d_backward.grad_input(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & _upsample_nearest_exact2d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt) { + return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, output_size, input_size, scales_h, scales_w, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & _upsample_nearest_exact2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt) { + return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, output_size, input_size, scales_h, scales_w, grad_input); + } +} + +// aten::_upsample_nearest_exact2d_backward.grad_input(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & _upsample_nearest_exact2d_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input) { + return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, output_size, input_size, scales_h, scales_w, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & _upsample_nearest_exact2d_backward_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input) { + return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, output_size, input_size, scales_h, scales_w, grad_input); + } +} + +// aten::_upsample_nearest_exact2d_backward(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None) -> Tensor +inline at::Tensor _upsample_nearest_exact2d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt) { + return at::_ops::_upsample_nearest_exact2d_backward::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w); +} +namespace symint { + template ::value>> + at::Tensor _upsample_nearest_exact2d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt) { + return at::_ops::_upsample_nearest_exact2d_backward::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w); + } +} + +// aten::_upsample_nearest_exact2d_backward(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None) -> Tensor +inline at::Tensor _upsample_nearest_exact2d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt) { + return at::_ops::_upsample_nearest_exact2d_backward::call(grad_output, output_size, input_size, scales_h, scales_w); +} +namespace symint { + template ::value>> + at::Tensor _upsample_nearest_exact2d_backward(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt) { + return at::_ops::_upsample_nearest_exact2d_backward::call(grad_output, output_size, input_size, scales_h, scales_w); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_int4pack_mm_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_int4pack_mm_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a9e414c5d502dd94436968afd7f8ed962e861fea --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_int4pack_mm_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 _weight_int4pack_mm(const at::Tensor & self, const at::Tensor & mat2, int64_t qGroupSize, const at::Tensor & qScaleAndZeros); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/absolute_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/absolute_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a2ae237bba787b66a9f92eff4ec1995308c84a2d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/absolute_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor absolute(const at::Tensor & self); +TORCH_API at::Tensor & absolute_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & absolute_(at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/alias_copy_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/alias_copy_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..4b2b8e738af1639b54020aa79537b872931fab5e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/alias_copy_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 alias_copy(const at::Tensor & self); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/arcsin.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/arcsin.h new file mode 100644 index 0000000000000000000000000000000000000000..f184090987c5cbe1953b7fc9cb93cbb6f920b188 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/arcsin.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::arcsin(Tensor self) -> Tensor +inline at::Tensor arcsin(const at::Tensor & self) { + return at::_ops::arcsin::call(self); +} + +// aten::arcsin_(Tensor(a!) self) -> Tensor(a!) +inline at::Tensor & arcsin_(at::Tensor & self) { + return at::_ops::arcsin_::call(self); +} + +// aten::arcsin.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & arcsin_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::arcsin_out::call(self, out); +} +// aten::arcsin.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & arcsin_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::arcsin_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/argmax_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/argmax_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8600211a8c3bfdc8df17740bfd4ea54d6b490e86 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/argmax_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 argmax(const at::Tensor & self, c10::optional dim=c10::nullopt, bool keepdim=false); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_backward_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_backward_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..18c982408091b18fc34d41973d1c0d5111a8fccc --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool3d_backward_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 avg_pool3d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override); +TORCH_API at::Tensor & avg_pool3d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override); +TORCH_API at::Tensor & avg_pool3d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, c10::optional divisor_override, at::Tensor & grad_input); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ccol_indices_copy_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ccol_indices_copy_native.h new file mode 100644 index 0000000000000000000000000000000000000000..8a61f5a6d263192e47f706fb07f5160dec9c60f4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ccol_indices_copy_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & ccol_indices_copy_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor ccol_indices_copy(const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diag_embed_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diag_embed_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..860f46781fb8638d9e7c5c708d65a6f75a94842f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diag_embed_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 diag_embed(const at::Tensor & self, int64_t offset=0, int64_t dim1=-2, int64_t dim2=-1); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..4ff1a04f05746db286bfff9ce9c9b4934d9aba1d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_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 digamma(const at::Tensor & self); +TORCH_API at::Tensor & digamma_(at::Tensor & self); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/eq_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/eq_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a15f08555822b803f3b3400e9b0da39ea3247f43 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/eq_cpu_dispatch.h @@ -0,0 +1,30 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor eq(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & eq_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & eq_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +TORCH_API at::Tensor & eq_(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor eq(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & eq_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & eq_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & eq_(at::Tensor & self, const at::Tensor & other); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/exp.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/exp.h new file mode 100644 index 0000000000000000000000000000000000000000..7c71dd13fcc89dd54e1d47d7991d518b5a48679e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/exp.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::exp(Tensor self) -> Tensor +inline at::Tensor exp(const at::Tensor & self) { + return at::_ops::exp::call(self); +} + +// aten::exp_(Tensor(a!) self) -> Tensor(a!) +inline at::Tensor & exp_(at::Tensor & self) { + return at::_ops::exp_::call(self); +} + +// aten::exp.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & exp_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::exp_out::call(self, out); +} +// aten::exp.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & exp_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::exp_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_rfft_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_rfft_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..1e021d8c80df0844c83614c67e5cd5e31ac88f5d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_rfft_compositeimplicitautograd_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 fft_rfft(const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt); +TORCH_API at::Tensor fft_rfft_symint(const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt); +TORCH_API at::Tensor & fft_rfft_out(at::Tensor & out, const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt); +TORCH_API at::Tensor & fft_rfft_outf(const at::Tensor & self, c10::optional n, int64_t dim, c10::optional norm, at::Tensor & out); +TORCH_API at::Tensor & fft_rfft_symint_out(at::Tensor & out, const at::Tensor & self, c10::optional n=c10::nullopt, int64_t dim=-1, c10::optional norm=c10::nullopt); +TORCH_API at::Tensor & fft_rfft_symint_outf(const at::Tensor & self, c10::optional n, int64_t dim, c10::optional norm, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_rfftfreq_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_rfftfreq_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..f0b099d134d1e485b36b9cb1a6977b5a85eebf11 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_rfftfreq_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 fft_rfftfreq { + using schema = at::Tensor (int64_t, double, c10::optional, c10::optional, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::fft_rfftfreq") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fft_rfftfreq(int n, float d=1.0, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor") + static at::Tensor call(int64_t n, double d, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, int64_t n, double d, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +}; + +struct TORCH_API fft_rfftfreq_out { + using schema = at::Tensor & (int64_t, double, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::fft_rfftfreq") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fft_rfftfreq.out(int n, float d=1.0, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(int64_t n, double d, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, int64_t n, double d, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool2d_backward_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool2d_backward_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..24d179e058031f4970cb95a7ed71fdebaf01e6a4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool2d_backward_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#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 fractional_max_pool2d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices); +TORCH_API at::Tensor & fractional_max_pool2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices); +TORCH_API at::Tensor & fractional_max_pool2d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices, at::Tensor & grad_input); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gcd_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gcd_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..2a7689c34f2e206bbe5a4565d0fb10ffcf1c281b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gcd_cpu_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor gcd(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & gcd_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & gcd_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & gcd_(at::Tensor & self, const at::Tensor & other); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_2d_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_2d_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..b744da0c27d61d5c1597849c0691396d8a86e56f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_2d_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 grid_sampler_2d { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, int64_t, int64_t, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::grid_sampler_2d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "grid_sampler_2d(Tensor input, Tensor grid, int interpolation_mode, int padding_mode, bool align_corners) -> Tensor") + static at::Tensor call(const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners); +}; + +struct TORCH_API grid_sampler_2d_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, int64_t, int64_t, 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::grid_sampler_2d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "grid_sampler_2d.out(Tensor input, Tensor grid, int interpolation_mode, int padding_mode, bool align_corners, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gru_cell_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gru_cell_native.h new file mode 100644 index 0000000000000000000000000000000000000000..2000de3bb92a9aa2b6041de337c094e21ec6963d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gru_cell_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor gru_cell(const at::Tensor & input, const at::Tensor & hx, const at::Tensor & w_ih, const at::Tensor & w_hh, const c10::optional & b_ih={}, const c10::optional & b_hh={}); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_add_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_add_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..748abba02340adc8369eb4136e2affc9ea6cb50a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_add_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 index_add_out { + using schema = at::Tensor & (const at::Tensor &, int64_t, const 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::index_add") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_add.out(Tensor self, int dim, Tensor index, Tensor source, *, Scalar alpha=1, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha, at::Tensor & out); +}; + +struct TORCH_API index_add_ { + using schema = at::Tensor & (at::Tensor &, int64_t, const 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::index_add_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_add_(Tensor(a!) self, int dim, Tensor index, Tensor source, *, Scalar alpha=1) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha); +}; + +struct TORCH_API index_add { + using schema = at::Tensor (const at::Tensor &, int64_t, const 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::index_add") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_add(Tensor self, int dim, Tensor index, Tensor source, *, Scalar alpha=1) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha); +}; + +struct TORCH_API index_add_dimname { + using schema = at::Tensor (const at::Tensor &, at::Dimname, const 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::index_add") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dimname") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_add.dimname(Tensor self, Dimname dim, Tensor index, Tensor source, *, Scalar alpha=1) -> Tensor") + static at::Tensor call(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & source, const at::Scalar & alpha); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_reduce_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_reduce_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..3a00a88d2ae924fec00dff4a255dd40aa4fe1912 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_reduce_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 index_reduce(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, c10::string_view reduce, bool include_self=true); +TORCH_API at::Tensor & index_reduce_out(at::Tensor & out, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, c10::string_view reduce, bool include_self=true); +TORCH_API at::Tensor & index_reduce_outf(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, c10::string_view reduce, bool include_self, at::Tensor & out); +TORCH_API at::Tensor & index_reduce_(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, c10::string_view reduce, bool include_self=true); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_conj_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_conj_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..360e209a38b494a62fb6dedff9bfab1ccadeda6d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_conj_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API bool is_conj(const at::Tensor & self); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_distributed_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_distributed_native.h new file mode 100644 index 0000000000000000000000000000000000000000..d1bea57a3b35a2e987a85bcfb07abb6c344d8c34 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_distributed_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 bool is_distributed(const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_inference_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_inference_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..aef9b50ca1ad3cd122240094412b6d1fe5c84ea6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_inference_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API bool is_inference(const at::Tensor & self); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isclose_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isclose_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..f4ffcff61bb1758edbe9402f242f8070e51967eb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isclose_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 isclose { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, double, double, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::isclose") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "isclose(Tensor self, Tensor other, float rtol=1e-05, float atol=1e-08, bool equal_nan=False) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & other, double rtol, double atol, bool equal_nan); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, double rtol, double atol, bool equal_nan); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_native.h new file mode 100644 index 0000000000000000000000000000000000000000..697fb8dae63cc4b0a1134c4c0ef65deab85a6332 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_linalg_lu_solve_out : public at::meta::structured_linalg_lu_solve { +void impl(const at::Tensor & LU, const at::Tensor & pivots, const at::Tensor & B, bool left, bool adjoint, const at::Tensor & out); +}; +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_norm.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_norm.h new file mode 100644 index 0000000000000000000000000000000000000000..4f380885d32e7c0b2c5928e89e7032a3f6c8fda6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_norm.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::linalg_norm(Tensor self, Scalar? ord=None, int[1]? dim=None, bool keepdim=False, *, ScalarType? dtype=None) -> Tensor +inline at::Tensor linalg_norm(const at::Tensor & self, const c10::optional & ord=c10::nullopt, at::OptionalIntArrayRef dim=c10::nullopt, bool keepdim=false, c10::optional dtype=c10::nullopt) { + return at::_ops::linalg_norm::call(self, ord, dim, keepdim, dtype); +} + +// aten::linalg_norm.ord_str(Tensor self, str ord, int[1]? dim=None, bool keepdim=False, *, ScalarType? dtype=None) -> Tensor +inline at::Tensor linalg_norm(const at::Tensor & self, c10::string_view ord, at::OptionalIntArrayRef dim=c10::nullopt, bool keepdim=false, c10::optional dtype=c10::nullopt) { + return at::_ops::linalg_norm_ord_str::call(self, ord, dim, keepdim, dtype); +} + +// aten::linalg_norm.out(Tensor self, Scalar? ord=None, int[1]? dim=None, bool keepdim=False, *, ScalarType? dtype=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & linalg_norm_out(at::Tensor & out, const at::Tensor & self, const c10::optional & ord=c10::nullopt, at::OptionalIntArrayRef dim=c10::nullopt, bool keepdim=false, c10::optional dtype=c10::nullopt) { + return at::_ops::linalg_norm_out::call(self, ord, dim, keepdim, dtype, out); +} +// aten::linalg_norm.out(Tensor self, Scalar? ord=None, int[1]? dim=None, bool keepdim=False, *, ScalarType? dtype=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & linalg_norm_outf(const at::Tensor & self, const c10::optional & ord, at::OptionalIntArrayRef dim, bool keepdim, c10::optional dtype, at::Tensor & out) { + return at::_ops::linalg_norm_out::call(self, ord, dim, keepdim, dtype, out); +} + +// aten::linalg_norm.ord_str_out(Tensor self, str ord, int[1]? dim=None, bool keepdim=False, *, ScalarType? dtype=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & linalg_norm_out(at::Tensor & out, const at::Tensor & self, c10::string_view ord, at::OptionalIntArrayRef dim=c10::nullopt, bool keepdim=false, c10::optional dtype=c10::nullopt) { + return at::_ops::linalg_norm_ord_str_out::call(self, ord, dim, keepdim, dtype, out); +} +// aten::linalg_norm.ord_str_out(Tensor self, str ord, int[1]? dim=None, bool keepdim=False, *, ScalarType? dtype=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & linalg_norm_outf(const at::Tensor & self, c10::string_view ord, at::OptionalIntArrayRef dim, bool keepdim, c10::optional dtype, at::Tensor & out) { + return at::_ops::linalg_norm_ord_str_out::call(self, ord, dim, keepdim, dtype, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logit_backward_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logit_backward_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..9cdebee676db088e052c2c9d2f4c66ef96189cb6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logit_backward_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 logit_backward(const at::Tensor & grad_output, const at::Tensor & self, c10::optional eps=c10::nullopt); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logsumexp_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logsumexp_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..322dffc2fbb5f961743483b844984c9018ba91ef --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logsumexp_compositeimplicitautograd_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor logsumexp(const at::Tensor & self, at::DimnameList dim, bool keepdim=false); +TORCH_API at::Tensor & logsumexp_out(at::Tensor & out, const at::Tensor & self, at::DimnameList dim, bool keepdim=false); +TORCH_API at::Tensor & logsumexp_outf(const at::Tensor & self, at::DimnameList dim, bool keepdim, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matrix_exp_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matrix_exp_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..e0588d0e3d873f0894e9fbbb4cab0672b15bc9b4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matrix_exp_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 matrix_exp { + using schema = at::Tensor (const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::matrix_exp") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "matrix_exp(Tensor self) -> Tensor") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_adaptive_avg_pool2d_backward.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_adaptive_avg_pool2d_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..d726d7cd6d4cf1b6660864566597f4c554c1d0a9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_adaptive_avg_pool2d_backward.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::mkldnn_adaptive_avg_pool2d_backward(Tensor grad_output, Tensor self) -> Tensor +inline at::Tensor mkldnn_adaptive_avg_pool2d_backward(const at::Tensor & grad_output, const at::Tensor & self) { + return at::_ops::mkldnn_adaptive_avg_pool2d_backward::call(grad_output, self); +} + +// aten::mkldnn_adaptive_avg_pool2d_backward.out(Tensor grad_output, Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & mkldnn_adaptive_avg_pool2d_backward_out(at::Tensor & out, const at::Tensor & grad_output, const at::Tensor & self) { + return at::_ops::mkldnn_adaptive_avg_pool2d_backward_out::call(grad_output, self, out); +} +// aten::mkldnn_adaptive_avg_pool2d_backward.out(Tensor grad_output, Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & mkldnn_adaptive_avg_pool2d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & out) { + return at::_ops::mkldnn_adaptive_avg_pool2d_backward_out::call(grad_output, self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_linear_backward_input_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_linear_backward_input_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..f66d6e9d012c31752dbae177c68a7d9fecb4d34b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_linear_backward_input_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & mkldnn_linear_backward_input_out(at::Tensor & out, at::IntArrayRef input_size, const at::Tensor & grad_output, const at::Tensor & weight); +TORCH_API at::Tensor & mkldnn_linear_backward_input_outf(at::IntArrayRef input_size, const at::Tensor & grad_output, const at::Tensor & weight, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_rnn_layer_backward_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_rnn_layer_backward_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..21d5d7e9e9491e79390f7ea48b30a1f894352094 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_rnn_layer_backward_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API ::std::tuple mkldnn_rnn_layer_backward_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3, at::Tensor & out4, at::Tensor & out5, at::Tensor & out6, const at::Tensor & input, const at::Tensor & weight1, const at::Tensor & weight2, const at::Tensor & weight3, const at::Tensor & weight4, const at::Tensor & hx_, const at::Tensor & cx_tmp, const at::Tensor & output, const at::Tensor & hy_, const at::Tensor & cy_, const c10::optional & grad_output, const c10::optional & grad_hy, const c10::optional & grad_cy, bool reverse, int64_t mode, int64_t hidden_size, int64_t num_layers, bool has_biases, bool train, bool bidirectional, at::IntArrayRef batch_sizes, bool batch_first, const at::Tensor & workspace); +TORCH_API ::std::tuple mkldnn_rnn_layer_backward_outf(const at::Tensor & input, const at::Tensor & weight1, const at::Tensor & weight2, const at::Tensor & weight3, const at::Tensor & weight4, const at::Tensor & hx_, const at::Tensor & cx_tmp, const at::Tensor & output, const at::Tensor & hy_, const at::Tensor & cy_, const c10::optional & grad_output, const c10::optional & grad_hy, const c10::optional & grad_cy, bool reverse, int64_t mode, int64_t hidden_size, int64_t num_layers, bool has_biases, bool train, bool bidirectional, at::IntArrayRef batch_sizes, bool batch_first, const at::Tensor & workspace, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3, at::Tensor & out4, at::Tensor & out5, at::Tensor & out6); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_batch_norm_backward.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_batch_norm_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..e91f96306178ca2f8b678331b4a60f9fa9b6387c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_batch_norm_backward.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::native_batch_norm_backward(Tensor grad_out, Tensor input, Tensor? weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_invstd, bool train, float eps, bool[3] output_mask) -> (Tensor, Tensor, Tensor) +inline ::std::tuple native_batch_norm_backward(const at::Tensor & grad_out, const at::Tensor & input, const c10::optional & weight, const c10::optional & running_mean, const c10::optional & running_var, const c10::optional & save_mean, const c10::optional & save_invstd, bool train, double eps, ::std::array output_mask) { + return at::_ops::native_batch_norm_backward::call(grad_out, input, weight, running_mean, running_var, save_mean, save_invstd, train, eps, output_mask); +} + +// aten::native_batch_norm_backward.out(Tensor grad_out, Tensor input, Tensor? weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_invstd, bool train, float eps, bool[3] output_mask, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!)) +inline ::std::tuple native_batch_norm_backward_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, const at::Tensor & grad_out, const at::Tensor & input, const c10::optional & weight, const c10::optional & running_mean, const c10::optional & running_var, const c10::optional & save_mean, const c10::optional & save_invstd, bool train, double eps, ::std::array output_mask) { + return at::_ops::native_batch_norm_backward_out::call(grad_out, input, weight, running_mean, running_var, save_mean, save_invstd, train, eps, output_mask, out0, out1, out2); +} +// aten::native_batch_norm_backward.out(Tensor grad_out, Tensor input, Tensor? weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_invstd, bool train, float eps, bool[3] output_mask, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!)) +inline ::std::tuple native_batch_norm_backward_outf(const at::Tensor & grad_out, const at::Tensor & input, const c10::optional & weight, const c10::optional & running_mean, const c10::optional & running_var, const c10::optional & save_mean, const c10::optional & save_invstd, bool train, double eps, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2) { + return at::_ops::native_batch_norm_backward_out::call(grad_out, input, weight, running_mean, running_var, save_mean, save_invstd, train, eps, output_mask, out0, out1, out2); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_layer_norm_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_layer_norm_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..814a88d116834f88290c3c8f9fe4f1ab8942bc2f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_layer_norm_backward_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 native_layer_norm_backward_out_symint(const at::Tensor & grad_out, const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const at::Tensor & mean, const at::Tensor & rstd, const c10::optional & weight, const c10::optional & bias, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); +TORCH_API ::std::tuple layer_norm_backward_cpu(const at::Tensor & grad_out, const at::Tensor & input, at::IntArrayRef normalized_shape, const at::Tensor & mean, const at::Tensor & rstd, const c10::optional & weight, const c10::optional & bias, ::std::array output_mask); +TORCH_API ::std::tuple layer_norm_backward_cuda(const at::Tensor & grad_out, const at::Tensor & input, at::IntArrayRef normalized_shape, const at::Tensor & mean, const at::Tensor & rstd, const c10::optional & weight, const c10::optional & bias, ::std::array output_mask); +TORCH_API ::std::tuple layer_norm_backward_nested(const at::Tensor & grad_out, const at::Tensor & input, at::IntArrayRef normalized_shape, const at::Tensor & mean, const at::Tensor & rstd, const c10::optional & weight, const c10::optional & bias, ::std::array output_mask); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_norm_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_norm_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..09e3fe7a74fb363bd1b39d438922be67e03fffe0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_norm_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & native_norm_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & p=2); +TORCH_API at::Tensor & native_norm_outf(const at::Tensor & self, const at::Scalar & p, at::Tensor & out); +TORCH_API at::Tensor & native_norm_out(at::Tensor & out, const at::Tensor & self, const c10::optional & p, at::IntArrayRef dim, bool keepdim, c10::optional dtype); +TORCH_API at::Tensor & native_norm_outf(const at::Tensor & self, const c10::optional & p, at::IntArrayRef dim, bool keepdim, c10::optional dtype, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/new_empty_strided_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/new_empty_strided_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..90d1b56b8c67fe6783ff09327fde60ec696d3ac4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/new_empty_strided_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & new_empty_strided_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef size, at::IntArrayRef stride); +TORCH_API at::Tensor & new_empty_strided_outf(const at::Tensor & self, at::IntArrayRef size, at::IntArrayRef stride, at::Tensor & out); +TORCH_API at::Tensor & new_empty_strided_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef size, c10::SymIntArrayRef stride); +TORCH_API at::Tensor & new_empty_strided_symint_outf(const at::Tensor & self, c10::SymIntArrayRef size, c10::SymIntArrayRef stride, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pow_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pow_native.h new file mode 100644 index 0000000000000000000000000000000000000000..b80eb5f1577e96b78cf5d9996c57c794f5f0f4b8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pow_native.h @@ -0,0 +1,31 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_pow_Tensor_Tensor_out : public at::meta::structured_pow_Tensor_Tensor { +void impl(const at::Tensor & self, const at::Tensor & exponent, const at::Tensor & out); +}; +struct TORCH_API structured_pow_Scalar_out : public at::meta::structured_pow_Scalar { +void impl(const at::Scalar & self, const at::Tensor & exponent, const at::Tensor & out); +}; +struct TORCH_API structured_pow_Tensor_Scalar_out : public at::meta::structured_pow_Tensor_Scalar { +void impl(const at::Tensor & self, const at::Scalar & exponent, const at::Tensor & out); +}; +TORCH_API at::Tensor pow_sparse_scalar(const at::Tensor & self, const at::Scalar & exponent); +TORCH_API at::Tensor & pow_out_sparse_scalar(const at::Tensor & self, const at::Scalar & exponent, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rand_like_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rand_like_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..6378e1436c1bd0c5694da4c5704b93fdc3ba7899 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rand_like_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 rand_like { + using schema = at::Tensor (const at::Tensor &, c10::optional, c10::optional, c10::optional, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::rand_like") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "rand_like(Tensor self, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format); +}; + +struct TORCH_API rand_like_out { + using schema = at::Tensor & (const at::Tensor &, 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::rand_like") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "rand_like.out(Tensor self, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, c10::optional memory_format, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::optional memory_format, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/replication_pad3d_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/replication_pad3d_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..4855552acf96c1053084147bc9ed1356a5ec2904 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/replication_pad3d_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 replication_pad3d(const at::Tensor & self, at::IntArrayRef padding); +TORCH_API at::Tensor replication_pad3d_symint(const at::Tensor & self, c10::SymIntArrayRef padding); +TORCH_API at::Tensor & replication_pad3d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef padding); +TORCH_API at::Tensor & replication_pad3d_outf(const at::Tensor & self, at::IntArrayRef padding, at::Tensor & out); +TORCH_API at::Tensor & replication_pad3d_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef padding); +TORCH_API at::Tensor & replication_pad3d_symint_outf(const at::Tensor & self, c10::SymIntArrayRef padding, at::Tensor & out); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/resize.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/resize.h new file mode 100644 index 0000000000000000000000000000000000000000..2121dc2824a1674c7ed70e9c007d093c0ac59f3f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/resize.h @@ -0,0 +1,105 @@ +#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 { + + +namespace symint { + template ::value>> + const at::Tensor & resize_(const at::Tensor & self, at::IntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize_::call(self, c10::fromIntArrayRefSlow(size), memory_format); + } +} + +namespace symint { + template ::value>> + const at::Tensor & resize_(const at::Tensor & self, c10::SymIntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize_::call(self, size, memory_format); + } +} + +// aten::resize.out(Tensor self, SymInt[] size, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline const at::Tensor & resize_out(const at::Tensor & out, const at::Tensor & self, at::IntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize_out::call(self, c10::fromIntArrayRefSlow(size), memory_format, out); +} +namespace symint { + template ::value>> + const at::Tensor & resize_out(const at::Tensor & out, const at::Tensor & self, at::IntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize_out::call(self, c10::fromIntArrayRefSlow(size), memory_format, out); + } +} + +// aten::resize.out(Tensor self, SymInt[] size, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline const at::Tensor & resize_outf(const at::Tensor & self, at::IntArrayRef size, c10::optional memory_format, const at::Tensor & out) { + return at::_ops::resize_out::call(self, c10::fromIntArrayRefSlow(size), memory_format, out); +} +namespace symint { + template ::value>> + const at::Tensor & resize_outf(const at::Tensor & self, at::IntArrayRef size, c10::optional memory_format, const at::Tensor & out) { + return at::_ops::resize_out::call(self, c10::fromIntArrayRefSlow(size), memory_format, out); + } +} + +// aten::resize.out(Tensor self, SymInt[] size, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline const at::Tensor & resize_symint_out(const at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize_out::call(self, size, memory_format, out); +} +namespace symint { + template ::value>> + const at::Tensor & resize_out(const at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize_out::call(self, size, memory_format, out); + } +} + +// aten::resize.out(Tensor self, SymInt[] size, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!) +inline const at::Tensor & resize_symint_outf(const at::Tensor & self, c10::SymIntArrayRef size, c10::optional memory_format, const at::Tensor & out) { + return at::_ops::resize_out::call(self, size, memory_format, out); +} +namespace symint { + template ::value>> + const at::Tensor & resize_outf(const at::Tensor & self, c10::SymIntArrayRef size, c10::optional memory_format, const at::Tensor & out) { + return at::_ops::resize_out::call(self, size, memory_format, out); + } +} + +// aten::resize(Tensor self, SymInt[] size, *, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor resize(const at::Tensor & self, at::IntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize::call(self, c10::fromIntArrayRefSlow(size), memory_format); +} +namespace symint { + template ::value>> + at::Tensor resize(const at::Tensor & self, at::IntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize::call(self, c10::fromIntArrayRefSlow(size), memory_format); + } +} + +// aten::resize(Tensor self, SymInt[] size, *, MemoryFormat? memory_format=None) -> Tensor +inline at::Tensor resize_symint(const at::Tensor & self, c10::SymIntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize::call(self, size, memory_format); +} +namespace symint { + template ::value>> + at::Tensor resize(const at::Tensor & self, c10::SymIntArrayRef size, c10::optional memory_format=c10::nullopt) { + return at::_ops::resize::call(self, size, memory_format); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/select_backward_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/select_backward_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..5943043edf69bcf684e26b0f8d6aeb7c3455fcb1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/select_backward_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & select_backward_out(at::Tensor & out, const at::Tensor & grad_output, at::IntArrayRef input_sizes, int64_t dim, int64_t index); +TORCH_API at::Tensor & select_backward_outf(const at::Tensor & grad_output, at::IntArrayRef input_sizes, int64_t dim, int64_t index, at::Tensor & out); +TORCH_API at::Tensor & select_backward_symint_out(at::Tensor & out, const at::Tensor & grad_output, c10::SymIntArrayRef input_sizes, int64_t dim, c10::SymInt index); +TORCH_API at::Tensor & select_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef input_sizes, int64_t dim, c10::SymInt index, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softplus_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softplus_native.h new file mode 100644 index 0000000000000000000000000000000000000000..e7b8d071054b8dfcf0fd54188b13920b1cceff69 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softplus_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_softplus_out : public at::meta::structured_softplus { +void impl(const at::Tensor & self, const at::Scalar & beta, const at::Scalar & threshold, const at::Tensor & out); +}; +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_ndtr_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_ndtr_native.h new file mode 100644 index 0000000000000000000000000000000000000000..b28a5fffa2e28402ca6cc89ebb0e4d45cdfd4152 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_ndtr_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor special_ndtr(const at::Tensor & self); +TORCH_API at::Tensor & special_ndtr_out(const at::Tensor & self, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_shifted_chebyshev_polynomial_u_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_shifted_chebyshev_polynomial_u_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..d4e8afac66c70c9584f4db0d55cf03ea8ea69b11 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_shifted_chebyshev_polynomial_u_ops.h @@ -0,0 +1,83 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API special_shifted_chebyshev_polynomial_u { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_shifted_chebyshev_polynomial_u") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_shifted_chebyshev_polynomial_u(Tensor x, Tensor n) -> Tensor") + static at::Tensor call(const at::Tensor & x, const at::Tensor & n); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & x, const at::Tensor & n); +}; + +struct TORCH_API special_shifted_chebyshev_polynomial_u_x_scalar { + using schema = at::Tensor (const at::Scalar &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_shifted_chebyshev_polynomial_u") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "x_scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_shifted_chebyshev_polynomial_u.x_scalar(Scalar x, Tensor n) -> Tensor") + static at::Tensor call(const at::Scalar & x, const at::Tensor & n); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & x, const at::Tensor & n); +}; + +struct TORCH_API special_shifted_chebyshev_polynomial_u_n_scalar { + 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::special_shifted_chebyshev_polynomial_u") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "n_scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_shifted_chebyshev_polynomial_u.n_scalar(Tensor x, Scalar n) -> Tensor") + static at::Tensor call(const at::Tensor & x, const at::Scalar & n); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & x, const at::Scalar & n); +}; + +struct TORCH_API special_shifted_chebyshev_polynomial_u_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_shifted_chebyshev_polynomial_u") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_shifted_chebyshev_polynomial_u.out(Tensor x, Tensor n, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & x, const at::Tensor & n, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & x, const at::Tensor & n, at::Tensor & out); +}; + +struct TORCH_API special_shifted_chebyshev_polynomial_u_x_scalar_out { + using schema = at::Tensor & (const at::Scalar &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_shifted_chebyshev_polynomial_u") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "x_scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_shifted_chebyshev_polynomial_u.x_scalar_out(Scalar x, Tensor n, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Scalar & x, const at::Tensor & n, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & x, const at::Tensor & n, at::Tensor & out); +}; + +struct TORCH_API special_shifted_chebyshev_polynomial_u_n_scalar_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::special_shifted_chebyshev_polynomial_u") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "n_scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_shifted_chebyshev_polynomial_u.n_scalar_out(Tensor x, Scalar n, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & x, const at::Scalar & n, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & x, const at::Scalar & n, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_sinc_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_sinc_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..c1b9e30bcede0db64f456136754ffec1056fb88c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_sinc_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 special_sinc { + using schema = at::Tensor (const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_sinc") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_sinc(Tensor self) -> Tensor") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +struct TORCH_API special_sinc_out { + using schema = at::Tensor & (const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::special_sinc") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "special_sinc.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/squeeze_copy_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/squeeze_copy_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8b7eb181051fc30d3e5e8745be9e1b718999e292 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/squeeze_copy_compositeexplicitautogradnonfunctional_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 compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor squeeze_copy(const at::Tensor & self); +TORCH_API at::Tensor squeeze_copy(const at::Tensor & self, int64_t dim); +TORCH_API at::Tensor squeeze_copy(const at::Tensor & self, at::IntArrayRef dim); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_padded_tensor.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_padded_tensor.h new file mode 100644 index 0000000000000000000000000000000000000000..21f2d6f7aa1e48b28275340948e0a77fc0590a5a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_padded_tensor.h @@ -0,0 +1,83 @@ +#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 { + + +namespace symint { + template ::value>> + at::Tensor to_padded_tensor(const at::Tensor & self, double padding, at::OptionalIntArrayRef output_size=c10::nullopt) { + return at::_ops::to_padded_tensor::call(self, padding, output_size.has_value() ? c10::make_optional(c10::fromIntArrayRefSlow(*output_size)) : c10::nullopt); + } +} + +namespace symint { + template ::value>> + at::Tensor to_padded_tensor(const at::Tensor & self, double padding, at::OptionalSymIntArrayRef output_size=c10::nullopt) { + return at::_ops::to_padded_tensor::call(self, padding, output_size); + } +} + +// aten::to_padded_tensor.out(Tensor self, float padding, SymInt[]? output_size=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & to_padded_tensor_out(at::Tensor & out, const at::Tensor & self, double padding, at::OptionalIntArrayRef output_size=c10::nullopt) { + return at::_ops::to_padded_tensor_out::call(self, padding, output_size.has_value() ? c10::make_optional(c10::fromIntArrayRefSlow(*output_size)) : c10::nullopt, out); +} +namespace symint { + template ::value>> + at::Tensor & to_padded_tensor_out(at::Tensor & out, const at::Tensor & self, double padding, at::OptionalIntArrayRef output_size=c10::nullopt) { + return at::_ops::to_padded_tensor_out::call(self, padding, output_size.has_value() ? c10::make_optional(c10::fromIntArrayRefSlow(*output_size)) : c10::nullopt, out); + } +} + +// aten::to_padded_tensor.out(Tensor self, float padding, SymInt[]? output_size=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & to_padded_tensor_outf(const at::Tensor & self, double padding, at::OptionalIntArrayRef output_size, at::Tensor & out) { + return at::_ops::to_padded_tensor_out::call(self, padding, output_size.has_value() ? c10::make_optional(c10::fromIntArrayRefSlow(*output_size)) : c10::nullopt, out); +} +namespace symint { + template ::value>> + at::Tensor & to_padded_tensor_outf(const at::Tensor & self, double padding, at::OptionalIntArrayRef output_size, at::Tensor & out) { + return at::_ops::to_padded_tensor_out::call(self, padding, output_size.has_value() ? c10::make_optional(c10::fromIntArrayRefSlow(*output_size)) : c10::nullopt, out); + } +} + +// aten::to_padded_tensor.out(Tensor self, float padding, SymInt[]? output_size=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & to_padded_tensor_symint_out(at::Tensor & out, const at::Tensor & self, double padding, at::OptionalSymIntArrayRef output_size=c10::nullopt) { + return at::_ops::to_padded_tensor_out::call(self, padding, output_size, out); +} +namespace symint { + template ::value>> + at::Tensor & to_padded_tensor_out(at::Tensor & out, const at::Tensor & self, double padding, at::OptionalSymIntArrayRef output_size=c10::nullopt) { + return at::_ops::to_padded_tensor_out::call(self, padding, output_size, out); + } +} + +// aten::to_padded_tensor.out(Tensor self, float padding, SymInt[]? output_size=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & to_padded_tensor_symint_outf(const at::Tensor & self, double padding, at::OptionalSymIntArrayRef output_size, at::Tensor & out) { + return at::_ops::to_padded_tensor_out::call(self, padding, output_size, out); +} +namespace symint { + template ::value>> + at::Tensor & to_padded_tensor_outf(const at::Tensor & self, double padding, at::OptionalSymIntArrayRef output_size, at::Tensor & out) { + return at::_ops::to_padded_tensor_out::call(self, padding, output_size, out); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unbind_copy_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unbind_copy_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..66be9d3fabf684b97acf382b385ba54d205dec8e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unbind_copy_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API void unbind_copy_out(at::TensorList out, const at::Tensor & self, int64_t dim=0); +TORCH_API void unbind_copy_outf(const at::Tensor & self, int64_t dim, at::TensorList out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/uniform_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/uniform_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..965546f6281f135ddcf241fe1b5919e4a6d6eea4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/uniform_meta_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor & uniform_(at::Tensor & self, double from=0, double to=1, c10::optional generator=c10::nullopt); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bilinear2d_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bilinear2d_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..775f364d7b4cdd0d3a16ae1b47915a22919262cb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_bilinear2d_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_bilinear2d(const at::Tensor & self, at::IntArrayRef output_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor upsample_bilinear2d_symint(const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_bilinear2d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_bilinear2d_outf(const at::Tensor & self, at::IntArrayRef output_size, bool align_corners, c10::optional scales_h, c10::optional scales_w, at::Tensor & out); +TORCH_API at::Tensor & upsample_bilinear2d_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & upsample_bilinear2d_symint_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, 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/values_copy_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/values_copy_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..1adfee09a3c2e7d76caca99f1f327ebe2c8140a0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/values_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 values_copy { + using schema = at::Tensor (const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::values_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "values_copy(Tensor self) -> Tensor") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +struct TORCH_API values_copy_out { + using schema = at::Tensor & (const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::values_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "values_copy.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Tensor & out); +}; + +}} // namespace at::_ops