diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_amp_foreach_non_finite_check_and_unscale_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_amp_foreach_non_finite_check_and_unscale_native.h new file mode 100644 index 0000000000000000000000000000000000000000..41da53fee3b34c884562a2e21f8c264914278a05 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_amp_foreach_non_finite_check_and_unscale_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<::std::vector,at::Tensor> _amp_foreach_non_finite_check_and_unscale(at::TensorList self, const at::Tensor & found_inf, const at::Tensor & inv_scale); +TORCH_API void _amp_foreach_non_finite_check_and_unscale_out(at::TensorList self, at::Tensor & found_inf, const at::Tensor & inv_scale, at::TensorList out); +TORCH_API void _amp_foreach_non_finite_check_and_unscale_cpu_(at::TensorList self, at::Tensor & found_inf, const at::Tensor & inv_scale); +TORCH_API void _amp_foreach_non_finite_check_and_unscale_cuda_(at::TensorList self, at::Tensor & found_inf, const at::Tensor & inv_scale); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_impl_index_backward_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_impl_index_backward_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..944890c75838e2fd4374aeeafb40c77a874c5399 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_impl_index_backward_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 ::std::tuple _batch_norm_impl_index_backward(int64_t impl_index, const at::Tensor & input, const at::Tensor & grad_output, const c10::optional & weight, const c10::optional & running_mean, const c10::optional & running_var, const c10::optional & save_mean, const c10::optional & save_var_transform, bool train, double eps, ::std::array output_mask, const at::Tensor & reservedSpace); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_impl_index_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_impl_index_native.h new file mode 100644 index 0000000000000000000000000000000000000000..4a4c33cb1969a86a38963763fddee47e09dcdbd8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_impl_index_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 ::std::tuple _batch_norm_impl_index(const at::Tensor & input, const c10::optional & weight, const c10::optional & bias, const c10::optional & running_mean, const c10::optional & running_var, bool training, double momentum, double eps, bool cudnn_enabled); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convolution_mode_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convolution_mode_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8044e7a33056fdf93d50ebf924210c33b398b418 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convolution_mode_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 _convolution_mode(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, c10::string_view padding, at::IntArrayRef dilation, int64_t groups); +TORCH_API at::Tensor _convolution_mode_symint(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::string_view padding, c10::SymIntArrayRef dilation, c10::SymInt groups); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_debug_has_internal_overlap.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_debug_has_internal_overlap.h new file mode 100644 index 0000000000000000000000000000000000000000..b190cedb62c83dbe2ba61714ea7691c09c503547 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_debug_has_internal_overlap.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::_debug_has_internal_overlap(Tensor self) -> int +inline int64_t _debug_has_internal_overlap(const at::Tensor & self) { + return at::_ops::_debug_has_internal_overlap::call(self); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_clamp_max_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_clamp_max_native.h new file mode 100644 index 0000000000000000000000000000000000000000..93fe48906bd7d3199217209ef8c58b69b7a5ef81 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_clamp_max_native.h @@ -0,0 +1,35 @@ +#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_clamp_max_Scalar_out(at::TensorList self, const at::Scalar & scalar, at::TensorList out); +TORCH_API ::std::vector foreach_tensor_clamp_max_scalar_kernel_slow(at::TensorList self, const at::Scalar & scalar); +TORCH_API void foreach_tensor_clamp_max_scalar_kernel_slow_(at::TensorList self, const at::Scalar & scalar); +TORCH_API ::std::vector foreach_tensor_clamp_max_scalar_kernel_cuda(at::TensorList self, const at::Scalar & scalar); +TORCH_API void foreach_tensor_clamp_max_scalar_kernel_cuda_(at::TensorList self, const at::Scalar & scalar); +TORCH_API void _foreach_clamp_max_List_out(at::TensorList self, at::TensorList other, at::TensorList out); +TORCH_API ::std::vector foreach_tensor_clamp_max_list_kernel_slow(at::TensorList self, at::TensorList other); +TORCH_API void foreach_tensor_clamp_max_list_kernel_slow_(at::TensorList self, at::TensorList other); +TORCH_API ::std::vector foreach_tensor_clamp_max_list_kernel_cuda(at::TensorList self, at::TensorList other); +TORCH_API void foreach_tensor_clamp_max_list_kernel_cuda_(at::TensorList self, at::TensorList other); +TORCH_API void _foreach_clamp_max_ScalarList_out(at::TensorList self, at::ArrayRef scalars, at::TensorList out); +TORCH_API ::std::vector foreach_tensor_clamp_max_scalarlist_kernel_slow(at::TensorList self, at::ArrayRef scalars); +TORCH_API void foreach_tensor_clamp_max_scalarlist_kernel_slow_(at::TensorList self, at::ArrayRef scalars); +TORCH_API ::std::vector foreach_tensor_clamp_max_scalarlist_kernel_cuda(at::TensorList self, at::ArrayRef scalars); +TORCH_API void foreach_tensor_clamp_max_scalarlist_kernel_cuda_(at::TensorList self, at::ArrayRef scalars); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_exp_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_exp_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..3336692ddb41f4dbb6532842233f09e890c9b86f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_exp_cuda_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::vector _foreach_exp(at::TensorList self); +TORCH_API void _foreach_exp_(at::TensorList self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log2_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log2_native.h new file mode 100644 index 0000000000000000000000000000000000000000..0b64f1804ede7042c313f091a017dc65adda2083 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log2_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_log2_out(at::TensorList self, at::TensorList out); +TORCH_API ::std::vector foreach_tensor_log2_slow(at::TensorList self); +TORCH_API void foreach_tensor_log2_slow_(at::TensorList self); +TORCH_API ::std::vector foreach_tensor_log2_cuda(at::TensorList self); +TORCH_API void foreach_tensor_log2_cuda_(at::TensorList self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_tan_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_tan_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..fa6c0437fae5e206a9dc56e9390e40b5278925c6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_tan_cpu_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API ::std::vector _foreach_tan(at::TensorList self); +TORCH_API void _foreach_tan_(at::TensorList self); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_trunc_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_trunc_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..2e76048965ddab344ad9285f932ca154881df83e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_trunc_cuda_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::vector _foreach_trunc(at::TensorList self); +TORCH_API void _foreach_trunc_(at::TensorList self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_adamw_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_adamw_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c5ad41b10890d56b9c0a0b642d675c5ade57a761 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_adamw_compositeexplicitautograd_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 compositeexplicitautograd { + +TORCH_API ::std::tuple<::std::vector,::std::vector,::std::vector,::std::vector,::std::vector> _fused_adamw(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const c10::optional & grad_scale={}, const c10::optional & found_inf={}); +TORCH_API void _fused_adamw_out(at::TensorList out, at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const c10::optional & grad_scale={}, const c10::optional & found_inf={}); +TORCH_API void _fused_adamw_outf(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const c10::optional & grad_scale, const c10::optional & found_inf, at::TensorList out); +TORCH_API ::std::tuple<::std::vector,::std::vector,::std::vector,::std::vector,::std::vector> _fused_adamw(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const c10::optional & grad_scale={}, const c10::optional & found_inf={}); +TORCH_API void _fused_adamw_out(at::TensorList out, at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const c10::optional & grad_scale={}, const c10::optional & found_inf={}); +TORCH_API void _fused_adamw_outf(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const c10::optional & grad_scale, const c10::optional & found_inf, at::TensorList out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_dropout_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_dropout_native.h new file mode 100644 index 0000000000000000000000000000000000000000..088312fc2a2bf33259665f10dc3159234f294771 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_dropout_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 _fused_dropout_out(const at::Tensor & self, double p, c10::optional generator, at::Tensor & out0, at::Tensor & out1); +TORCH_API ::std::tuple fused_dropout_cuda(const at::Tensor & self, double p, c10::optional generator=c10::nullopt); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_eigvals_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_eigvals_native.h new file mode 100644 index 0000000000000000000000000000000000000000..3235cbef2f1e5ec863211f9183cffae97f456460 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_eigvals_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 _linalg_eigvals(const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_slogdet_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_slogdet_native.h new file mode 100644 index 0000000000000000000000000000000000000000..05d92df30daf7040a848b0091b9bef2f078d971f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_slogdet_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_slogdet_out : public at::meta::structured__linalg_slogdet { +void impl(const at::Tensor & A, const at::Tensor & sign, const at::Tensor & logabsdet, const at::Tensor & LU, const at::Tensor & pivots); +}; +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_backward_data_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_backward_data_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..079f33e581df1fa69ec7ef77ce3e490217f12b35 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_backward_data_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 _log_softmax_backward_data { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, int64_t, at::ScalarType); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_log_softmax_backward_data") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_log_softmax_backward_data(Tensor grad_output, Tensor output, int dim, ScalarType input_dtype) -> Tensor") + static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype); +}; + +struct TORCH_API _log_softmax_backward_data_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, int64_t, at::ScalarType, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_log_softmax_backward_data") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_log_softmax_backward_data.out(Tensor grad_output, Tensor output, int dim, ScalarType input_dtype, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ec852764e1b940f141eee456d0bfc73133987ffa --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_log_softmax_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 _log_softmax(const at::Tensor & self, int64_t dim, bool half_to_float); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_slow_conv2d_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_slow_conv2d_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..086d41b79118a22f08d8bf035dbebf6e6d5bc457 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_slow_conv2d_backward_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 ::std::tuple slow_conv2d_backward_out_cpu(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::Tensor & grad_input, at::Tensor & grad_weight, at::Tensor & grad_bias); +TORCH_API ::std::tuple slow_conv2d_backward_out_cuda(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::Tensor & grad_input, at::Tensor & grad_weight, at::Tensor & grad_bias); +TORCH_API ::std::tuple _slow_conv2d_backward_output_mask_out_symint(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); +TORCH_API ::std::tuple slow_conv2d_backward_cpu(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_cuda(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); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_softmax_backward_data_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_softmax_backward_data_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8bae65384b773b1a6e5024971ceabbe6c762067b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_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_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_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/adaptive_max_pool2d_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..1d5976942a8760fa00b269000a1ad3b7d45605f8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_backward_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API adaptive_max_pool2d_backward_grad_input { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::adaptive_max_pool2d_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "adaptive_max_pool2d_backward.grad_input(Tensor grad_output, Tensor self, Tensor indices, *, Tensor(a!) grad_input) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & indices, at::Tensor & grad_input); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & indices, at::Tensor & grad_input); +}; + +struct TORCH_API adaptive_max_pool2d_backward { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::adaptive_max_pool2d_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "adaptive_max_pool2d_backward(Tensor grad_output, Tensor self, Tensor indices) -> Tensor") + static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & indices); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & indices); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..215891fb78f2d629f6a5af4398a06f4a8b25adab --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_adaptive_max_pool2d : public at::impl::MetaBase { + + + void meta(const at::Tensor & self, at::IntArrayRef output_size); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bartlett_window.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bartlett_window.h new file mode 100644 index 0000000000000000000000000000000000000000..7ba80c1ed3842a4c0e9ee2eedc6edcb0e19018d0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bartlett_window.h @@ -0,0 +1,61 @@ +#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::bartlett_window(int window_length, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor bartlett_window(int64_t window_length, at::TensorOptions options={}) { + return at::_ops::bartlett_window::call(window_length, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt()); +} +// aten::bartlett_window(int window_length, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor bartlett_window(int64_t window_length, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory) { + return at::_ops::bartlett_window::call(window_length, dtype, layout, device, pin_memory); +} + +// aten::bartlett_window.periodic(int window_length, bool periodic, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor bartlett_window(int64_t window_length, bool periodic, at::TensorOptions options={}) { + return at::_ops::bartlett_window_periodic::call(window_length, periodic, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt()); +} +// aten::bartlett_window.periodic(int window_length, bool periodic, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor +inline at::Tensor bartlett_window(int64_t window_length, bool periodic, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory) { + return at::_ops::bartlett_window_periodic::call(window_length, periodic, dtype, layout, device, pin_memory); +} + +// aten::bartlett_window.out(int window_length, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & bartlett_window_out(at::Tensor & out, int64_t window_length) { + return at::_ops::bartlett_window_out::call(window_length, out); +} +// aten::bartlett_window.out(int window_length, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & bartlett_window_outf(int64_t window_length, at::Tensor & out) { + return at::_ops::bartlett_window_out::call(window_length, out); +} + +// aten::bartlett_window.periodic_out(int window_length, bool periodic, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & bartlett_window_out(at::Tensor & out, int64_t window_length, bool periodic) { + return at::_ops::bartlett_window_periodic_out::call(window_length, periodic, out); +} +// aten::bartlett_window.periodic_out(int window_length, bool periodic, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & bartlett_window_outf(int64_t window_length, bool periodic, at::Tensor & out) { + return at::_ops::bartlett_window_periodic_out::call(window_length, periodic, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/batch_norm_backward_elemt_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/batch_norm_backward_elemt_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..3ca62b525a31fe0c9133aacf52db0a507ae0e564 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/batch_norm_backward_elemt_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 & batch_norm_backward_elemt_out(at::Tensor & out, const at::Tensor & grad_out, const at::Tensor & input, const at::Tensor & mean, const at::Tensor & invstd, const c10::optional & weight, const at::Tensor & sum_dy, const at::Tensor & sum_dy_xmu, const at::Tensor & count); +TORCH_API at::Tensor & batch_norm_backward_elemt_outf(const at::Tensor & grad_out, const at::Tensor & input, const at::Tensor & mean, const at::Tensor & invstd, const c10::optional & weight, const at::Tensor & sum_dy, const at::Tensor & sum_dy_xmu, const at::Tensor & count, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bincount_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bincount_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..f290700d4ee6984e8e8e16c678957736ee92c600 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bincount_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 bincount(const at::Tensor & self, const c10::optional & weights={}, int64_t minlength=0); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_or_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_or_native.h new file mode 100644 index 0000000000000000000000000000000000000000..268e2b6462bb99e73fbdc8bf07e4c6a2eca32c00 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_or_native.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_bitwise_or_out : public at::meta::structured_bitwise_or_Tensor { +void impl(const at::Tensor & self, const at::Tensor & other, const at::Tensor & out); +}; +TORCH_API at::Tensor bitwise_or(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & bitwise_or_out(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +TORCH_API at::Tensor & bitwise_or_(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor bitwise_or(const at::Scalar & self, const at::Tensor & other); +TORCH_API at::Tensor & bitwise_or_Scalar_Tensor_out(const at::Scalar & self, const at::Tensor & other, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cdist_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cdist_native.h new file mode 100644 index 0000000000000000000000000000000000000000..81d03bd67f6b8af15ad56bf5b71b91690dacf703 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cdist_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 cdist(const at::Tensor & x1, const at::Tensor & x2, double p=2, c10::optional compute_mode=c10::nullopt); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cholesky_inverse.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cholesky_inverse.h new file mode 100644 index 0000000000000000000000000000000000000000..68443b394857aa8d547e857ce52999e5eea176cb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cholesky_inverse.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::cholesky_inverse(Tensor self, bool upper=False) -> Tensor +inline at::Tensor cholesky_inverse(const at::Tensor & self, bool upper=false) { + return at::_ops::cholesky_inverse::call(self, upper); +} + +// aten::cholesky_inverse.out(Tensor self, bool upper=False, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cholesky_inverse_out(at::Tensor & out, const at::Tensor & self, bool upper=false) { + return at::_ops::cholesky_inverse_out::call(self, upper, out); +} +// aten::cholesky_inverse.out(Tensor self, bool upper=False, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cholesky_inverse_outf(const at::Tensor & self, bool upper, at::Tensor & out) { + return at::_ops::cholesky_inverse_out::call(self, upper, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/contiguous_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/contiguous_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..c4f554ad217d17f3b7af13cf884f810f0fc5a220 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/contiguous_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 contiguous { + using schema = at::Tensor (const at::Tensor &, at::MemoryFormat); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::contiguous") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "contiguous(Tensor(a) self, *, MemoryFormat memory_format=contiguous_format) -> Tensor(a)") + static at::Tensor call(const at::Tensor & self, at::MemoryFormat memory_format); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::MemoryFormat memory_format); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumulative_trapezoid_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumulative_trapezoid_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0e78ed19fef37909b2a0d7b5f87fecacdf735909 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumulative_trapezoid_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 cumulative_trapezoid(const at::Tensor & y, const at::Tensor & x, int64_t dim=-1); +TORCH_API at::Tensor cumulative_trapezoid(const at::Tensor & y, const at::Scalar & dx=1, int64_t dim=-1); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_linear_int8_weight_fp32_activation_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_linear_int8_weight_fp32_activation_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..2d66e8b04e5db132e860f74d59aad7a4b8ff4106 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_linear_int8_weight_fp32_activation_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor fbgemm_linear_int8_weight_fp32_activation(const at::Tensor & input, const at::Tensor & weight, const at::Tensor & packed, const at::Tensor & col_offsets, const at::Scalar & weight_scale, const at::Scalar & weight_zero_point, const at::Tensor & bias); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_fftn_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_fftn_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..5123627879e168151a8b9a22440101658c9ac0e4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_fftn_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_fftn(const at::Tensor & self, at::OptionalIntArrayRef s=c10::nullopt, at::OptionalIntArrayRef dim=c10::nullopt, c10::optional norm=c10::nullopt); +TORCH_API at::Tensor fft_fftn_symint(const at::Tensor & self, at::OptionalSymIntArrayRef s=c10::nullopt, at::OptionalIntArrayRef dim=c10::nullopt, c10::optional norm=c10::nullopt); +TORCH_API at::Tensor & fft_fftn_out(at::Tensor & out, const at::Tensor & self, at::OptionalIntArrayRef s=c10::nullopt, at::OptionalIntArrayRef dim=c10::nullopt, c10::optional norm=c10::nullopt); +TORCH_API at::Tensor & fft_fftn_outf(const at::Tensor & self, at::OptionalIntArrayRef s, at::OptionalIntArrayRef dim, c10::optional norm, at::Tensor & out); +TORCH_API at::Tensor & fft_fftn_symint_out(at::Tensor & out, const at::Tensor & self, at::OptionalSymIntArrayRef s=c10::nullopt, at::OptionalIntArrayRef dim=c10::nullopt, c10::optional norm=c10::nullopt); +TORCH_API at::Tensor & fft_fftn_symint_outf(const at::Tensor & self, at::OptionalSymIntArrayRef s, at::OptionalIntArrayRef 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/floor_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/floor_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..058fdbb796fc12e1238b3a76ade4575504b0f0f0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/floor_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 floor(const at::Tensor & self); +TORCH_API at::Tensor & floor_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & floor_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & floor_(at::Tensor & self); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/frac_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/frac_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..38ac00e29c3b6debf665298fa897277bcdf9c8c5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/frac_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_frac : public TensorIteratorBase { + + + void meta(const at::Tensor & self); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ge_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ge_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..08d64b25aec0e85534e321c9ba402d33f7692003 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ge_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 ge_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::ge") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "ge.Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +}; + +struct TORCH_API ge_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::ge") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "ge.Scalar(Tensor self, Scalar other) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Scalar & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Scalar & other); +}; + +struct TORCH_API ge_Tensor_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::ge") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "ge.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +}; + +struct TORCH_API ge_Tensor { + 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::ge") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "ge.Tensor(Tensor self, Tensor other) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other); +}; + +struct TORCH_API ge__Scalar { + using schema = at::Tensor & (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::ge_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "ge_.Scalar(Tensor(a!) self, Scalar other) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, const at::Scalar & other); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, const at::Scalar & other); +}; + +struct TORCH_API ge__Tensor { + using schema = at::Tensor & (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::ge_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "ge_.Tensor(Tensor(a!) self, Tensor other) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, const at::Tensor & other); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, const at::Tensor & other); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gt_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gt_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..5ff67ef575640c5baa83463aef7c3e4bb0904923 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gt_cuda_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 cuda { + +TORCH_API at::Tensor gt(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & gt_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & gt_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +TORCH_API at::Tensor & gt_(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor gt(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & gt_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & gt_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & gt_(at::Tensor & self, const at::Tensor & other); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardsigmoid_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardsigmoid_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ddedaf4cb1de8e242b8313803308e3c73eb75a42 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardsigmoid_backward_cuda_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor hardsigmoid_backward(const at::Tensor & grad_output, const at::Tensor & self); +TORCH_API at::Tensor & hardsigmoid_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self); +TORCH_API at::Tensor & hardsigmoid_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & grad_input); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/igammac_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/igammac_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..a3e479bdccd111a7f1d336dda768241b9c8919bf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/igammac_ops.h @@ -0,0 +1,50 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API igammac_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::igammac") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "igammac.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +}; + +struct TORCH_API igammac { + 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::igammac") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "igammac(Tensor self, Tensor other) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other); +}; + +struct TORCH_API igammac_ { + using schema = at::Tensor & (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::igammac_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "igammac_(Tensor(a!) self, Tensor other) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, const at::Tensor & other); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, const at::Tensor & other); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_select_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_select_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..771318844ccbb86af2de4dbbf4226a3450314e72 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_select_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 index_select(const at::Tensor & self, int64_t dim, const at::Tensor & index); +TORCH_API at::Tensor & index_select_out(at::Tensor & out, const at::Tensor & self, int64_t dim, const at::Tensor & index); +TORCH_API at::Tensor & index_select_outf(const at::Tensor & self, int64_t dim, const at::Tensor & index, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/inner_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/inner_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..2146bb8226cf21e71f1e7ac9a044b8e95e4918ad --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/inner_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 inner { + 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::inner") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "inner(Tensor self, Tensor other) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other); +}; + +struct TORCH_API inner_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::inner") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "inner.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/instance_norm.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/instance_norm.h new file mode 100644 index 0000000000000000000000000000000000000000..8373799b35a254400d5815eaa852944f9fc0b343 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/instance_norm.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::instance_norm(Tensor input, Tensor? weight, Tensor? bias, Tensor? running_mean, Tensor? running_var, bool use_input_stats, float momentum, float eps, bool cudnn_enabled) -> Tensor +inline at::Tensor instance_norm(const at::Tensor & input, const c10::optional & weight, const c10::optional & bias, const c10::optional & running_mean, const c10::optional & running_var, bool use_input_stats, double momentum, double eps, bool cudnn_enabled) { + return at::_ops::instance_norm::call(input, weight, bias, running_mean, running_var, use_input_stats, momentum, eps, cudnn_enabled); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_vulkan_available_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_vulkan_available_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..67645c74f872ed505d143f35e6b67929eb133236 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_vulkan_available_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_vulkan_available(); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isreal_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isreal_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a4e7e3a6ae918f437a65a7efa32ebca8ad052dd9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isreal_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor isreal(const at::Tensor & self); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lstsq_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lstsq_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..a4bae4781dd3089499e07d18467a539d3e531c85 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lstsq_compositeexplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API ::std::tuple linalg_lstsq(const at::Tensor & self, const at::Tensor & b, c10::optional rcond=c10::nullopt, c10::optional driver=c10::nullopt); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lstsq_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lstsq_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..4f803b9a9fd41aeee20374f382d5f0de95ab70e0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lstsq_cpu_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API ::std::tuple linalg_lstsq_out(at::Tensor & solution, at::Tensor & residuals, at::Tensor & rank, at::Tensor & singular_values, const at::Tensor & self, const at::Tensor & b, c10::optional rcond=c10::nullopt, c10::optional driver=c10::nullopt); +TORCH_API ::std::tuple linalg_lstsq_outf(const at::Tensor & self, const at::Tensor & b, c10::optional rcond, c10::optional driver, at::Tensor & solution, at::Tensor & residuals, at::Tensor & rank, at::Tensor & singular_values); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_multi_dot_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_multi_dot_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..8b7b5f8329f39f67a3e037f8c9689cde3f82d909 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_multi_dot_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API linalg_multi_dot { + using schema = at::Tensor (at::TensorList); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_multi_dot") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_multi_dot(Tensor[] tensors) -> Tensor") + static at::Tensor call(at::TensorList tensors); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList tensors); +}; + +struct TORCH_API linalg_multi_dot_out { + using schema = at::Tensor & (at::TensorList, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_multi_dot") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_multi_dot.out(Tensor[] tensors, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(at::TensorList tensors, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList tensors, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_tensorsolve_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_tensorsolve_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..fa5ae0902298786ca7f4b0f5edf367c6943f0b30 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_tensorsolve_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API linalg_tensorsolve { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, at::OptionalIntArrayRef); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_tensorsolve") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_tensorsolve(Tensor self, Tensor other, int[]? dims=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & other, at::OptionalIntArrayRef dims); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, at::OptionalIntArrayRef dims); +}; + +struct TORCH_API linalg_tensorsolve_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, at::OptionalIntArrayRef, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_tensorsolve") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_tensorsolve.out(Tensor self, Tensor other, int[]? dims=None, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, at::OptionalIntArrayRef dims, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, at::OptionalIntArrayRef dims, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_xor_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_xor_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..2d587f1a4a38ad074d5e84659e00e39d9db55b83 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logical_xor_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 logical_xor(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & logical_xor_(at::Tensor & self, const at::Tensor & other); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_select_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_select_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..aa7e1b90f74ced952851dda55a2805563370c7a0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_select_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API masked_select_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::masked_select") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "masked_select.out(Tensor self, Tensor mask, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & mask, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & mask, at::Tensor & out); +}; + +struct TORCH_API masked_select { + 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::masked_select") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "masked_select(Tensor self, Tensor mask) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & mask); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & mask); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mul_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mul_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..620bb8fd62954006b2d7873d261a5fb463cc601f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mul_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 mul(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & mul_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & mul_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & mul_(at::Tensor & self, const at::Tensor & other); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/multilabel_margin_loss_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/multilabel_margin_loss_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..bba7d0ecbf4c171e540a9447cfa3f7589e1d6c39 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/multilabel_margin_loss_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 multilabel_margin_loss(const at::Tensor & self, const at::Tensor & target, int64_t reduction=at::Reduction::Mean); +TORCH_API at::Tensor & multilabel_margin_loss_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & target, int64_t reduction=at::Reduction::Mean); +TORCH_API at::Tensor & multilabel_margin_loss_outf(const at::Tensor & self, const at::Tensor & target, int64_t reduction, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mv_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mv_native.h new file mode 100644 index 0000000000000000000000000000000000000000..80dc717cdd6b0eb5776103e61ad34db1a6597bca --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mv_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 mv(const at::Tensor & self, const at::Tensor & vec); +TORCH_API at::Tensor & mv_out(const at::Tensor & self, const at::Tensor & vec, at::Tensor & out); +TORCH_API at::Tensor mv_sparse(const at::Tensor & self, const at::Tensor & vec); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/neg_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/neg_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..690d24220b768a96ea9c14035037407b5bcbb617 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/neg_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 neg(const at::Tensor & self); +TORCH_API at::Tensor & neg_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & neg_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & neg_(at::Tensor & self); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/polar_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/polar_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0f2a0c07cbcd7888f6c4100dbea45f742e2d738e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/polar_cpu_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor & polar_out(at::Tensor & out, const at::Tensor & abs, const at::Tensor & angle); +TORCH_API at::Tensor & polar_outf(const at::Tensor & abs, const at::Tensor & angle, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/polygamma_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/polygamma_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0243e48e27c4f8d6cc2fe0f4619498d28892b5c9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/polygamma_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor polygamma(int64_t n, const at::Tensor & self); +TORCH_API at::Tensor & polygamma_out(at::Tensor & out, int64_t n, const at::Tensor & self); +TORCH_API at::Tensor & polygamma_outf(int64_t n, const at::Tensor & self, at::Tensor & out); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/q_per_channel_zero_points_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/q_per_channel_zero_points_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..02e538c13e5a0bb044320787665a1f180159c13b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/q_per_channel_zero_points_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 q_per_channel_zero_points { + 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::q_per_channel_zero_points") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "q_per_channel_zero_points(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 q_per_channel_zero_points_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::q_per_channel_zero_points") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "q_per_channel_zero_points.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/quantized_rnn_relu_cell_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_rnn_relu_cell_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0b8cc6c660d8c4a9c23fe20c0c75a77f559edeef --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_rnn_relu_cell_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor quantized_rnn_relu_cell(const at::Tensor & input, const at::Tensor & hx, const at::Tensor & w_ih, const at::Tensor & w_hh, const at::Tensor & b_ih, const at::Tensor & b_hh, const at::Tensor & packed_ih, const at::Tensor & packed_hh, const at::Tensor & col_offsets_ih, const at::Tensor & col_offsets_hh, const at::Scalar & scale_ih, const at::Scalar & scale_hh, const at::Scalar & zero_point_ih, const at::Scalar & zero_point_hh); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randn_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randn_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ca843b40a3a2e1f879cd5d3eb2a5d48632faed7a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randn_compositeexplicitautograd_dispatch.h @@ -0,0 +1,46 @@ +#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 randn(at::IntArrayRef size, at::TensorOptions options={}); +TORCH_API at::Tensor randn(at::IntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor randn_symint(c10::SymIntArrayRef size, at::TensorOptions options={}); +TORCH_API at::Tensor randn_symint(c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor randn(at::IntArrayRef size, c10::optional generator, at::TensorOptions options={}); +TORCH_API at::Tensor randn(at::IntArrayRef size, c10::optional generator, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor randn_symint(c10::SymIntArrayRef size, c10::optional generator, at::TensorOptions options={}); +TORCH_API at::Tensor randn_symint(c10::SymIntArrayRef size, c10::optional generator, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor randn(at::IntArrayRef size, c10::optional names, at::TensorOptions options={}); +TORCH_API at::Tensor randn(at::IntArrayRef size, c10::optional names, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor randn_symint(c10::SymIntArrayRef size, c10::optional names, at::TensorOptions options={}); +TORCH_API at::Tensor randn_symint(c10::SymIntArrayRef size, c10::optional names, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor & randn_out(at::Tensor & out, at::IntArrayRef size, c10::optional names); +TORCH_API at::Tensor & randn_outf(at::IntArrayRef size, c10::optional names, at::Tensor & out); +TORCH_API at::Tensor & randn_symint_out(at::Tensor & out, c10::SymIntArrayRef size, c10::optional names); +TORCH_API at::Tensor & randn_symint_outf(c10::SymIntArrayRef size, c10::optional names, at::Tensor & out); +TORCH_API at::Tensor randn(at::IntArrayRef size, c10::optional generator, c10::optional names, at::TensorOptions options={}); +TORCH_API at::Tensor randn(at::IntArrayRef size, c10::optional generator, c10::optional names, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor randn_symint(c10::SymIntArrayRef size, c10::optional generator, c10::optional names, at::TensorOptions options={}); +TORCH_API at::Tensor randn_symint(c10::SymIntArrayRef size, c10::optional generator, c10::optional names, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +TORCH_API at::Tensor & randn_out(at::Tensor & out, at::IntArrayRef size, c10::optional generator, c10::optional names); +TORCH_API at::Tensor & randn_outf(at::IntArrayRef size, c10::optional generator, c10::optional names, at::Tensor & out); +TORCH_API at::Tensor & randn_symint_out(at::Tensor & out, c10::SymIntArrayRef size, c10::optional generator, c10::optional names); +TORCH_API at::Tensor & randn_symint_outf(c10::SymIntArrayRef size, c10::optional generator, c10::optional names, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reflection_pad2d_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reflection_pad2d_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..7cf53b66cacf2495a866b5f24e8afe1201ecffe6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reflection_pad2d_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 at::Tensor reflection_pad2d_backward_cpu(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding); +TORCH_API at::Tensor & reflection_pad2d_backward_out_cpu(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding, at::Tensor & grad_input); +TORCH_API at::Tensor reflection_pad2d_backward_cuda(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding); +TORCH_API at::Tensor & reflection_pad2d_backward_out_cuda(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding, at::Tensor & grad_input); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/resize_as_sparse_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/resize_as_sparse_native.h new file mode 100644 index 0000000000000000000000000000000000000000..beb06b63ff4c927d7bad11dbc81e3e7ebff682f3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/resize_as_sparse_native.h @@ -0,0 +1,24 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor resize_as_sparse(const at::Tensor & self, const at::Tensor & the_template); +TORCH_API const at::Tensor & resize_as_sparse_out(const at::Tensor & self, const at::Tensor & the_template, const at::Tensor & out); +TORCH_API const at::Tensor & resize_as_sparse_(const at::Tensor & self, const at::Tensor & the_template); +TORCH_API const at::Tensor & resize_as_sparse_compressed_(const at::Tensor & self, const at::Tensor & the_template); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/roll.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/roll.h new file mode 100644 index 0000000000000000000000000000000000000000..2a9a6fc898c619ad870667b1b4ec834b9f997908 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/roll.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::roll(Tensor self, SymInt[1] shifts, int[1] dims=[]) -> Tensor +inline at::Tensor roll(const at::Tensor & self, at::IntArrayRef shifts, at::IntArrayRef dims={}) { + return at::_ops::roll::call(self, c10::fromIntArrayRefSlow(shifts), dims); +} +namespace symint { + template ::value>> + at::Tensor roll(const at::Tensor & self, at::IntArrayRef shifts, at::IntArrayRef dims={}) { + return at::_ops::roll::call(self, c10::fromIntArrayRefSlow(shifts), dims); + } +} + +// aten::roll(Tensor self, SymInt[1] shifts, int[1] dims=[]) -> Tensor +inline at::Tensor roll_symint(const at::Tensor & self, c10::SymIntArrayRef shifts, at::IntArrayRef dims={}) { + return at::_ops::roll::call(self, shifts, dims); +} +namespace symint { + template ::value>> + at::Tensor roll(const at::Tensor & self, c10::SymIntArrayRef shifts, at::IntArrayRef dims={}) { + return at::_ops::roll::call(self, shifts, dims); + } +} + +// aten::roll.out(Tensor self, SymInt[1] shifts, int[1] dims=[], *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & roll_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef shifts, at::IntArrayRef dims={}) { + return at::_ops::roll_out::call(self, c10::fromIntArrayRefSlow(shifts), dims, out); +} +namespace symint { + template ::value>> + at::Tensor & roll_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef shifts, at::IntArrayRef dims={}) { + return at::_ops::roll_out::call(self, c10::fromIntArrayRefSlow(shifts), dims, out); + } +} + +// aten::roll.out(Tensor self, SymInt[1] shifts, int[1] dims=[], *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & roll_outf(const at::Tensor & self, at::IntArrayRef shifts, at::IntArrayRef dims, at::Tensor & out) { + return at::_ops::roll_out::call(self, c10::fromIntArrayRefSlow(shifts), dims, out); +} +namespace symint { + template ::value>> + at::Tensor & roll_outf(const at::Tensor & self, at::IntArrayRef shifts, at::IntArrayRef dims, at::Tensor & out) { + return at::_ops::roll_out::call(self, c10::fromIntArrayRefSlow(shifts), dims, out); + } +} + +// aten::roll.out(Tensor self, SymInt[1] shifts, int[1] dims=[], *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & roll_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef shifts, at::IntArrayRef dims={}) { + return at::_ops::roll_out::call(self, shifts, dims, out); +} +namespace symint { + template ::value>> + at::Tensor & roll_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef shifts, at::IntArrayRef dims={}) { + return at::_ops::roll_out::call(self, shifts, dims, out); + } +} + +// aten::roll.out(Tensor self, SymInt[1] shifts, int[1] dims=[], *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & roll_symint_outf(const at::Tensor & self, c10::SymIntArrayRef shifts, at::IntArrayRef dims, at::Tensor & out) { + return at::_ops::roll_out::call(self, shifts, dims, out); +} +namespace symint { + template ::value>> + at::Tensor & roll_outf(const at::Tensor & self, c10::SymIntArrayRef shifts, at::IntArrayRef dims, at::Tensor & out) { + return at::_ops::roll_out::call(self, shifts, dims, out); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/select_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/select_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..97a02ee0b41b3058d27fefafe0d2365cd09e9228 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/select_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor select(const at::Tensor & self, at::Dimname dim, int64_t index); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/smooth_l1_loss_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/smooth_l1_loss_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..13b3b9b45fff4f24a3a6a62fe137e3f6cbe1272b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/smooth_l1_loss_cuda_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor smooth_l1_loss(const at::Tensor & self, const at::Tensor & target, int64_t reduction=at::Reduction::Mean, double beta=1.0); +TORCH_API at::Tensor & smooth_l1_loss_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & target, int64_t reduction=at::Reduction::Mean, double beta=1.0); +TORCH_API at::Tensor & smooth_l1_loss_outf(const at::Tensor & self, const at::Tensor & target, int64_t reduction, double beta, at::Tensor & out); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_resize_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_resize_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..fbdaa39b990e2d4aecadda81a29656cb32032394 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_resize_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 const at::Tensor & sparse_resize_(const at::Tensor & self, at::IntArrayRef size, int64_t sparse_dim, int64_t dense_dim); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_bessel_j1_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_bessel_j1_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..300df779b2a6ce36824f30ba946fea0b638b1631 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_bessel_j1_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor special_bessel_j1(const at::Tensor & self); +TORCH_API at::Tensor & special_bessel_j1_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & special_bessel_j1_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_hermite_polynomial_h_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_hermite_polynomial_h_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..7c3a44d48df6fac3b67de2171910ced647f1a05c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_hermite_polynomial_h_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor special_hermite_polynomial_h(const at::Tensor & x, const at::Tensor & n); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_modified_bessel_k0_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_modified_bessel_k0_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0207b570537da2f4761aa29146312be579cd3cc6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_modified_bessel_k0_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor special_modified_bessel_k0(const at::Tensor & self); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_modified_bessel_k0_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_modified_bessel_k0_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..59a36708fc82a865b6621aab1f23f72e8659db16 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_modified_bessel_k0_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor special_modified_bessel_k0(const at::Tensor & self); +TORCH_API at::Tensor & special_modified_bessel_k0_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & special_modified_bessel_k0_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_psi.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_psi.h new file mode 100644 index 0000000000000000000000000000000000000000..35ecd7f1dc24882da09d9e2dff9577a7d93d91d3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_psi.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::special_psi(Tensor self) -> Tensor +inline at::Tensor special_psi(const at::Tensor & self) { + return at::_ops::special_psi::call(self); +} + +// aten::special_psi.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & special_psi_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::special_psi_out::call(self, out); +} +// aten::special_psi.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & special_psi_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::special_psi_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_shifted_chebyshev_polynomial_w_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_shifted_chebyshev_polynomial_w_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..2f8d1ce4ece672c1ff818d5394662e259449f08f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_shifted_chebyshev_polynomial_w_compositeexplicitautograd_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 compositeexplicitautograd { + +TORCH_API at::Tensor special_shifted_chebyshev_polynomial_w(const at::Scalar & x, const at::Tensor & n); +TORCH_API at::Tensor & special_shifted_chebyshev_polynomial_w_out(at::Tensor & out, const at::Scalar & x, const at::Tensor & n); +TORCH_API at::Tensor & special_shifted_chebyshev_polynomial_w_outf(const at::Scalar & x, const at::Tensor & n, at::Tensor & out); +TORCH_API at::Tensor special_shifted_chebyshev_polynomial_w(const at::Tensor & x, const at::Scalar & n); +TORCH_API at::Tensor & special_shifted_chebyshev_polynomial_w_out(at::Tensor & out, const at::Tensor & x, const at::Scalar & n); +TORCH_API at::Tensor & special_shifted_chebyshev_polynomial_w_outf(const at::Tensor & x, const at::Scalar & n, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/split_copy_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/split_copy_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..117be69cc3651cd6eb47ad13e020be29a8535655 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/split_copy_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 void split_copy_out(at::TensorList out, const at::Tensor & self, int64_t split_size, int64_t dim=0); +TORCH_API void split_copy_outf(const at::Tensor & self, int64_t split_size, int64_t dim, at::TensorList out); +TORCH_API void split_copy_symint_out(at::TensorList out, const at::Tensor & self, c10::SymInt split_size, int64_t dim=0); +TORCH_API void split_copy_symint_outf(const at::Tensor & self, c10::SymInt split_size, 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/take_along_dim_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/take_along_dim_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..003c227e51d32b75a7d12310d7bbc280aa230c7f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/take_along_dim_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 take_along_dim(const at::Tensor & self, const at::Tensor & indices, c10::optional dim=c10::nullopt); +TORCH_API at::Tensor & take_along_dim_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & indices, c10::optional dim=c10::nullopt); +TORCH_API at::Tensor & take_along_dim_outf(const at::Tensor & self, const at::Tensor & indices, c10::optional dim, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/tanh_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/tanh_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..222cbb467d5c5f2772afed9916c0a9bfd55ec994 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/tanh_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 tanh(const at::Tensor & self); +TORCH_API at::Tensor & tanh_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & tanh_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & tanh_(at::Tensor & self); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_mkldnn.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_mkldnn.h new file mode 100644 index 0000000000000000000000000000000000000000..9430f7b93d85c9fa573f80df83e18d5c6cf2c968 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_mkldnn.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_mkldnn.out(Tensor self, ScalarType? dtype=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & to_mkldnn_out(at::Tensor & out, const at::Tensor & self, c10::optional dtype=c10::nullopt) { + return at::_ops::to_mkldnn_out::call(self, dtype, out); +} +// aten::to_mkldnn.out(Tensor self, ScalarType? dtype=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & to_mkldnn_outf(const at::Tensor & self, c10::optional dtype, at::Tensor & out) { + return at::_ops::to_mkldnn_out::call(self, dtype, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_linear1d_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_linear1d_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..b113090ceaa437211af1df76da575f4278a16cb0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_linear1d_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 upsample_linear1d(const at::Tensor & self, at::IntArrayRef output_size, bool align_corners, c10::optional scales=c10::nullopt); +TORCH_API at::Tensor upsample_linear1d_symint(const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional scales=c10::nullopt); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_trilinear3d_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_trilinear3d_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..3b9bbb4f54dbaafa52f8961c4d36611b9f7a37ae --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_trilinear3d_backward_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API upsample_trilinear3d_backward_grad_input { + using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, bool, c10::optional, c10::optional, c10::optional, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::upsample_trilinear3d_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "upsample_trilinear3d_backward.grad_input(Tensor grad_output, SymInt[3] output_size, SymInt[5] input_size, bool align_corners, float? scales_d=None, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); +}; + +struct TORCH_API upsample_trilinear3d_backward { + using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, bool, 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::upsample_trilinear3d_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "upsample_trilinear3d_backward(Tensor grad_output, SymInt[3] output_size, SymInt[5] input_size, bool align_corners, float? scales_d=None, float? scales_h=None, float? scales_w=None) -> Tensor") + static at::Tensor call(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_d, c10::optional scales_h, c10::optional scales_w); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/view_as_complex_copy_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/view_as_complex_copy_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..f4b30f1e4255c304de258635fb05c93ad58e7711 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/view_as_complex_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 & view_as_complex_copy_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & view_as_complex_copy_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at