diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_add_batch_dim_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_add_batch_dim_native.h new file mode 100644 index 0000000000000000000000000000000000000000..1d7afaac43659404f572f92ba2d5c8fd19e5c4e0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_add_batch_dim_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 _add_batch_dim(const at::Tensor & self, int64_t batch_dim, int64_t level); +} // 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.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_impl_index_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..5e2829eaa3dfe255d555694ab36f29dbc5243296 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_impl_index_backward.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_batch_norm_impl_index_backward(int impl_index, Tensor input, Tensor grad_output, Tensor? weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_var_transform, bool train, float eps, bool[3] output_mask, Tensor reservedSpace) -> (Tensor, Tensor, Tensor) +inline ::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) { + return at::_ops::_batch_norm_impl_index_backward::call(impl_index, input, grad_output, weight, running_mean, running_var, save_mean, save_var_transform, train, eps, output_mask, reservedSpace); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cdist_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cdist_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..1caa0989425a2a374613794d0044f6edbbb60c29 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cdist_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 _cdist_backward { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Tensor &, double, 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::_cdist_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_cdist_backward(Tensor grad, Tensor x1, Tensor x2, float p, Tensor cdist) -> Tensor") + static at::Tensor call(const at::Tensor & grad, const at::Tensor & x1, const at::Tensor & x2, double p, const at::Tensor & cdist); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad, const at::Tensor & x1, const at::Tensor & x2, double p, const at::Tensor & cdist); +}; + +struct TORCH_API _cdist_backward_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const at::Tensor &, double, 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::_cdist_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_cdist_backward.out(Tensor grad, Tensor x1, Tensor x2, float p, Tensor cdist, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & grad, const at::Tensor & x1, const at::Tensor & x2, double p, const at::Tensor & cdist, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad, const at::Tensor & x1, const at::Tensor & x2, double p, const at::Tensor & cdist, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convolution_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convolution_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..68da8d0bf7334f9883c269b9322c67306d7ce4c8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convolution_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 _convolution(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool transposed, at::IntArrayRef output_padding, int64_t groups, bool benchmark, bool deterministic, bool cudnn_enabled, bool allow_tf32); +TORCH_API at::Tensor _convolution_symint(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, bool transposed, c10::SymIntArrayRef output_padding, c10::SymInt groups, bool benchmark, bool deterministic, bool cudnn_enabled, bool allow_tf32); +TORCH_API at::Tensor & _convolution_out(at::Tensor & out, const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool transposed, at::IntArrayRef output_padding, int64_t groups, bool benchmark, bool deterministic, bool cudnn_enabled, bool allow_tf32); +TORCH_API at::Tensor & _convolution_outf(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool transposed, at::IntArrayRef output_padding, int64_t groups, bool benchmark, bool deterministic, bool cudnn_enabled, bool allow_tf32, at::Tensor & out); +TORCH_API at::Tensor & _convolution_symint_out(at::Tensor & out, const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, bool transposed, c10::SymIntArrayRef output_padding, c10::SymInt groups, bool benchmark, bool deterministic, bool cudnn_enabled, bool allow_tf32); +TORCH_API at::Tensor & _convolution_symint_outf(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, bool transposed, c10::SymIntArrayRef output_padding, c10::SymInt groups, bool benchmark, bool deterministic, bool cudnn_enabled, bool allow_tf32, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_addcdiv_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_addcdiv_native.h new file mode 100644 index 0000000000000000000000000000000000000000..356e3b948bfad49101468e220d9318a3cdddded0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_addcdiv_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_addcdiv_Scalar_out(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Scalar & value, at::TensorList out); +TORCH_API ::std::vector foreach_tensor_addcdiv_scalar_slow(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Scalar & value=1); +TORCH_API void foreach_tensor_addcdiv_scalar_slow_(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Scalar & value=1); +TORCH_API ::std::vector foreach_tensor_addcdiv_scalar_cuda(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Scalar & value=1); +TORCH_API void foreach_tensor_addcdiv_scalar_cuda_(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Scalar & value=1); +TORCH_API void _foreach_addcdiv_ScalarList_out(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, at::ArrayRef scalars, at::TensorList out); +TORCH_API ::std::vector foreach_tensor_addcdiv_scalarlist_slow(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, at::ArrayRef scalars); +TORCH_API void foreach_tensor_addcdiv_scalarlist_slow_(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, at::ArrayRef scalars); +TORCH_API ::std::vector foreach_tensor_addcdiv_scalarlist_cuda(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, at::ArrayRef scalars); +TORCH_API void foreach_tensor_addcdiv_scalarlist_cuda_(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, at::ArrayRef scalars); +TORCH_API void _foreach_addcdiv_Tensor_out(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Tensor & scalars, at::TensorList out); +TORCH_API ::std::vector foreach_tensor_addcdiv_tensor_slow(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Tensor & scalars); +TORCH_API void foreach_tensor_addcdiv_tensor_slow_(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Tensor & scalars); +TORCH_API ::std::vector foreach_tensor_addcdiv_tensor_cuda(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Tensor & scalars); +TORCH_API void foreach_tensor_addcdiv_tensor_cuda_(at::TensorList self, at::TensorList tensor1, at::TensorList tensor2, const at::Tensor & scalars); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sub_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sub_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..e1262cbef5739b0eba94e0836fa86dcac7474c74 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sub_cpu_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API ::std::vector _foreach_sub(at::TensorList self, const at::Scalar & scalar); +TORCH_API void _foreach_sub_(at::TensorList self, const at::Scalar & scalar); +TORCH_API ::std::vector _foreach_sub(at::TensorList self, at::TensorList other, const at::Scalar & alpha=1); +TORCH_API void _foreach_sub_(at::TensorList self, at::TensorList other, const at::Scalar & alpha=1); +TORCH_API ::std::vector _foreach_sub(at::TensorList self, at::ArrayRef scalars); +TORCH_API void _foreach_sub_(at::TensorList self, at::ArrayRef scalars); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_eigh_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_eigh_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c3a9e5693f2938ee925eb0c1aa786985a9bce3fd --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_eigh_cuda_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple _linalg_eigh(const at::Tensor & A, c10::string_view UPLO="L", bool compute_v=true); +TORCH_API ::std::tuple _linalg_eigh_out(at::Tensor & eigenvalues, at::Tensor & eigenvectors, const at::Tensor & A, c10::string_view UPLO="L", bool compute_v=true); +TORCH_API ::std::tuple _linalg_eigh_outf(const at::Tensor & A, c10::string_view UPLO, bool compute_v, at::Tensor & eigenvalues, at::Tensor & eigenvectors); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_logcumsumexp_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_logcumsumexp_native.h new file mode 100644 index 0000000000000000000000000000000000000000..04dfb9838e33df316adba273ad31a5895b0f4d36 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_logcumsumexp_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 _logcumsumexp_cpu(const at::Tensor & self, int64_t dim); +TORCH_API at::Tensor & _logcumsumexp_out_cpu(const at::Tensor & self, int64_t dim, at::Tensor & out); +TORCH_API at::Tensor _logcumsumexp_cuda(const at::Tensor & self, int64_t dim); +TORCH_API at::Tensor & _logcumsumexp_out_cuda(const at::Tensor & self, int64_t dim, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_mkldnn_transpose_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_mkldnn_transpose_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..b0c8ac78e6442e19ae12e655bf04c900a5ad43d5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_mkldnn_transpose_meta_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor & _mkldnn_transpose_(at::Tensor & self, int64_t dim0, int64_t dim1); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_prelu_kernel_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_prelu_kernel_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..42fc04ecf6748ed42cf69d8dd4af39f6e1d8d8f6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_prelu_kernel_cpu_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor _prelu_kernel(const at::Tensor & self, const at::Tensor & weight); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_dense_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_dense_native.h new file mode 100644 index 0000000000000000000000000000000000000000..62f812925f15c4ff5556f1bb3f7eb648df15a61e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_dense_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 & _to_dense_out(const at::Tensor & self, c10::optional dtype, c10::optional masked_grad, at::Tensor & out); +TORCH_API at::Tensor sparse_to_dense(const at::Tensor & self, c10::optional dtype=c10::nullopt, c10::optional masked_grad=c10::nullopt); +TORCH_API at::Tensor sparse_compressed_to_dense(const at::Tensor & self, c10::optional dtype=c10::nullopt, c10::optional masked_grad=c10::nullopt); +TORCH_API at::Tensor mkldnn_to_dense(const at::Tensor & self, c10::optional dtype=c10::nullopt, c10::optional masked_grad=c10::nullopt); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_transformer_encoder_layer_fwd_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_transformer_encoder_layer_fwd_native.h new file mode 100644 index 0000000000000000000000000000000000000000..907ea2e376f3f9bb00952b6fce07553a7e250c46 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_transformer_encoder_layer_fwd_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & _transformer_encoder_layer_fwd_out(const at::Tensor & src, int64_t embed_dim, int64_t num_heads, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, bool use_gelu, bool norm_first, double eps, const at::Tensor & norm_weight_1, const at::Tensor & norm_bias_1, const at::Tensor & norm_weight_2, const at::Tensor & norm_bias_2, const at::Tensor & ffn_weight_1, const at::Tensor & ffn_bias_1, const at::Tensor & ffn_weight_2, const at::Tensor & ffn_bias_2, const c10::optional & mask, c10::optional mask_type, at::Tensor & out); +TORCH_API at::Tensor transformer_encoder_layer_forward(const at::Tensor & src, int64_t embed_dim, int64_t num_heads, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, bool use_gelu, bool norm_first, double eps, const at::Tensor & norm_weight_1, const at::Tensor & norm_bias_1, const at::Tensor & norm_weight_2, const at::Tensor & norm_bias_2, const at::Tensor & ffn_weight_1, const at::Tensor & ffn_bias_1, const at::Tensor & ffn_weight_2, const at::Tensor & ffn_bias_2, const c10::optional & mask={}, c10::optional mask_type=c10::nullopt); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_triton_multi_head_attention.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_triton_multi_head_attention.h new file mode 100644 index 0000000000000000000000000000000000000000..94c474d4d3dcd03422a3efdb12f1864669a5d788 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_triton_multi_head_attention.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::_triton_multi_head_attention(Tensor query, Tensor key, Tensor value, int embed_dim, int num_head, Tensor qkv_weight, Tensor qkv_bias, Tensor proj_weight, Tensor proj_bias, Tensor? mask=None) -> Tensor +inline at::Tensor _triton_multi_head_attention(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, int64_t embed_dim, int64_t num_head, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, const c10::optional & mask={}) { + return at::_ops::_triton_multi_head_attention::call(query, key, value, embed_dim, num_head, qkv_weight, qkv_bias, proj_weight, proj_bias, mask); +} + +// aten::_triton_multi_head_attention.out(Tensor query, Tensor key, Tensor value, int embed_dim, int num_head, Tensor qkv_weight, Tensor qkv_bias, Tensor proj_weight, Tensor proj_bias, Tensor? mask=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _triton_multi_head_attention_out(at::Tensor & out, const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, int64_t embed_dim, int64_t num_head, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, const c10::optional & mask={}) { + return at::_ops::_triton_multi_head_attention_out::call(query, key, value, embed_dim, num_head, qkv_weight, qkv_bias, proj_weight, proj_bias, mask, out); +} +// aten::_triton_multi_head_attention.out(Tensor query, Tensor key, Tensor value, int embed_dim, int num_head, Tensor qkv_weight, Tensor qkv_bias, Tensor proj_weight, Tensor proj_bias, Tensor? mask=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _triton_multi_head_attention_outf(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, int64_t embed_dim, int64_t num_head, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, const c10::optional & mask, at::Tensor & out) { + return at::_ops::_triton_multi_head_attention_out::call(query, key, value, embed_dim, num_head, qkv_weight, qkv_bias, proj_weight, proj_bias, mask, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bilinear2d_aa_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bilinear2d_aa_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..dd7dd1c14cb6530f3e156ce10d8be1ab9a980fc3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_bilinear2d_aa_backward_cuda_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor _upsample_bilinear2d_aa_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor _upsample_bilinear2d_aa_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & _upsample_bilinear2d_aa_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & _upsample_bilinear2d_aa_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, bool align_corners, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); +TORCH_API at::Tensor & _upsample_bilinear2d_aa_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor & _upsample_bilinear2d_aa_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, bool align_corners, c10::optional scales_h, c10::optional scales_w, at::Tensor & grad_input); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact1d_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact1d_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..dfadbef4b311f1b45466fb67f9d8fd93743228f0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact1d_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured__upsample_nearest_exact1d : public at::impl::MetaBase { + + + void meta(const at::Tensor & self, at::ArrayRef output_size, c10::optional scales); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm_native.h new file mode 100644 index 0000000000000000000000000000000000000000..db3a3e51be2c6dffac8566efa76e27e9fda58273 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_weight_norm_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor _weight_norm(const at::Tensor & v, const at::Tensor & g, int64_t dim=0); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adjoint.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adjoint.h new file mode 100644 index 0000000000000000000000000000000000000000..315f8046deccb7c2fff8174e88418a1c209a235f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adjoint.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::adjoint(Tensor(a) self) -> Tensor(a) +inline at::Tensor adjoint(const at::Tensor & self) { + return at::_ops::adjoint::call(self); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/as_strided_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/as_strided_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..d0386a7f2f2cb51431988a2268f2962a827b27a1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/as_strided_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 as_strided(const at::Tensor & self, at::IntArrayRef size, at::IntArrayRef stride, c10::optional storage_offset=c10::nullopt); +TORCH_API at::Tensor as_strided_symint(const at::Tensor & self, c10::SymIntArrayRef size, c10::SymIntArrayRef stride, c10::optional storage_offset=c10::nullopt); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool1d_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool1d_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a8a0753f44fac5815a1c33b34a2eb243737ebede --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool1d_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 avg_pool1d(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, bool ceil_mode=false, bool count_include_pad=true); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_or_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_or_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..ecfc21114d55ae0c4fca7be8cc2d861dbdc3f0b9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/bitwise_or_ops.h @@ -0,0 +1,105 @@ +#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 bitwise_or_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::bitwise_or") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bitwise_or.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 bitwise_or_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::bitwise_or") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bitwise_or.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 bitwise_or_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::bitwise_or") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bitwise_or.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 bitwise_or_Scalar_Tensor { + using schema = at::Tensor (const at::Scalar &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::bitwise_or") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bitwise_or.Scalar_Tensor(Scalar self, Tensor other) -> Tensor") + static at::Tensor call(const at::Scalar & self, const at::Tensor & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & self, const at::Tensor & other); +}; + +struct TORCH_API bitwise_or_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::bitwise_or") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bitwise_or.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 bitwise_or__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::bitwise_or_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bitwise_or_.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 bitwise_or__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::bitwise_or_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bitwise_or_.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); +}; + +struct TORCH_API bitwise_or_Scalar_Tensor_out { + using schema = at::Tensor & (const at::Scalar &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::bitwise_or") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_Tensor_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bitwise_or.Scalar_Tensor_out(Scalar self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Scalar & self, const at::Tensor & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & 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/ccol_indices_copy_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ccol_indices_copy_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..3748636504f01196de1d234388f72fe917fd2d2c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ccol_indices_copy_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor ccol_indices_copy(const at::Tensor & self); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/crow_indices_copy_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/crow_indices_copy_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..2e39202781c78a6bc21bf33a0e7b455c3e747d27 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/crow_indices_copy_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API crow_indices_copy { + using schema = at::Tensor (const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::crow_indices_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "crow_indices_copy(Tensor self) -> Tensor") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +struct TORCH_API crow_indices_copy_out { + using schema = at::Tensor & (const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::crow_indices_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "crow_indices_copy.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumprod_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumprod_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..74e54910b0e1e4efe117c4b8ca6544e4712991ce --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumprod_backward_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API cumprod_backward { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, int64_t, 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::cumprod_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "cumprod_backward(Tensor grad, Tensor input, int dim, Tensor output) -> Tensor") + static at::Tensor call(const at::Tensor & grad, const at::Tensor & input, int64_t dim, const at::Tensor & output); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad, const at::Tensor & input, int64_t dim, const at::Tensor & output); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..cdac1e43e8f4d5e5942fccd0bb261b6815fc1019 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_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 cumsum(const at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); +TORCH_API at::Tensor & cumsum_(at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach.h new file mode 100644 index 0000000000000000000000000000000000000000..2efaf2af2431fe3a9331d07fceb9b5113f9e08c5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach.h @@ -0,0 +1,35 @@ +#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::detach(Tensor(a) self) -> Tensor(a) +inline at::Tensor detach(const at::Tensor & self) { + return at::_ops::detach::call(self); +} + +// aten::detach_(Tensor(a!) self) -> Tensor(a!) +inline at::Tensor & detach_(at::Tensor & self) { + return at::_ops::detach_::call(self); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach_copy_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach_copy_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a813a2cc31bd11100eba233e339362be78d31dd8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach_copy_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & detach_copy_out(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor detach_copy(const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..1474453f1644deb4f2c42ebe05c3db4c7693e814 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_cpu_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor digamma(const at::Tensor & self); +TORCH_API at::Tensor & digamma_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & digamma_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & digamma_(at::Tensor & self); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/dstack_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/dstack_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..9c2c56de3d9ff3dda532f449bf2e6c7d1a9953f1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/dstack_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 dstack { + 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::dstack") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "dstack(Tensor[] tensors) -> Tensor") + static at::Tensor call(at::TensorList tensors); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList tensors); +}; + +struct TORCH_API dstack_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::dstack") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "dstack.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/empty_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/empty_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..21f68b046d716e2ee5e33aad1d99c77d0d4135be --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/empty_ops.h @@ -0,0 +1,61 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API empty_names { + using schema = at::Tensor (at::IntArrayRef, c10::optional, c10::optional, c10::optional, c10::optional, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::empty") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "names") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "empty.names(int[] size, *, Dimname[]? names, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor") + static at::Tensor call(at::IntArrayRef size, c10::optional names, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, at::IntArrayRef size, c10::optional names, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format); +}; + +struct TORCH_API empty_memory_format { + using schema = at::Tensor (c10::SymIntArrayRef, c10::optional, c10::optional, c10::optional, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::empty") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "memory_format") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "empty.memory_format(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None, MemoryFormat? memory_format=None) -> Tensor") + static at::Tensor call(c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional memory_format); +}; + +struct TORCH_API empty_out { + using schema = at::Tensor & (c10::SymIntArrayRef, 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::empty") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "empty.out(SymInt[] size, *, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(c10::SymIntArrayRef size, c10::optional memory_format, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymIntArrayRef size, c10::optional memory_format, at::Tensor & out); +}; + +struct TORCH_API empty_names_out { + using schema = at::Tensor & (at::IntArrayRef, 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::empty") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "names_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "empty.names_out(int[] size, *, Dimname[]? names, MemoryFormat? memory_format=None, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(at::IntArrayRef size, c10::optional names, c10::optional memory_format, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::IntArrayRef size, c10::optional names, c10::optional memory_format, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_ifft2_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_ifft2_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..1b98ee4b7d3f0b62f86698103a57efcfa27a2de4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_ifft2_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API fft_ifft2 { + using schema = at::Tensor (const at::Tensor &, at::OptionalSymIntArrayRef, at::IntArrayRef, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::fft_ifft2") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fft_ifft2(Tensor self, SymInt[1]? s=None, int[1] dim=[-2,-1], str? norm=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, at::OptionalSymIntArrayRef s, at::IntArrayRef dim, c10::optional norm); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::OptionalSymIntArrayRef s, at::IntArrayRef dim, c10::optional norm); +}; + +struct TORCH_API fft_ifft2_out { + using schema = at::Tensor & (const at::Tensor &, at::OptionalSymIntArrayRef, at::IntArrayRef, 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::fft_ifft2") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fft_ifft2.out(Tensor self, SymInt[1]? s=None, int[1] dim=[-2,-1], str? norm=None, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, at::OptionalSymIntArrayRef s, at::IntArrayRef dim, c10::optional norm, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::OptionalSymIntArrayRef s, at::IntArrayRef dim, c10::optional norm, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/float_power_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/float_power_native.h new file mode 100644 index 0000000000000000000000000000000000000000..15673ab38ecf26675a7141a311edeafdf0e7b3c9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/float_power_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 + + +namespace at { +namespace native { +TORCH_API at::Tensor float_power(const at::Tensor & self, const at::Tensor & exponent); +TORCH_API at::Tensor & float_power_out(const at::Tensor & self, const at::Tensor & exponent, at::Tensor & out); +TORCH_API at::Tensor & float_power_(at::Tensor & self, const at::Tensor & exponent); +TORCH_API at::Tensor float_power(const at::Scalar & self, const at::Tensor & exponent); +TORCH_API at::Tensor & float_power_out(const at::Scalar & self, const at::Tensor & exponent, at::Tensor & out); +TORCH_API at::Tensor float_power(const at::Tensor & self, const at::Scalar & exponent); +TORCH_API at::Tensor & float_power_out(const at::Tensor & self, const at::Scalar & exponent, at::Tensor & out); +TORCH_API at::Tensor & float_power_(at::Tensor & self, const at::Scalar & exponent); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_3d_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_3d_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..1d241244cf13dd3cace15c009af908c9f32a1ca4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_3d_backward_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API ::std::tuple grid_sampler_3d_backward_out(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1); +TORCH_API ::std::tuple grid_sampler_3d_backward_cpu(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, ::std::array output_mask); +TORCH_API ::std::tuple grid_sampler_3d_backward_cuda(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, ::std::array output_mask); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardsigmoid_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardsigmoid_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..642140c2e920127aa6d870cf48a07eef153b1213 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardsigmoid_cuda_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor hardsigmoid(const at::Tensor & self); +TORCH_API at::Tensor & hardsigmoid_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & hardsigmoid_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & hardsigmoid_(at::Tensor & self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardswish_backward_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardswish_backward_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..5630a0560540fa2db05115e868a4eb560172d067 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardswish_backward_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & hardswish_backward_out(at::Tensor & out, const at::Tensor & grad_output, const at::Tensor & self); +TORCH_API at::Tensor & hardswish_backward_outf(const at::Tensor & grad_output, 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/hardswish_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardswish_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0f29b0293645726c227b2264e4eaf866a1e179a4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardswish_cuda_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor hardswish(const at::Tensor & self); +TORCH_API at::Tensor & hardswish_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & hardswish_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & hardswish_(at::Tensor & self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/heaviside_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/heaviside_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c28380986264e55a7c971fa47cc11537ce793250 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/heaviside_cuda_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor heaviside(const at::Tensor & self, const at::Tensor & values); +TORCH_API at::Tensor & heaviside_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & values); +TORCH_API at::Tensor & heaviside_outf(const at::Tensor & self, const at::Tensor & values, at::Tensor & out); +TORCH_API at::Tensor & heaviside_(at::Tensor & self, const at::Tensor & values); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hsplit.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hsplit.h new file mode 100644 index 0000000000000000000000000000000000000000..bc9a990a41b1ba715333ec3221aa097c1f8e77c8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hsplit.h @@ -0,0 +1,35 @@ +#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::hsplit.int(Tensor(a -> *) self, int sections) -> Tensor(a)[] +inline ::std::vector hsplit(const at::Tensor & self, int64_t sections) { + return at::_ops::hsplit_int::call(self, sections); +} + +// aten::hsplit.array(Tensor(a -> *) self, int[] indices) -> Tensor(a)[] +inline ::std::vector hsplit(const at::Tensor & self, at::IntArrayRef indices) { + return at::_ops::hsplit_array::call(self, indices); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/huber_loss_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/huber_loss_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..534fd4019d4547dca8c8d7d498233daeac7124a2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/huber_loss_backward_cuda_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor & huber_loss_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & target, int64_t reduction, double delta); +TORCH_API at::Tensor & huber_loss_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & target, int64_t reduction, double delta, at::Tensor & grad_input); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..d9b4fdbae96838e24427e7b16eab9ecaec642d58 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_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 index(const at::Tensor & self, const c10::List> & indices); +TORCH_API at::Tensor & index_out(at::Tensor & out, const at::Tensor & self, const c10::List> & indices); +TORCH_API at::Tensor & index_outf(const at::Tensor & self, const c10::List> & indices, at::Tensor & out); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/istft_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/istft_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..443ad161b891233a67e478a68fd21863d0c790f7 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/istft_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 istft { + using schema = at::Tensor (const at::Tensor &, int64_t, c10::optional, c10::optional, const c10::optional &, bool, bool, c10::optional, c10::optional, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::istft") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "istft(Tensor self, int n_fft, int? hop_length=None, int? win_length=None, Tensor? window=None, bool center=True, bool normalized=False, bool? onesided=None, int? length=None, bool return_complex=False) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t n_fft, c10::optional hop_length, c10::optional win_length, const c10::optional & window, bool center, bool normalized, c10::optional onesided, c10::optional length, bool return_complex); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t n_fft, c10::optional hop_length, c10::optional win_length, const c10::optional & window, bool center, bool normalized, c10::optional onesided, c10::optional length, bool return_complex); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_cond_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_cond_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..65e3dfc3370c217b44e6dd9448b853143804901a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_cond_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 linalg_cond(const at::Tensor & self, const c10::optional & p=c10::nullopt); +TORCH_API at::Tensor & linalg_cond_out(at::Tensor & out, const at::Tensor & self, const c10::optional & p=c10::nullopt); +TORCH_API at::Tensor & linalg_cond_outf(const at::Tensor & self, const c10::optional & p, at::Tensor & out); +TORCH_API at::Tensor linalg_cond(const at::Tensor & self, c10::string_view p); +TORCH_API at::Tensor & linalg_cond_out(at::Tensor & out, const at::Tensor & self, c10::string_view p); +TORCH_API at::Tensor & linalg_cond_outf(const at::Tensor & self, c10::string_view p, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..deecd795b72a7db02f83fcf879633760898d3b99 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_lu_solve_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 linalg_lu_solve(const at::Tensor & LU, const at::Tensor & pivots, const at::Tensor & B, bool left=true, bool adjoint=false); +TORCH_API at::Tensor & linalg_lu_solve_out(at::Tensor & out, const at::Tensor & LU, const at::Tensor & pivots, const at::Tensor & B, bool left=true, bool adjoint=false); +TORCH_API at::Tensor & linalg_lu_solve_outf(const at::Tensor & LU, const at::Tensor & pivots, const at::Tensor & B, bool left, bool adjoint, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logspace_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logspace_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..cf0a6fe31c3d4aaaf8162cb08a9f5b2a719ea6c2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logspace_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 & logspace_out(at::Tensor & out, const at::Scalar & start, const at::Scalar & end, int64_t steps, double base=10.0); +TORCH_API at::Tensor & logspace_outf(const at::Scalar & start, const at::Scalar & end, int64_t steps, double base, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logsumexp_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logsumexp_native.h new file mode 100644 index 0000000000000000000000000000000000000000..ca86369240283a35a6ae33001ae1ac467badb472 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logsumexp_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 logsumexp(const at::Tensor & self, at::IntArrayRef dim, bool keepdim=false); +TORCH_API at::Tensor & logsumexp_out(const at::Tensor & self, at::IntArrayRef dim, bool keepdim, at::Tensor & out); +TORCH_API at::Tensor logsumexp(const at::Tensor & self, at::DimnameList dim, bool keepdim=false); +TORCH_API at::Tensor & logsumexp_out(const at::Tensor & self, at::DimnameList dim, bool keepdim, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mT_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mT_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..b6526e6604e20298e1c90b5391c7b5066ea72a7c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mT_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 mT { + 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::mT") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mT(Tensor(a) self) -> Tensor(a)") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matrix_H_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matrix_H_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..50dfbfbfba565a6a4bc500ccd95c09b172dbccc5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matrix_H_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 matrix_H(const at::Tensor & self); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_max_pool3d_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_max_pool3d_native.h new file mode 100644 index 0000000000000000000000000000000000000000..ec662c3cc0025ecec01ea8c8d67456b968d807c4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_max_pool3d_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & mkldnn_max_pool3d_out(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, at::Tensor & out); +TORCH_API at::Tensor mkldnn_max_pool3d(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, at::IntArrayRef dilation=1, bool ceil_mode=false); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mps_convolution_transpose_backward_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mps_convolution_transpose_backward_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..eb0ea998728cff73afd785f8cdcdfcff5d835d53 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mps_convolution_transpose_backward_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API ::std::tuple mps_convolution_transpose_backward_out(at::Tensor & out0, at::Tensor & out1, const at::Tensor & self, const at::Tensor & grad_output, const at::Tensor & weight, at::IntArrayRef padding, at::IntArrayRef output_padding, at::IntArrayRef stride, at::IntArrayRef dilation, int64_t groups, ::std::array output_mask); +TORCH_API ::std::tuple mps_convolution_transpose_backward_outf(const at::Tensor & self, const at::Tensor & grad_output, const at::Tensor & weight, at::IntArrayRef padding, at::IntArrayRef output_padding, at::IntArrayRef stride, at::IntArrayRef dilation, int64_t groups, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1); +TORCH_API ::std::tuple mps_convolution_transpose_backward_symint_out(at::Tensor & out0, at::Tensor & out1, const at::Tensor & self, const at::Tensor & grad_output, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef output_padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, ::std::array output_mask); +TORCH_API ::std::tuple mps_convolution_transpose_backward_symint_outf(const at::Tensor & self, const at::Tensor & grad_output, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef output_padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_layer_norm_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_layer_norm_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..d1803da7d837d43f073f4a9ca33a73b71c386ca5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_layer_norm_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 native_layer_norm_backward { + using schema = ::std::tuple (const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, const at::Tensor &, const at::Tensor &, const c10::optional &, const c10::optional &, ::std::array); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::native_layer_norm_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "native_layer_norm_backward(Tensor grad_out, Tensor input, SymInt[] normalized_shape, Tensor mean, Tensor rstd, Tensor? weight, Tensor? bias, bool[3] output_mask) -> (Tensor, Tensor, Tensor)") + static ::std::tuple call(const at::Tensor & grad_out, const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const at::Tensor & mean, const at::Tensor & rstd, const c10::optional & weight, const c10::optional & bias, ::std::array output_mask); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_out, const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const at::Tensor & mean, const at::Tensor & rstd, const c10::optional & weight, const c10::optional & bias, ::std::array output_mask); +}; + +struct TORCH_API native_layer_norm_backward_out { + using schema = ::std::tuple (const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, const at::Tensor &, const at::Tensor &, const c10::optional &, const c10::optional &, ::std::array, at::Tensor &, at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::native_layer_norm_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "native_layer_norm_backward.out(Tensor grad_out, Tensor input, SymInt[] normalized_shape, Tensor mean, Tensor rstd, Tensor? weight, Tensor? bias, bool[3] output_mask, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!))") + static ::std::tuple call(const at::Tensor & grad_out, const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const at::Tensor & mean, const at::Tensor & rstd, const c10::optional & weight, const c10::optional & bias, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_out, const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const at::Tensor & mean, const at::Tensor & rstd, const c10::optional & weight, const c10::optional & bias, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_layer_norm_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_layer_norm_native.h new file mode 100644 index 0000000000000000000000000000000000000000..e1a22aaf29cdd20006e5ce7c1d5353115207620a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/native_layer_norm_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 math_native_layer_norm(const at::Tensor & input, at::IntArrayRef normalized_shape, const c10::optional & weight, const c10::optional & bias, double eps); +TORCH_API ::std::tuple native_layer_norm_out_symint(const at::Tensor & input, c10::SymIntArrayRef normalized_shape, const c10::optional & weight, const c10::optional & bias, double eps, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); +TORCH_API ::std::tuple layer_norm_cpu(const at::Tensor & input, at::IntArrayRef normalized_shape, const c10::optional & weight, const c10::optional & bias, double eps); +TORCH_API ::std::tuple layer_norm_cuda(const at::Tensor & input, at::IntArrayRef normalized_shape, const c10::optional & weight, const c10::optional & bias, double eps); +TORCH_API ::std::tuple nested_layer_norm(const at::Tensor & input, at::IntArrayRef normalized_shape, const c10::optional & weight, const c10::optional & bias, double eps); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nextafter_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nextafter_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..02ecb3cb140971e7df2fe36a438fdea25eaa600f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nextafter_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 nextafter_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::nextafter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "nextafter.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 nextafter { + 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::nextafter") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "nextafter(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 nextafter_ { + 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::nextafter_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "nextafter_(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/poisson_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/poisson_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a50988f59dd2ad87153f3fc790b75643bbceb80d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/poisson_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 & poisson_out(const at::Tensor & self, c10::optional generator, at::Tensor & out); +TORCH_API at::Tensor _s_poisson_cpu(const at::Tensor & self, c10::optional generator=c10::nullopt); +TORCH_API at::Tensor _s_poisson_cuda(const at::Tensor & self, 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/positive_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/positive_native.h new file mode 100644 index 0000000000000000000000000000000000000000..165a50b2a7f2319cb81538c172d6540aaec77bdf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/positive_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 positive(const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pow_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pow_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..41d76bb8ada02710eb71c57ff21d47f878cd3cc9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/pow_cpu_dispatch.h @@ -0,0 +1,33 @@ +#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 pow(const at::Tensor & self, const at::Tensor & exponent); +TORCH_API at::Tensor & pow_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & exponent); +TORCH_API at::Tensor & pow_outf(const at::Tensor & self, const at::Tensor & exponent, at::Tensor & out); +TORCH_API at::Tensor & pow_(at::Tensor & self, const at::Tensor & exponent); +TORCH_API at::Tensor pow(const at::Scalar & self, const at::Tensor & exponent); +TORCH_API at::Tensor & pow_out(at::Tensor & out, const at::Scalar & self, const at::Tensor & exponent); +TORCH_API at::Tensor & pow_outf(const at::Scalar & self, const at::Tensor & exponent, at::Tensor & out); +TORCH_API at::Tensor pow(const at::Tensor & self, const at::Scalar & exponent); +TORCH_API at::Tensor & pow_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & exponent); +TORCH_API at::Tensor & pow_outf(const at::Tensor & self, const at::Scalar & exponent, at::Tensor & out); +TORCH_API at::Tensor & pow_(at::Tensor & self, const at::Scalar & exponent); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randint_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randint_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..6d2804dc14a5077ab907c230cbd0990622cba268 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/randint_ops.h @@ -0,0 +1,105 @@ +#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 randint { + using schema = at::Tensor (c10::SymInt, c10::SymIntArrayRef, c10::optional, c10::optional, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::randint") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "randint(SymInt high, SymInt[] size, *, ScalarType? dtype=long, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor") + static at::Tensor call(c10::SymInt high, c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymInt high, c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +}; + +struct TORCH_API randint_generator { + using schema = at::Tensor (c10::SymInt, c10::SymIntArrayRef, c10::optional, c10::optional, c10::optional, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::randint") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "generator") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "randint.generator(SymInt high, SymInt[] size, *, Generator? generator, ScalarType? dtype=long, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor") + static at::Tensor call(c10::SymInt high, c10::SymIntArrayRef size, c10::optional generator, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymInt high, c10::SymIntArrayRef size, c10::optional generator, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +}; + +struct TORCH_API randint_low { + using schema = at::Tensor (c10::SymInt, c10::SymInt, c10::SymIntArrayRef, c10::optional, c10::optional, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::randint") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "low") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "randint.low(SymInt low, SymInt high, SymInt[] size, *, ScalarType? dtype=long, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor") + static at::Tensor call(c10::SymInt low, c10::SymInt high, c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymInt low, c10::SymInt high, c10::SymIntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +}; + +struct TORCH_API randint_low_generator { + using schema = at::Tensor (c10::SymInt, c10::SymInt, c10::SymIntArrayRef, c10::optional, c10::optional, c10::optional, c10::optional, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::randint") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "low_generator") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "randint.low_generator(SymInt low, SymInt high, SymInt[] size, *, Generator? generator, ScalarType? dtype=long, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor") + static at::Tensor call(c10::SymInt low, c10::SymInt high, c10::SymIntArrayRef size, c10::optional generator, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymInt low, c10::SymInt high, c10::SymIntArrayRef size, c10::optional generator, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory); +}; + +struct TORCH_API randint_out { + using schema = at::Tensor & (c10::SymInt, c10::SymIntArrayRef, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::randint") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "randint.out(SymInt high, SymInt[] size, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(c10::SymInt high, c10::SymIntArrayRef size, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymInt high, c10::SymIntArrayRef size, at::Tensor & out); +}; + +struct TORCH_API randint_generator_out { + using schema = at::Tensor & (c10::SymInt, c10::SymIntArrayRef, 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::randint") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "generator_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "randint.generator_out(SymInt high, SymInt[] size, *, Generator? generator, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(c10::SymInt high, c10::SymIntArrayRef size, c10::optional generator, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymInt high, c10::SymIntArrayRef size, c10::optional generator, at::Tensor & out); +}; + +struct TORCH_API randint_low_out { + using schema = at::Tensor & (c10::SymInt, c10::SymInt, c10::SymIntArrayRef, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::randint") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "low_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "randint.low_out(SymInt low, SymInt high, SymInt[] size, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(c10::SymInt low, c10::SymInt high, c10::SymIntArrayRef size, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymInt low, c10::SymInt high, c10::SymIntArrayRef size, at::Tensor & out); +}; + +struct TORCH_API randint_low_generator_out { + using schema = at::Tensor & (c10::SymInt, c10::SymInt, c10::SymIntArrayRef, 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::randint") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "low_generator_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "randint.low_generator_out(SymInt low, SymInt high, SymInt[] size, *, Generator? generator, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(c10::SymInt low, c10::SymInt high, c10::SymIntArrayRef size, c10::optional generator, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, c10::SymInt low, c10::SymInt high, c10::SymIntArrayRef size, c10::optional generator, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/relu_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/relu_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..131b920493179ecfcb882e72806ae09291c8758a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/relu_meta_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor & relu_(at::Tensor & self); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/replication_pad2d_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/replication_pad2d_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c4a4ee4822e042140f185268150cd019e4d3afcc --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/replication_pad2d_meta_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 meta { + +TORCH_API at::Tensor replication_pad2d(const at::Tensor & self, at::IntArrayRef padding); +TORCH_API at::Tensor replication_pad2d_symint(const at::Tensor & self, c10::SymIntArrayRef padding); +TORCH_API at::Tensor & replication_pad2d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef padding); +TORCH_API at::Tensor & replication_pad2d_outf(const at::Tensor & self, at::IntArrayRef padding, at::Tensor & out); +TORCH_API at::Tensor & replication_pad2d_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef padding); +TORCH_API at::Tensor & replication_pad2d_symint_outf(const at::Tensor & self, c10::SymIntArrayRef padding, at::Tensor & out); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rsub_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rsub_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8f8be274c37f43d23a995b285ce2607c2273c498 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/rsub_compositeexplicitautograd_dispatch.h @@ -0,0 +1,27 @@ +#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 & rsub_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha=1); +TORCH_API at::Tensor & rsub_outf(const at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha, at::Tensor & out); +TORCH_API at::Tensor rsub(const at::Tensor & self, const at::Scalar & other, const at::Scalar & alpha=1); +TORCH_API at::Tensor & rsub_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other, const at::Scalar & alpha=1); +TORCH_API at::Tensor & rsub_outf(const at::Tensor & self, const at::Scalar & other, const at::Scalar & alpha, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/silu_backward_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/silu_backward_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ec03ea1a338d85ee542e71515f41a63d623d293d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/silu_backward_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor silu_backward(const at::Tensor & grad_output, const at::Tensor & self); +TORCH_API at::Tensor & silu_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self); +TORCH_API at::Tensor & silu_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & grad_input); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softshrink_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softshrink_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..10a168c401bd02980aef1ff68db97f0c8b58c424 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/softshrink_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 softshrink(const at::Tensor & self, const at::Scalar & lambd=0.5); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_coo_tensor_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_coo_tensor_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..14d9612f745c57fb07bc8fe3ef4a38629e156359 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_coo_tensor_compositeimplicitautograd_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 compositeimplicitautograd { + +TORCH_API at::Tensor sparse_coo_tensor(const at::Tensor & indices, const at::Tensor & values, at::TensorOptions options={}, c10::optional is_coalesced=c10::nullopt); +TORCH_API at::Tensor sparse_coo_tensor(const at::Tensor & indices, const at::Tensor & values, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional is_coalesced); +TORCH_API at::Tensor sparse_coo_tensor(const at::Tensor & indices, const at::Tensor & values, at::IntArrayRef size, at::TensorOptions options={}, c10::optional is_coalesced=c10::nullopt); +TORCH_API at::Tensor sparse_coo_tensor(const at::Tensor & indices, const at::Tensor & values, at::IntArrayRef size, c10::optional dtype, c10::optional layout, c10::optional device, c10::optional pin_memory, c10::optional is_coalesced); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_resize_and_clear_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_resize_and_clear_native.h new file mode 100644 index 0000000000000000000000000000000000000000..718cd921048c9b2c2e9c092dc1b8edc13cc1263a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sparse_resize_and_clear_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 sparse_resize_and_clear(const at::Tensor & self, at::IntArrayRef size, int64_t sparse_dim, int64_t dense_dim); +TORCH_API const at::Tensor & sparse_resize_and_clear_out(const at::Tensor & self, at::IntArrayRef size, int64_t sparse_dim, int64_t dense_dim, const at::Tensor & out); +TORCH_API const at::Tensor & sparse_resize_and_clear_(const at::Tensor & self, at::IntArrayRef size, int64_t sparse_dim, int64_t dense_dim); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_bessel_j1_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_bessel_j1_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ab8120bd229c623244c5ee02841608ef2dfd8dcc --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_bessel_j1_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor special_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 cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_expm1_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_expm1_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..7d961ab94eb108d301b76abba79cee6f031ef866 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_expm1_compositeimplicitautograd_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor special_expm1(const at::Tensor & self); +TORCH_API at::Tensor & special_expm1_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & special_expm1_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_gammainc_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_gammainc_native.h new file mode 100644 index 0000000000000000000000000000000000000000..0d05d59cc2e0c022402d942a9d738a74fedc2d6b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_gammainc_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor special_gammainc(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & special_gammainc_out(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_hermite_polynomial_he_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_hermite_polynomial_he_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..781f7c8eec6e8616c5d5a5ef445067e586d79220 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_hermite_polynomial_he_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor special_hermite_polynomial_he(const at::Tensor & x, const at::Tensor & n); +TORCH_API at::Tensor & special_hermite_polynomial_he_out(at::Tensor & out, const at::Tensor & x, const at::Tensor & n); +TORCH_API at::Tensor & special_hermite_polynomial_he_outf(const at::Tensor & x, const at::Tensor & n, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_modified_bessel_i0_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_modified_bessel_i0_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..6f57a5360904b20817e617360fb7c38ffba83001 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_modified_bessel_i0_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor special_modified_bessel_i0(const at::Tensor & self); +TORCH_API at::Tensor & special_modified_bessel_i0_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & special_modified_bessel_i0_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/squeeze_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/squeeze_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..81ac367200df3dec057ece4e3682cf3a156af660 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/squeeze_ops.h @@ -0,0 +1,105 @@ +#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 squeeze { + 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::squeeze") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "squeeze(Tensor(a) self) -> Tensor(a)") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +struct TORCH_API squeeze_dim { + using schema = at::Tensor (const at::Tensor &, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::squeeze") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dim") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "squeeze.dim(Tensor(a) self, int dim) -> Tensor(a)") + static at::Tensor call(const at::Tensor & self, int64_t dim); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim); +}; + +struct TORCH_API squeeze_dimname { + using schema = at::Tensor (const at::Tensor &, at::Dimname); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::squeeze") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dimname") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "squeeze.dimname(Tensor(a) self, Dimname dim) -> Tensor(a)") + static at::Tensor call(const at::Tensor & self, at::Dimname dim); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim); +}; + +struct TORCH_API squeeze_dims { + using schema = at::Tensor (const at::Tensor &, at::IntArrayRef); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::squeeze") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dims") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "squeeze.dims(Tensor(a) self, int[] dim) -> Tensor(a)") + static at::Tensor call(const at::Tensor & self, at::IntArrayRef dim); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::IntArrayRef dim); +}; + +struct TORCH_API squeeze_ { + using schema = at::Tensor & (at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::squeeze_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "squeeze_(Tensor(a!) self) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self); +}; + +struct TORCH_API squeeze__dim { + using schema = at::Tensor & (at::Tensor &, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::squeeze_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dim") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "squeeze_.dim(Tensor(a!) self, int dim) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, int64_t dim); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, int64_t dim); +}; + +struct TORCH_API squeeze__dims { + using schema = at::Tensor & (at::Tensor &, at::IntArrayRef); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::squeeze_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dims") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "squeeze_.dims(Tensor(a!) self, int[] dim) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, at::IntArrayRef dim); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, at::IntArrayRef dim); +}; + +struct TORCH_API squeeze__dimname { + using schema = at::Tensor & (at::Tensor &, at::Dimname); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::squeeze_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dimname") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "squeeze_.dimname(Tensor(a!) self, Dimname dim) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self, at::Dimname dim); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, at::Dimname dim); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/stride_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/stride_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..a396933fb4ecddf1bfc13871c7e6323b79f8c28e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/stride_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 stride_int { + using schema = int64_t (const at::Tensor &, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::stride") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "int") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "stride.int(Tensor self, int dim) -> int") + static int64_t call(const at::Tensor & self, int64_t dim); + static int64_t redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim); +}; + +struct TORCH_API stride_Dimname { + using schema = int64_t (const at::Tensor &, at::Dimname); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::stride") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Dimname") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "stride.Dimname(Tensor self, Dimname dim) -> int") + static int64_t call(const at::Tensor & self, at::Dimname dim); + static int64_t redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/take_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/take_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..e85629ac48f0b53748b405fe6394105cb9e5cc23 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/take_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 take_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::take") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "take.out(Tensor self, Tensor index, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & index, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & index, at::Tensor & out); +}; + +struct TORCH_API take { + 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::take") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "take(Tensor self, Tensor index) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & index); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & index); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unfold_copy_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unfold_copy_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..3d2b2d72d1002bb2d6cbc98e11f3d9e0c87eb29c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unfold_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 & unfold_copy_out(at::Tensor & out, const at::Tensor & self, int64_t dimension, int64_t size, int64_t step); +TORCH_API at::Tensor & unfold_copy_outf(const at::Tensor & self, int64_t dimension, int64_t size, int64_t step, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_dim.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_dim.h new file mode 100644 index 0000000000000000000000000000000000000000..fddfb249b720b043cd5006869606b6ff0b3b325d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_dim.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::unique_dim(Tensor self, int dim, bool sorted=True, bool return_inverse=False, bool return_counts=False) -> (Tensor, Tensor, Tensor) +inline ::std::tuple unique_dim(const at::Tensor & self, int64_t dim, bool sorted=true, bool return_inverse=false, bool return_counts=false) { + return at::_ops::unique_dim::call(self, dim, sorted, return_inverse, return_counts); +} + +// aten::unique_dim.out(Tensor self, int dim, bool sorted=True, bool return_inverse=False, bool return_counts=False, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!)) +inline ::std::tuple unique_dim_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, const at::Tensor & self, int64_t dim, bool sorted=true, bool return_inverse=false, bool return_counts=false) { + return at::_ops::unique_dim_out::call(self, dim, sorted, return_inverse, return_counts, out0, out1, out2); +} +// aten::unique_dim.out(Tensor self, int dim, bool sorted=True, bool return_inverse=False, bool return_counts=False, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!)) +inline ::std::tuple unique_dim_outf(const at::Tensor & self, int64_t dim, bool sorted, bool return_inverse, bool return_counts, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2) { + return at::_ops::unique_dim_out::call(self, dim, sorted, return_inverse, return_counts, out0, out1, out2); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest1d_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest1d_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..741fae22935fc5e8405a637582b2faa3d69466d7 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_nearest1d_ops.h @@ -0,0 +1,50 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API upsample_nearest1d_vec { + using schema = at::Tensor (const at::Tensor &, at::OptionalSymIntArrayRef, 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_nearest1d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "vec") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "upsample_nearest1d.vec(Tensor input, SymInt[]? output_size, float[]? scale_factors) -> Tensor") + static at::Tensor call(const at::Tensor & input, at::OptionalSymIntArrayRef output_size, c10::optional> scale_factors); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, at::OptionalSymIntArrayRef output_size, c10::optional> scale_factors); +}; + +struct TORCH_API upsample_nearest1d_out { + using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, 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_nearest1d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "upsample_nearest1d.out(Tensor self, SymInt[1] output_size, float? scales=None, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, c10::SymIntArrayRef output_size, c10::optional scales, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size, c10::optional scales, at::Tensor & out); +}; + +struct TORCH_API upsample_nearest1d { + using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef, 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_nearest1d") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "upsample_nearest1d(Tensor self, SymInt[1] output_size, float? scales=None) -> Tensor") + static at::Tensor call(const at::Tensor & self, c10::SymIntArrayRef output_size, c10::optional scales); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size, c10::optional scales); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/var_mean_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/var_mean_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..e8ade74d09e689d7e552ad1d6326cb11705a0efb --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/var_mean_cpu_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API ::std::tuple var_mean(const at::Tensor & self, at::OptionalIntArrayRef dim=c10::nullopt, const c10::optional & correction=c10::nullopt, bool keepdim=false); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/zero_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/zero_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..533113c43f0fdae65703d1221af15b4e30aa8f57 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/zero_cpu_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor & zero_(at::Tensor & self); + +} // namespace cpu +} // namespace at