Add files using upload-large-folder tool

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_addmm_activation_meta.h

@@ -0,0 +1,27 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeMetaFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/TensorIterator.h>
+#include <ATen/TensorMeta.h>
+#include <tuple>
+#include <vector>
+
+namespace at {
+namespace meta {
+
+struct TORCH_API structured__addmm_activation : public at::impl::MetaBase {
+    
+    
+    void meta(const at::Tensor & self, const at::Tensor & mat1, const at::Tensor & mat2, const at::Scalar & beta, const at::Scalar & alpha, bool use_gelu);
+};
+
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_impl_index_backward_native.h

@@ -0,0 +1,21 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor> _batch_norm_impl_index_backward(int64_t impl_index, const at::Tensor & input, const at::Tensor & grad_output, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & running_mean, const ::std::optional<at::Tensor> & running_var, const ::std::optional<at::Tensor> & save_mean, const ::std::optional<at::Tensor> & save_var_transform, bool train, double eps, ::std::array<bool,3> output_mask, const at::Tensor & reservedSpace);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_convert_indices_from_coo_to_csr_native.h

@@ -0,0 +1,26 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+#include <ATen/ops/_convert_indices_from_coo_to_csr_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured__convert_indices_from_coo_to_csr_structured_cpu : public at::meta::structured__convert_indices_from_coo_to_csr {
+void impl(const at::Tensor & self, int64_t size, bool out_int32, const at::Tensor & out);
+};
+struct TORCH_API structured__convert_indices_from_coo_to_csr_structured_cuda : public at::meta::structured__convert_indices_from_coo_to_csr {
+void impl(const at::Tensor & self, int64_t size, bool out_int32, const at::Tensor & out);
+};
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cudnn_rnn_backward_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API void _cudnn_rnn_backward_out_symint(const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const ::std::optional<at::Tensor> & cx, const at::Tensor & output, const ::std::optional<at::Tensor> & grad_output, const ::std::optional<at::Tensor> & grad_hy, const ::std::optional<at::Tensor> & grad_cy, int64_t mode, c10::SymInt hidden_size, c10::SymInt proj_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, c10::SymIntArrayRef batch_sizes, const ::std::optional<at::Tensor> & dropout_state, const at::Tensor & reserve, ::std::array<bool,4> output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::TensorList out3);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor,::std::vector<at::Tensor>> _cudnn_rnn_backward(const at::Tensor & input, at::TensorList weight, int64_t weight_stride0, const at::Tensor & weight_buf, const at::Tensor & hx, const ::std::optional<at::Tensor> & cx, const at::Tensor & output, const ::std::optional<at::Tensor> & grad_output, const ::std::optional<at::Tensor> & grad_hy, const ::std::optional<at::Tensor> & grad_cy, int64_t mode, int64_t hidden_size, int64_t proj_size, int64_t num_layers, bool batch_first, double dropout, bool train, bool bidirectional, at::IntArrayRef batch_sizes, const ::std::optional<at::Tensor> & dropout_state, const at::Tensor & reserve, ::std::array<bool,4> output_mask);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_add_compositeexplicitautograd_dispatch.h

@@ -0,0 +1,38 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeexplicitautograd {
+
+TORCH_API ::std::vector<at::Tensor> _foreach_add(at::TensorList self, const at::Scalar & scalar);
+TORCH_API void _foreach_add_out(at::TensorList out, at::TensorList self, const at::Scalar & scalar);
+TORCH_API void _foreach_add_outf(at::TensorList self, const at::Scalar & scalar, at::TensorList out);
+TORCH_API void _foreach_add_(at::TensorList self, const at::Scalar & scalar);
+TORCH_API ::std::vector<at::Tensor> _foreach_add(at::TensorList self, at::TensorList other, const at::Scalar & alpha=1);
+TORCH_API void _foreach_add_out(at::TensorList out, at::TensorList self, at::TensorList other, const at::Scalar & alpha=1);
+TORCH_API void _foreach_add_outf(at::TensorList self, at::TensorList other, const at::Scalar & alpha, at::TensorList out);
+TORCH_API void _foreach_add_(at::TensorList self, at::TensorList other, const at::Scalar & alpha=1);
+TORCH_API ::std::vector<at::Tensor> _foreach_add(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API void _foreach_add_out(at::TensorList out, at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API void _foreach_add_outf(at::TensorList self, at::ArrayRef<at::Scalar> scalars, at::TensorList out);
+TORCH_API void _foreach_add_(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API ::std::vector<at::Tensor> _foreach_add(at::TensorList self, const at::Tensor & other, const at::Scalar & alpha=1);
+TORCH_API void _foreach_add_out(at::TensorList out, at::TensorList self, const at::Tensor & other, const at::Scalar & alpha=1);
+TORCH_API void _foreach_add_outf(at::TensorList self, const at::Tensor & other, const at::Scalar & alpha, at::TensorList out);
+TORCH_API void _foreach_add_(at::TensorList self, const at::Tensor & other, const at::Scalar & alpha=1);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_maximum_native.h

@@ -0,0 +1,35 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_min_scalar_kernel_slow(at::TensorList self, const at::Scalar & scalar);
+TORCH_API void _foreach_maximum_Scalar_out(at::TensorList self, const at::Scalar & scalar, at::TensorList out);
+TORCH_API void foreach_tensor_clamp_min_scalar_kernel_slow_(at::TensorList self, const at::Scalar & scalar);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_min_scalar_kernel_cuda(at::TensorList self, const at::Scalar & scalar);
+TORCH_API void foreach_tensor_clamp_min_scalar_kernel_cuda_(at::TensorList self, const at::Scalar & scalar);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_min_list_kernel_slow(at::TensorList self, at::TensorList other);
+TORCH_API void _foreach_maximum_List_out(at::TensorList self, at::TensorList other, at::TensorList out);
+TORCH_API void foreach_tensor_clamp_min_list_kernel_slow_(at::TensorList self, at::TensorList other);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_min_list_kernel_cuda(at::TensorList self, at::TensorList other);
+TORCH_API void foreach_tensor_clamp_min_list_kernel_cuda_(at::TensorList self, at::TensorList other);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_min_scalarlist_kernel_slow(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API void _foreach_maximum_ScalarList_out(at::TensorList self, at::ArrayRef<at::Scalar> scalars, at::TensorList out);
+TORCH_API void foreach_tensor_clamp_min_scalarlist_kernel_slow_(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_min_scalarlist_kernel_cuda(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API void foreach_tensor_clamp_min_scalarlist_kernel_cuda_(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sign_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API ::std::vector<at::Tensor> _foreach_sign(at::TensorList self);
+TORCH_API void _foreach_sign_(at::TensorList self);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_tanh_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API ::std::vector<at::Tensor> _foreach_tanh(at::TensorList self);
+TORCH_API void _foreach_tanh_(at::TensorList self);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API at::Tensor & _index_put_impl_(at::Tensor & self, const c10::List<::std::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_index_put_impl_native.h

@@ -0,0 +1,25 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor _index_put_impl(const at::Tensor & self, const c10::List<::std::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false);
+TORCH_API at::Tensor & _index_put_impl_out(const at::Tensor & self, const c10::List<::std::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate, bool unsafe, at::Tensor & out);
+TORCH_API at::Tensor & _index_put_impl_(at::Tensor & self, const c10::List<::std::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false);
+TORCH_API at::Tensor & _index_put_impl_quantized_cpu_(at::Tensor & self, const c10::List<::std::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false);
+TORCH_API at::Tensor & _index_put_impl_quantized_cuda_(at::Tensor & self, const c10::List<::std::optional<at::Tensor>> & indices, const at::Tensor & values, bool accumulate=false, bool unsafe=false);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_neg_view_native.h

@@ -0,0 +1,21 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor _neg_view(const at::Tensor & self);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_storage_offsets_ops.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API _nested_tensor_storage_offsets {
+  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::_nested_tensor_storage_offsets")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_tensor_storage_offsets(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 _nested_tensor_storage_offsets_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::_nested_tensor_storage_offsets")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_tensor_storage_offsets.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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_prelu_kernel_backward_ops.h

@@ -0,0 +1,28 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API _prelu_kernel_backward {
+  using schema = ::std::tuple<at::Tensor,at::Tensor> (const at::Tensor &, const at::Tensor &, const at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_prelu_kernel_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_prelu_kernel_backward(Tensor grad_output, Tensor self, Tensor weight) -> (Tensor, Tensor)")
+  static ::std::tuple<at::Tensor,at::Tensor> call(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight);
+  static ::std::tuple<at::Tensor,at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_reshape_alias_copy_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor & _reshape_alias_copy_out_symint(const at::Tensor & self, c10::SymIntArrayRef size, c10::SymIntArrayRef stride, at::Tensor & out);
+TORCH_API at::Tensor _reshape_alias_copy_symint(const at::Tensor & self, c10::SymIntArrayRef size, c10::SymIntArrayRef stride);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_bsr_tensor_unsafe.h

@@ -0,0 +1,34 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/_sparse_bsr_tensor_unsafe_ops.h>
+
+namespace at {
+
+
+// aten::_sparse_bsr_tensor_unsafe(Tensor crow_indices, Tensor col_indices, Tensor values, int[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor _sparse_bsr_tensor_unsafe(const at::Tensor & crow_indices, const at::Tensor & col_indices, const at::Tensor & values, at::IntArrayRef size, at::TensorOptions options={}) {
+    return at::_ops::_sparse_bsr_tensor_unsafe::call(crow_indices, col_indices, values, size, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+}
+// aten::_sparse_bsr_tensor_unsafe(Tensor crow_indices, Tensor col_indices, Tensor values, int[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor _sparse_bsr_tensor_unsafe(const at::Tensor & crow_indices, const at::Tensor & col_indices, const at::Tensor & values, at::IntArrayRef size, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory) {
+    return at::_ops::_sparse_bsr_tensor_unsafe::call(crow_indices, col_indices, values, size, dtype, layout, device, pin_memory);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_csc_tensor_unsafe.h

@@ -0,0 +1,34 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/_sparse_csc_tensor_unsafe_ops.h>
+
+namespace at {
+
+
+// aten::_sparse_csc_tensor_unsafe(Tensor ccol_indices, Tensor row_indices, Tensor values, int[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor _sparse_csc_tensor_unsafe(const at::Tensor & ccol_indices, const at::Tensor & row_indices, const at::Tensor & values, at::IntArrayRef size, at::TensorOptions options={}) {
+    return at::_ops::_sparse_csc_tensor_unsafe::call(ccol_indices, row_indices, values, size, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+}
+// aten::_sparse_csc_tensor_unsafe(Tensor ccol_indices, Tensor row_indices, Tensor values, int[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor _sparse_csc_tensor_unsafe(const at::Tensor & ccol_indices, const at::Tensor & row_indices, const at::Tensor & values, at::IntArrayRef size, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory) {
+    return at::_ops::_sparse_csc_tensor_unsafe::call(ccol_indices, row_indices, values, size, dtype, layout, device, pin_memory);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_csr_sum_native.h

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor & _sparse_csr_sum_dim_dtype_out(const at::Tensor & self, at::IntArrayRef dim, bool keepdim, ::std::optional<at::ScalarType> dtype, at::Tensor & out);
+TORCH_API at::Tensor _sparse_csr_sum_cpu(const at::Tensor & self, at::IntArrayRef dim, bool keepdim=false, ::std::optional<at::ScalarType> dtype=::std::nullopt);
+TORCH_API at::Tensor _sparse_csr_sum_cuda(const at::Tensor & self, at::IntArrayRef dim, bool keepdim=false, ::std::optional<at::ScalarType> dtype=::std::nullopt);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_serialization_subcmul.h

@@ -0,0 +1,30 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/_test_serialization_subcmul_ops.h>
+
+namespace at {
+
+
+// aten::_test_serialization_subcmul(Tensor self, Tensor other, Scalar alpha=1) -> Tensor
+inline at::Tensor _test_serialization_subcmul(const at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha=1) {
+    return at::_ops::_test_serialization_subcmul::call(self, other, alpha);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unsafe_masked_index_put_accumulate_native.h

@@ -0,0 +1,21 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor _unsafe_masked_index_put_accumulate(const at::Tensor & self, const at::Tensor & mask, const c10::List<::std::optional<at::Tensor>> & indices, const at::Tensor & values);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_ops.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API adaptive_avg_pool2d_out {
+  using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::adaptive_avg_pool2d")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "adaptive_avg_pool2d.out(Tensor self, SymInt[2] output_size, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, c10::SymIntArrayRef output_size, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size, at::Tensor & out);
+};
+
+struct TORCH_API adaptive_avg_pool2d {
+  using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::adaptive_avg_pool2d")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "adaptive_avg_pool2d(Tensor self, SymInt[2] output_size) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, c10::SymIntArrayRef output_size);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addmm_meta.h

@@ -0,0 +1,27 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeMetaFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/TensorIterator.h>
+#include <ATen/TensorMeta.h>
+#include <tuple>
+#include <vector>
+
+namespace at {
+namespace meta {
+
+struct TORCH_API structured_addmm : public at::impl::MetaBase {
+    
+    
+    void meta(const at::Tensor & self, const at::Tensor & mat1, const at::Tensor & mat2, const at::Scalar & beta, const at::Scalar & alpha);
+};
+
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/arccos.h

@@ -0,0 +1,44 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/arccos_ops.h>
+
+namespace at {
+
+
+// aten::arccos(Tensor self) -> Tensor
+inline at::Tensor arccos(const at::Tensor & self) {
+    return at::_ops::arccos::call(self);
+}
+
+// aten::arccos_(Tensor(a!) self) -> Tensor(a!)
+inline at::Tensor & arccos_(at::Tensor & self) {
+    return at::_ops::arccos_::call(self);
+}
+
+// aten::arccos.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & arccos_out(at::Tensor & out, const at::Tensor & self) {
+    return at::_ops::arccos_out::call(self, out);
+}
+// aten::arccos.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & arccos_outf(const at::Tensor & self, at::Tensor & out) {
+    return at::_ops::arccos_out::call(self, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool1d.h

@@ -0,0 +1,30 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/avg_pool1d_ops.h>
+
+namespace at {
+
+
+// aten::avg_pool1d(Tensor self, int[1] kernel_size, int[1] stride=[], int[1] padding=0, bool ceil_mode=False, bool count_include_pad=True) -> Tensor
+inline 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) {
+    return at::_ops::avg_pool1d::call(self, kernel_size, stride, padding, ceil_mode, count_include_pad);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/batch_norm_backward_reduce_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeexplicitautograd {
+
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &,at::Tensor &> batch_norm_backward_reduce_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3, const at::Tensor & grad_out, const at::Tensor & input, const at::Tensor & mean, const at::Tensor & invstd, const ::std::optional<at::Tensor> & weight, bool input_g, bool weight_g, bool bias_g);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &,at::Tensor &> batch_norm_backward_reduce_outf(const at::Tensor & grad_out, const at::Tensor & input, const at::Tensor & mean, const at::Tensor & invstd, const ::std::optional<at::Tensor> & weight, bool input_g, bool weight_g, bool bias_g, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, at::Tensor & out3);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/chain_matmul_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor chain_matmul(at::TensorList matrices);
+TORCH_API at::Tensor & chain_matmul_out(at::TensorList matrices, at::Tensor & out);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_convolution_relu.h

@@ -0,0 +1,91 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/cudnn_convolution_relu_ops.h>
+
+namespace at {
+
+
+// aten::cudnn_convolution_relu(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups) -> Tensor
+inline at::Tensor cudnn_convolution_relu(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups) {
+    return at::_ops::cudnn_convolution_relu::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor cudnn_convolution_relu(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups) {
+    return at::_ops::cudnn_convolution_relu::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups);
+  }
+}
+
+// aten::cudnn_convolution_relu(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups) -> Tensor
+inline at::Tensor cudnn_convolution_relu_symint(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups) {
+    return at::_ops::cudnn_convolution_relu::call(self, weight, bias, stride, padding, dilation, groups);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor cudnn_convolution_relu(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups) {
+    return at::_ops::cudnn_convolution_relu::call(self, weight, bias, stride, padding, dilation, groups);
+  }
+}
+
+// aten::cudnn_convolution_relu.out(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & cudnn_convolution_relu_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups) {
+    return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & cudnn_convolution_relu_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups) {
+    return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups, out);
+  }
+}
+
+// aten::cudnn_convolution_relu.out(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & cudnn_convolution_relu_outf(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups, at::Tensor & out) {
+    return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & cudnn_convolution_relu_outf(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, int64_t groups, at::Tensor & out) {
+    return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, c10::fromIntArrayRefSlow(stride), c10::fromIntArrayRefSlow(padding), c10::fromIntArrayRefSlow(dilation), groups, out);
+  }
+}
+
+// aten::cudnn_convolution_relu.out(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & cudnn_convolution_relu_symint_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups) {
+    return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, stride, padding, dilation, groups, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & cudnn_convolution_relu_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups) {
+    return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, stride, padding, dilation, groups, out);
+  }
+}
+
+// aten::cudnn_convolution_relu.out(Tensor self, Tensor weight, Tensor? bias, SymInt[] stride, SymInt[] padding, SymInt[] dilation, SymInt groups, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & cudnn_convolution_relu_symint_outf(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups, at::Tensor & out) {
+    return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, stride, padding, dilation, groups, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & cudnn_convolution_relu_outf(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, c10::SymInt groups, at::Tensor & out) {
+    return at::_ops::cudnn_convolution_relu_out::call(self, weight, bias, stride, padding, dilation, groups, out);
+  }
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cummax_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeimplicitautograd {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> cummax(const at::Tensor & self, at::Dimname dim);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> cummax_out(at::Tensor & values, at::Tensor & indices, const at::Tensor & self, at::Dimname dim);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> cummax_outf(const at::Tensor & self, at::Dimname dim, at::Tensor & values, at::Tensor & indices);
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/dense_dim.h

@@ -0,0 +1,26 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/dense_dim_ops.h>
+
+namespace at {
+
+
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor detach(const at::Tensor & self);
+TORCH_API at::Tensor & detach_(at::Tensor & self);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagflat_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeimplicitautograd {
+
+TORCH_API at::Tensor diagflat(const at::Tensor & self, int64_t offset=0);
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_native.h

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+#include <ATen/ops/digamma_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_digamma_out : public at::meta::structured_digamma {
+void impl(const at::Tensor & self, const at::Tensor & out);
+};
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/eq_compositeexplicitautogradnonfunctional_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeexplicitautogradnonfunctional {
+
+TORCH_API at::Tensor eq(const at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor & eq_(at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor eq(const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & eq_(at::Tensor & self, const at::Tensor & other);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_linear_fp16_weight_fp32_activation_compositeimplicitautograd_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeimplicitautograd {
+
+TORCH_API at::Tensor fbgemm_linear_fp16_weight_fp32_activation(const at::Tensor & input, const at::Tensor & packed_weight, const at::Tensor & bias);
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_fftfreq_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor fft_fftfreq(int64_t n, double d=1.0, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={});
+TORCH_API at::Tensor & fft_fftfreq_out(int64_t n, double d, at::Tensor & out);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/flip.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/flip_ops.h>
+
+namespace at {
+
+
+// aten::flip(Tensor self, int[] dims) -> Tensor
+inline at::Tensor flip(const at::Tensor & self, at::IntArrayRef dims) {
+    return at::_ops::flip::call(self, dims);
+}
+
+// aten::flip.out(Tensor self, int[] dims, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & flip_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef dims) {
+    return at::_ops::flip_out::call(self, dims, out);
+}
+// aten::flip.out(Tensor self, int[] dims, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & flip_outf(const at::Tensor & self, at::IntArrayRef dims, at::Tensor & out) {
+    return at::_ops::flip_out::call(self, dims, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gelu_native.h

@@ -0,0 +1,32 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+#include <ATen/ops/gelu_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_gelu_out_cpu : public at::meta::structured_gelu {
+void impl(const at::Tensor & self, c10::string_view approximate, const at::Tensor & out);
+};
+struct TORCH_API structured_gelu_out_cuda : public at::meta::structured_gelu {
+void impl(const at::Tensor & self, c10::string_view approximate, const at::Tensor & out);
+};
+TORCH_API at::Tensor NestedTensor_gelu(const at::Tensor & self, c10::string_view approximate="none");
+TORCH_API at::Tensor & NestedTensor_gelu_(at::Tensor & self, c10::string_view approximate="none");
+TORCH_API at::Tensor mkldnn_gelu(const at::Tensor & self, c10::string_view approximate="none");
+TORCH_API at::Tensor gelu_quantized_cpu(const at::Tensor & self, c10::string_view approximate="none");
+TORCH_API at::Tensor & gelu_quantized_cpu_(at::Tensor & self, c10::string_view approximate="none");
+TORCH_API at::Tensor gelu_quantized_cuda(const at::Tensor & self, c10::string_view approximate="none");
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ger.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/ger_ops.h>
+
+namespace at {
+
+
+// aten::ger(Tensor self, Tensor vec2) -> Tensor
+inline at::Tensor ger(const at::Tensor & self, const at::Tensor & vec2) {
+    return at::_ops::ger::call(self, vec2);
+}
+
+// aten::ger.out(Tensor self, Tensor vec2, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & ger_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & vec2) {
+    return at::_ops::ger_out::call(self, vec2, out);
+}
+// aten::ger.out(Tensor self, Tensor vec2, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & ger_outf(const at::Tensor & self, const at::Tensor & vec2, at::Tensor & out) {
+    return at::_ops::ger_out::call(self, vec2, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu_backward_cuda_dispatch.h

@@ -0,0 +1,25 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API at::Tensor glu_backward(const at::Tensor & grad_output, const at::Tensor & self, int64_t dim);
+TORCH_API at::Tensor & glu_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, int64_t dim);
+TORCH_API at::Tensor & glu_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, int64_t dim, at::Tensor & grad_input);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gru_cell_compositeimplicitautograd_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace compositeimplicitautograd {
+
+TORCH_API at::Tensor gru_cell(const at::Tensor & input, const at::Tensor & hx, const at::Tensor & w_ih, const at::Tensor & w_hh, const ::std::optional<at::Tensor> & b_ih={}, const ::std::optional<at::Tensor> & b_hh={});
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/hardshrink_ops.h>
+
+namespace at {
+
+
+// aten::hardshrink.out(Tensor self, Scalar lambd=0.5, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & hardshrink_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & lambd=0.5) {
+    return at::_ops::hardshrink_out::call(self, lambd, out);
+}
+// aten::hardshrink.out(Tensor self, Scalar lambd=0.5, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & hardshrink_outf(const at::Tensor & self, const at::Scalar & lambd, at::Tensor & out) {
+    return at::_ops::hardshrink_out::call(self, lambd, out);
+}
+
+// aten::hardshrink(Tensor self, Scalar lambd=0.5) -> Tensor
+inline at::Tensor hardshrink(const at::Tensor & self, const at::Scalar & lambd=0.5) {
+    return at::_ops::hardshrink::call(self, lambd);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardswish_native.h

@@ -0,0 +1,23 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor hardswish(const at::Tensor & self);
+TORCH_API at::Tensor & hardswish_out(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & hardswish_(at::Tensor & self);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/im2col.h

@@ -0,0 +1,39 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/im2col_ops.h>
+
+namespace at {
+
+
+// aten::im2col.out(Tensor self, int[2] kernel_size, int[2] dilation, int[2] padding, int[2] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & im2col_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::im2col_out::call(self, kernel_size, dilation, padding, stride, out);
+}
+// aten::im2col.out(Tensor self, int[2] kernel_size, int[2] dilation, int[2] padding, int[2] stride, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & im2col_outf(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride, at::Tensor & out) {
+    return at::_ops::im2col_out::call(self, kernel_size, dilation, padding, stride, out);
+}
+
+// aten::im2col(Tensor self, int[2] kernel_size, int[2] dilation, int[2] padding, int[2] stride) -> Tensor
+inline at::Tensor im2col(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride) {
+    return at::_ops::im2col::call(self, kernel_size, dilation, padding, stride);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isin_cpu_dispatch.h

@@ -0,0 +1,31 @@
+#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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API at::Tensor isin(const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false);
+TORCH_API at::Tensor & isin_out(at::Tensor & out, const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false);
+TORCH_API at::Tensor & isin_outf(const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out);
+TORCH_API at::Tensor isin(const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique=false, bool invert=false);
+TORCH_API at::Tensor & isin_out(at::Tensor & out, const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique=false, bool invert=false);
+TORCH_API at::Tensor & isin_outf(const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique, bool invert, at::Tensor & out);
+TORCH_API at::Tensor isin(const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false);
+TORCH_API at::Tensor & isin_out(at::Tensor & out, const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique=false, bool invert=false);
+TORCH_API at::Tensor & isin_outf(const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/istft_native.h

@@ -0,0 +1,21 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor istft(const at::Tensor & self, int64_t n_fft, ::std::optional<int64_t> hop_length=::std::nullopt, ::std::optional<int64_t> win_length=::std::nullopt, const ::std::optional<at::Tensor> & window={}, bool center=true, bool normalized=false, ::std::optional<bool> onesided=::std::nullopt, ::std::optional<int64_t> length=::std::nullopt, bool return_complex=false);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log_sigmoid_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor log_sigmoid(const at::Tensor & self);
+TORCH_API at::Tensor & log_sigmoid_out(const at::Tensor & self, at::Tensor & out);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logit_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace meta {
+
+TORCH_API at::Tensor logit_backward(const at::Tensor & grad_output, const at::Tensor & self, ::std::optional<double> eps=::std::nullopt);
+TORCH_API at::Tensor & logit_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, ::std::optional<double> eps=::std::nullopt);
+TORCH_API at::Tensor & logit_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, ::std::optional<double> eps, at::Tensor & grad_input);
+
+} // namespace meta
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_convolution_transpose_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor & miopen_convolution_transpose_out_symint(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef output_padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, at::Tensor & out);
+TORCH_API at::Tensor miopen_convolution_transpose(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, at::IntArrayRef padding, at::IntArrayRef output_padding, at::IntArrayRef stride, at::IntArrayRef dilation, int64_t groups, bool benchmark, bool deterministic);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_convolution_native.h

@@ -0,0 +1,22 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+#include <c10/core/QScheme.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <tuple>
+#include <vector>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor mkldnn_convolution(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, at::IntArrayRef padding, at::IntArrayRef stride, at::IntArrayRef dilation, int64_t groups);
+TORCH_API at::Tensor & mkldnn_convolution_out_symint(const at::Tensor & self, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, at::Tensor & out);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mkldnn_rnn_layer_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor> mkldnn_rnn_layer(const at::Tensor & input, const at::Tensor & weight0, const at::Tensor & weight1, const at::Tensor & weight2, const at::Tensor & weight3, const at::Tensor & hx_, const at::Tensor & cx_, bool reverse, at::IntArrayRef batch_sizes, int64_t mode, int64_t hidden_size, int64_t num_layers, bool has_biases, bool bidirectional, bool batch_first, bool train);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/nansum_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 <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// 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 <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API at::Tensor nansum(const at::Tensor & self, at::OptionalIntArrayRef dim=::std::nullopt, bool keepdim=false, ::std::optional<at::ScalarType> dtype=::std::nullopt);
+TORCH_API at::Tensor & nansum_out(at::Tensor & out, const at::Tensor & self, at::OptionalIntArrayRef dim=::std::nullopt, bool keepdim=false, ::std::optional<at::ScalarType> dtype=::std::nullopt);
+TORCH_API at::Tensor & nansum_outf(const at::Tensor & self, at::OptionalIntArrayRef dim, bool keepdim, ::std::optional<at::ScalarType> dtype, at::Tensor & out);
+
+} // namespace cuda
+} // namespace at