Add files using upload-large-folder tool

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_adaptive_avg_pool2d_backward_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_backward {
+  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::_adaptive_avg_pool2d_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_adaptive_avg_pool2d_backward(Tensor grad_output, Tensor self) -> Tensor")
+  static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & self);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self);
+};
+
+struct TORCH_API _adaptive_avg_pool2d_backward_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::_adaptive_avg_pool2d_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_adaptive_avg_pool2d_backward.out(Tensor grad_output, Tensor self, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & out);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_assert_scalar_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 void _assert_scalar(const at::Scalar & self, c10::string_view assert_msg);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cast_Double_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 _cast_Double(const at::Tensor & self, bool non_blocking=false);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cdist_backward_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 _cdist_backward(const at::Tensor & grad, const at::Tensor & x1, const at::Tensor & x2, double p, const at::Tensor & cdist);
+
+} // namespace cpu
+} // namespace at

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

@@ -0,0 +1,30 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <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/_debug_has_internal_overlap_ops.h>
+
+namespace at {
+
+
+// aten::_debug_has_internal_overlap(Tensor self) -> int
+inline int64_t _debug_has_internal_overlap(const at::Tensor & self) {
+    return at::_ops::_debug_has_internal_overlap::call(self);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_embedding_bag_sparse_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 at::Tensor _embedding_bag_sparse_backward_symint(const at::Tensor & grad, const at::Tensor & indices, const at::Tensor & offsets, const at::Tensor & offset2bag, const at::Tensor & bag_size, c10::SymInt num_weights, bool scale_grad_by_freq, int64_t mode, const ::std::optional<at::Tensor> & per_sample_weights, int64_t padding_idx=-1);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_acos_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 <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_acos(at::TensorList self);
+TORCH_API void _foreach_acos_out(at::TensorList out, at::TensorList self);
+TORCH_API void _foreach_acos_outf(at::TensorList self, at::TensorList out);
+TORCH_API void _foreach_acos_(at::TensorList self);
+
+} // namespace compositeexplicitautograd
+} // namespace at

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

@@ -0,0 +1,35 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <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_max_scalar_kernel_slow(at::TensorList self, const at::Scalar & scalar);
+TORCH_API void _foreach_clamp_max_Scalar_out(at::TensorList self, const at::Scalar & scalar, at::TensorList out);
+TORCH_API void foreach_tensor_clamp_max_scalar_kernel_slow_(at::TensorList self, const at::Scalar & scalar);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_max_scalar_kernel_cuda(at::TensorList self, const at::Scalar & scalar);
+TORCH_API void foreach_tensor_clamp_max_scalar_kernel_cuda_(at::TensorList self, const at::Scalar & scalar);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_max_list_kernel_slow(at::TensorList self, at::TensorList other);
+TORCH_API void _foreach_clamp_max_List_out(at::TensorList self, at::TensorList other, at::TensorList out);
+TORCH_API void foreach_tensor_clamp_max_list_kernel_slow_(at::TensorList self, at::TensorList other);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_max_list_kernel_cuda(at::TensorList self, at::TensorList other);
+TORCH_API void foreach_tensor_clamp_max_list_kernel_cuda_(at::TensorList self, at::TensorList other);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_max_scalarlist_kernel_slow(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API void _foreach_clamp_max_ScalarList_out(at::TensorList self, at::ArrayRef<at::Scalar> scalars, at::TensorList out);
+TORCH_API void foreach_tensor_clamp_max_scalarlist_kernel_slow_(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_clamp_max_scalarlist_kernel_cuda(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API void foreach_tensor_clamp_max_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_clamp_min.h

@@ -0,0 +1,82 @@
+#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/_foreach_clamp_min_ops.h>
+
+namespace at {
+
+
+// aten::_foreach_clamp_min.Scalar(Tensor[] self, Scalar scalar) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_clamp_min(at::TensorList self, const at::Scalar & scalar) {
+    return at::_ops::_foreach_clamp_min_Scalar::call(self, scalar);
+}
+
+// aten::_foreach_clamp_min_.Scalar(Tensor(a!)[] self, Scalar scalar) -> ()
+inline void _foreach_clamp_min_(at::TensorList self, const at::Scalar & scalar) {
+    return at::_ops::_foreach_clamp_min__Scalar::call(self, scalar);
+}
+
+// aten::_foreach_clamp_min.List(Tensor[] self, Tensor[] other) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_clamp_min(at::TensorList self, at::TensorList other) {
+    return at::_ops::_foreach_clamp_min_List::call(self, other);
+}
+
+// aten::_foreach_clamp_min_.List(Tensor(a!)[] self, Tensor[] other) -> ()
+inline void _foreach_clamp_min_(at::TensorList self, at::TensorList other) {
+    return at::_ops::_foreach_clamp_min__List::call(self, other);
+}
+
+// aten::_foreach_clamp_min.ScalarList(Tensor[] self, Scalar[] scalars) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_clamp_min(at::TensorList self, at::ArrayRef<at::Scalar> scalars) {
+    return at::_ops::_foreach_clamp_min_ScalarList::call(self, scalars);
+}
+
+// aten::_foreach_clamp_min_.ScalarList(Tensor(a!)[] self, Scalar[] scalars) -> ()
+inline void _foreach_clamp_min_(at::TensorList self, at::ArrayRef<at::Scalar> scalars) {
+    return at::_ops::_foreach_clamp_min__ScalarList::call(self, scalars);
+}
+
+// aten::_foreach_clamp_min.Scalar_out(Tensor[] self, Scalar scalar, *, Tensor(a!)[] out) -> ()
+inline void _foreach_clamp_min_out(at::TensorList out, at::TensorList self, const at::Scalar & scalar) {
+    return at::_ops::_foreach_clamp_min_Scalar_out::call(self, scalar, out);
+}
+// aten::_foreach_clamp_min.Scalar_out(Tensor[] self, Scalar scalar, *, Tensor(a!)[] out) -> ()
+inline void _foreach_clamp_min_outf(at::TensorList self, const at::Scalar & scalar, at::TensorList out) {
+    return at::_ops::_foreach_clamp_min_Scalar_out::call(self, scalar, out);
+}
+
+// aten::_foreach_clamp_min.List_out(Tensor[] self, Tensor[] other, *, Tensor(a!)[] out) -> ()
+inline void _foreach_clamp_min_out(at::TensorList out, at::TensorList self, at::TensorList other) {
+    return at::_ops::_foreach_clamp_min_List_out::call(self, other, out);
+}
+// aten::_foreach_clamp_min.List_out(Tensor[] self, Tensor[] other, *, Tensor(a!)[] out) -> ()
+inline void _foreach_clamp_min_outf(at::TensorList self, at::TensorList other, at::TensorList out) {
+    return at::_ops::_foreach_clamp_min_List_out::call(self, other, out);
+}
+
+// aten::_foreach_clamp_min.ScalarList_out(Tensor[] self, Scalar[] scalars, *, Tensor(a!)[] out) -> ()
+inline void _foreach_clamp_min_out(at::TensorList out, at::TensorList self, at::ArrayRef<at::Scalar> scalars) {
+    return at::_ops::_foreach_clamp_min_ScalarList_out::call(self, scalars, out);
+}
+// aten::_foreach_clamp_min.ScalarList_out(Tensor[] self, Scalar[] scalars, *, Tensor(a!)[] out) -> ()
+inline void _foreach_clamp_min_outf(at::TensorList self, at::ArrayRef<at::Scalar> scalars, at::TensorList out) {
+    return at::_ops::_foreach_clamp_min_ScalarList_out::call(self, scalars, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_clamp_min_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 <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_clamp_min(at::TensorList self, const at::Scalar & scalar);
+TORCH_API void _foreach_clamp_min_(at::TensorList self, const at::Scalar & scalar);
+TORCH_API ::std::vector<at::Tensor> _foreach_clamp_min(at::TensorList self, at::TensorList other);
+TORCH_API void _foreach_clamp_min_(at::TensorList self, at::TensorList other);
+TORCH_API ::std::vector<at::Tensor> _foreach_clamp_min(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+TORCH_API void _foreach_clamp_min_(at::TensorList self, at::ArrayRef<at::Scalar> scalars);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log2_native.h

@@ -0,0 +1,25 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeFunction.h
+
+#include <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_log2_slow(at::TensorList self);
+TORCH_API void _foreach_log2_out(at::TensorList self, at::TensorList out);
+TORCH_API void foreach_tensor_log2_slow_(at::TensorList self);
+TORCH_API ::std::vector<at::Tensor> foreach_tensor_log2_cuda(at::TensorList self);
+TORCH_API void foreach_tensor_log2_cuda_(at::TensorList self);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,58 @@
+#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/_native_batch_norm_legit_ops.h>
+
+namespace at {
+
+
+// aten::_native_batch_norm_legit(Tensor input, Tensor? weight, Tensor? bias, Tensor(a!) running_mean, Tensor(b!) running_var, bool training, float momentum, float eps) -> (Tensor, Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor,at::Tensor> _native_batch_norm_legit(const at::Tensor & input, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & bias, at::Tensor & running_mean, at::Tensor & running_var, bool training, double momentum, double eps) {
+    return at::_ops::_native_batch_norm_legit::call(input, weight, bias, running_mean, running_var, training, momentum, eps);
+}
+
+// aten::_native_batch_norm_legit.out(Tensor input, Tensor? weight, Tensor? bias, Tensor(a!) running_mean, Tensor(b!) running_var, bool training, float momentum, float eps, *, Tensor(d!) out, Tensor(e!) save_mean, Tensor(f!) save_invstd) -> (Tensor(d!), Tensor(e!), Tensor(f!))
+inline ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> _native_batch_norm_legit_out(at::Tensor & out, at::Tensor & save_mean, at::Tensor & save_invstd, const at::Tensor & input, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & bias, at::Tensor & running_mean, at::Tensor & running_var, bool training, double momentum, double eps) {
+    return at::_ops::_native_batch_norm_legit_out::call(input, weight, bias, running_mean, running_var, training, momentum, eps, out, save_mean, save_invstd);
+}
+// aten::_native_batch_norm_legit.out(Tensor input, Tensor? weight, Tensor? bias, Tensor(a!) running_mean, Tensor(b!) running_var, bool training, float momentum, float eps, *, Tensor(d!) out, Tensor(e!) save_mean, Tensor(f!) save_invstd) -> (Tensor(d!), Tensor(e!), Tensor(f!))
+inline ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> _native_batch_norm_legit_outf(const at::Tensor & input, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & bias, at::Tensor & running_mean, at::Tensor & running_var, bool training, double momentum, double eps, at::Tensor & out, at::Tensor & save_mean, at::Tensor & save_invstd) {
+    return at::_ops::_native_batch_norm_legit_out::call(input, weight, bias, running_mean, running_var, training, momentum, eps, out, save_mean, save_invstd);
+}
+
+// aten::_native_batch_norm_legit.no_stats(Tensor input, Tensor? weight, Tensor? bias, bool training, float momentum, float eps) -> (Tensor, Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor,at::Tensor> _native_batch_norm_legit(const at::Tensor & input, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & bias, bool training, double momentum, double eps) {
+    return at::_ops::_native_batch_norm_legit_no_stats::call(input, weight, bias, training, momentum, eps);
+}
+
+// aten::_native_batch_norm_legit.no_stats_out(Tensor input, Tensor? weight, Tensor? bias, bool training, float momentum, float eps, *, Tensor(a!) out, Tensor(b!) save_mean, Tensor(c!) save_invstd) -> (Tensor(a!), Tensor(b!), Tensor(c!))
+inline ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> _native_batch_norm_legit_out(at::Tensor & out, at::Tensor & save_mean, at::Tensor & save_invstd, const at::Tensor & input, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & bias, bool training, double momentum, double eps) {
+    return at::_ops::_native_batch_norm_legit_no_stats_out::call(input, weight, bias, training, momentum, eps, out, save_mean, save_invstd);
+}
+// aten::_native_batch_norm_legit.no_stats_out(Tensor input, Tensor? weight, Tensor? bias, bool training, float momentum, float eps, *, Tensor(a!) out, Tensor(b!) save_mean, Tensor(c!) save_invstd) -> (Tensor(a!), Tensor(b!), Tensor(c!))
+inline ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> _native_batch_norm_legit_outf(const at::Tensor & input, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & bias, bool training, double momentum, double eps, at::Tensor & out, at::Tensor & save_mean, at::Tensor & save_invstd) {
+    return at::_ops::_native_batch_norm_legit_no_stats_out::call(input, weight, bias, training, momentum, eps, out, save_mean, save_invstd);
+}
+
+// aten::_native_batch_norm_legit_functional(Tensor input, Tensor? weight, Tensor? bias, Tensor running_mean, Tensor running_var, bool training, float momentum, float eps) -> (Tensor, Tensor, Tensor, Tensor running_mean_out, Tensor running_var_out)
+inline ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor,at::Tensor> _native_batch_norm_legit_functional(const at::Tensor & input, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & bias, const at::Tensor & running_mean, const at::Tensor & running_var, bool training, double momentum, double eps) {
+    return at::_ops::_native_batch_norm_legit_functional::call(input, weight, bias, running_mean, running_var, training, momentum, eps);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_tensor_strides_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_strides {
+  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_strides")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_tensor_strides(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_strides_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_strides")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nested_tensor_strides.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/_nested_view_from_buffer_cuda_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 _nested_view_from_buffer(const at::Tensor & self, const at::Tensor & nested_size, const at::Tensor & nested_strides, const at::Tensor & offsets);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sample_dirichlet_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 & _sample_dirichlet_out(const at::Tensor & self, ::std::optional<at::Generator> generator, at::Tensor & out);
+TORCH_API at::Tensor _s_dirichlet_cpu(const at::Tensor & self, ::std::optional<at::Generator> generator=::std::nullopt);
+TORCH_API at::Tensor _s_dirichlet_cuda(const at::Tensor & self, ::std::optional<at::Generator> generator=::std::nullopt);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_autograd_multiple_dispatch_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 _test_autograd_multiple_dispatch_fullcoverage(const at::Tensor & self);
+TORCH_API at::Tensor & _test_autograd_multiple_dispatch_fullcoverage_out(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor _test_autograd_multiple_dispatch_ntonly(const at::Tensor & self, bool b);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_cpu.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/_to_cpu_ops.h>
+
+namespace at {
+
+
+// aten::_to_cpu(Tensor[] tensors) -> Tensor[]
+inline ::std::vector<at::Tensor> _to_cpu(at::TensorList tensors) {
+    return at::_ops::_to_cpu::call(tensors);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact1d_backward.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/_upsample_nearest_exact1d_backward_ops.h>
+
+namespace at {
+
+
+// aten::_upsample_nearest_exact1d_backward.grad_input(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, float? scales=None, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact1d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact1d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales, grad_input);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & _upsample_nearest_exact1d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact1d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales, grad_input);
+  }
+}
+
+// aten::_upsample_nearest_exact1d_backward.grad_input(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, float? scales=None, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact1d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales, at::Tensor & grad_input) {
+    return at::_ops::_upsample_nearest_exact1d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales, grad_input);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & _upsample_nearest_exact1d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales, at::Tensor & grad_input) {
+    return at::_ops::_upsample_nearest_exact1d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales, grad_input);
+  }
+}
+
+// aten::_upsample_nearest_exact1d_backward.grad_input(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, float? scales=None, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact1d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact1d_backward_grad_input::call(grad_output, output_size, input_size, scales, grad_input);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & _upsample_nearest_exact1d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact1d_backward_grad_input::call(grad_output, output_size, input_size, scales, grad_input);
+  }
+}
+
+// aten::_upsample_nearest_exact1d_backward.grad_input(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, float? scales=None, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact1d_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales, at::Tensor & grad_input) {
+    return at::_ops::_upsample_nearest_exact1d_backward_grad_input::call(grad_output, output_size, input_size, scales, grad_input);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & _upsample_nearest_exact1d_backward_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales, at::Tensor & grad_input) {
+    return at::_ops::_upsample_nearest_exact1d_backward_grad_input::call(grad_output, output_size, input_size, scales, grad_input);
+  }
+}
+
+// aten::_upsample_nearest_exact1d_backward(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, float? scales=None) -> Tensor
+inline at::Tensor _upsample_nearest_exact1d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact1d_backward::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor _upsample_nearest_exact1d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact1d_backward::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales);
+  }
+}
+
+// aten::_upsample_nearest_exact1d_backward(Tensor grad_output, SymInt[1] output_size, SymInt[3] input_size, float? scales=None) -> Tensor
+inline at::Tensor _upsample_nearest_exact1d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact1d_backward::call(grad_output, output_size, input_size, scales);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor _upsample_nearest_exact1d_backward(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact1d_backward::call(grad_output, output_size, input_size, scales);
+  }
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/acos_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 <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 acos(const at::Tensor & self);
+TORCH_API at::Tensor & acos_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & acos_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & acos_(at::Tensor & self);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool3d_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 adaptive_avg_pool3d_backward_grad_input {
+  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::adaptive_avg_pool3d_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "adaptive_avg_pool3d_backward.grad_input(Tensor grad_output, Tensor self, *, Tensor(a!) grad_input) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & grad_input);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & grad_input);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_meta.h

@@ -0,0 +1,27 @@
+#pragma once
+
+// @generated by torchgen/gen.py from NativeMetaFunction.h
+
+#include <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_adaptive_max_pool2d : public at::impl::MetaBase {
+    
+    
+    void meta(const at::Tensor & self, at::IntArrayRef output_size);
+};
+
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addr.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/addr_ops.h>
+
+namespace at {
+
+
+// aten::addr(Tensor self, Tensor vec1, Tensor vec2, *, Scalar beta=1, Scalar alpha=1) -> Tensor
+inline at::Tensor addr(const at::Tensor & self, const at::Tensor & vec1, const at::Tensor & vec2, const at::Scalar & beta=1, const at::Scalar & alpha=1) {
+    return at::_ops::addr::call(self, vec1, vec2, beta, alpha);
+}
+
+// aten::addr.out(Tensor self, Tensor vec1, Tensor vec2, *, Scalar beta=1, Scalar alpha=1, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & addr_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & vec1, const at::Tensor & vec2, const at::Scalar & beta=1, const at::Scalar & alpha=1) {
+    return at::_ops::addr_out::call(self, vec1, vec2, beta, alpha, out);
+}
+// aten::addr.out(Tensor self, Tensor vec1, Tensor vec2, *, Scalar beta=1, Scalar alpha=1, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & addr_outf(const at::Tensor & self, const at::Tensor & vec1, const at::Tensor & vec2, const at::Scalar & beta, const at::Scalar & alpha, at::Tensor & out) {
+    return at::_ops::addr_out::call(self, vec1, vec2, beta, alpha, out);
+}
+
+}

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

@@ -0,0 +1,24 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 affine_grid_generator_backward(const at::Tensor & grad, at::IntArrayRef size, bool align_corners);
+TORCH_API at::Tensor affine_grid_generator_backward_symint(const at::Tensor & grad, c10::SymIntArrayRef size, bool align_corners);
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

@@ -0,0 +1,50 @@
+#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 arctan {
+  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::arctan")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "arctan(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 arctan_ {
+  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::arctan_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "arctan_(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 arctan_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::arctan")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "arctan.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/asinh_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 <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 asinh(const at::Tensor & self);
+TORCH_API at::Tensor & asinh_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & asinh_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & asinh_(at::Tensor & self);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/avg_pool2d_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 avg_pool2d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, ::std::optional<int64_t> divisor_override);
+TORCH_API at::Tensor & avg_pool2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, ::std::optional<int64_t> divisor_override);
+TORCH_API at::Tensor & avg_pool2d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, bool ceil_mode, bool count_include_pad, ::std::optional<int64_t> divisor_override, at::Tensor & grad_input);
+
+} // namespace cuda
+} // namespace at

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

@@ -0,0 +1,61 @@
+#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 bartlett_window {
+  using schema = at::Tensor (int64_t, ::std::optional<at::ScalarType>, ::std::optional<at::Layout>, ::std::optional<at::Device>, ::std::optional<bool>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::bartlett_window")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bartlett_window(int window_length, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor")
+  static at::Tensor call(int64_t window_length, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, int64_t window_length, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory);
+};
+
+struct TORCH_API bartlett_window_periodic {
+  using schema = at::Tensor (int64_t, bool, ::std::optional<at::ScalarType>, ::std::optional<at::Layout>, ::std::optional<at::Device>, ::std::optional<bool>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::bartlett_window")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "periodic")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bartlett_window.periodic(int window_length, bool periodic, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor")
+  static at::Tensor call(int64_t window_length, bool periodic, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, int64_t window_length, bool periodic, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory);
+};
+
+struct TORCH_API bartlett_window_out {
+  using schema = at::Tensor & (int64_t, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::bartlett_window")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bartlett_window.out(int window_length, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(int64_t window_length, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, int64_t window_length, at::Tensor & out);
+};
+
+struct TORCH_API bartlett_window_periodic_out {
+  using schema = at::Tensor & (int64_t, bool, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::bartlett_window")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "periodic_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "bartlett_window.periodic_out(int window_length, bool periodic, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(int64_t window_length, bool periodic, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, int64_t window_length, bool periodic, at::Tensor & out);
+};
+
+}} // namespace at::_ops

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

@@ -0,0 +1,28 @@
+#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/bitwise_left_shift_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_bitwise_left_shift_out : public at::meta::structured_bitwise_left_shift_Tensor {
+void impl(const at::Tensor & self, const at::Tensor & other, const at::Tensor & out);
+};
+TORCH_API at::Tensor bitwise_left_shift(const at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor & bitwise_left_shift_out(const at::Tensor & self, const at::Scalar & other, at::Tensor & out);
+TORCH_API at::Tensor & bitwise_left_shift_(at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor bitwise_left_shift(const at::Scalar & self, const at::Tensor & other);
+TORCH_API at::Tensor & bitwise_left_shift_Scalar_Tensor_out(const at::Scalar & self, const at::Tensor & other, at::Tensor & out);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,26 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 bitwise_not(const at::Tensor & self);
+TORCH_API at::Tensor & bitwise_not_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & bitwise_not_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & bitwise_not_(at::Tensor & self);
+
+} // namespace meta
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/clamp_max_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/clamp_max_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_clamp_max_out : public at::meta::structured_clamp_max {
+void impl(const at::Tensor & self, const at::Scalar & max, const at::Tensor & out);
+};
+struct TORCH_API structured_clamp_max_Tensor_out : public at::meta::structured_clamp_max_Tensor {
+void impl(const at::Tensor & self, const at::Tensor & max, const at::Tensor & out);
+};
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,32 @@
+#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_clamp_min : public TensorIteratorBase {
+    
+    
+    void meta(const at::Tensor & self, const at::Scalar & min);
+};
+struct TORCH_API structured_clamp_min_Tensor : public TensorIteratorBase {
+    
+    
+    void meta(const at::Tensor & self, const at::Tensor & min);
+};
+
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_backward.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/cudnn_batch_norm_backward_ops.h>
+
+namespace at {
+
+
+// aten::cudnn_batch_norm_backward(Tensor input, Tensor grad_output, Tensor weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_var, float epsilon, Tensor reserveSpace) -> (Tensor, Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor,at::Tensor> cudnn_batch_norm_backward(const at::Tensor & input, const at::Tensor & grad_output, const 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, double epsilon, const at::Tensor & reserveSpace) {
+    return at::_ops::cudnn_batch_norm_backward::call(input, grad_output, weight, running_mean, running_var, save_mean, save_var, epsilon, reserveSpace);
+}
+
+// aten::cudnn_batch_norm_backward.out(Tensor input, Tensor grad_output, Tensor weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_var, float epsilon, Tensor reserveSpace, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!))
+inline ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> cudnn_batch_norm_backward_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, const at::Tensor & input, const at::Tensor & grad_output, const 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, double epsilon, const at::Tensor & reserveSpace) {
+    return at::_ops::cudnn_batch_norm_backward_out::call(input, grad_output, weight, running_mean, running_var, save_mean, save_var, epsilon, reserveSpace, out0, out1, out2);
+}
+// aten::cudnn_batch_norm_backward.out(Tensor input, Tensor grad_output, Tensor weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_var, float epsilon, Tensor reserveSpace, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!))
+inline ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> cudnn_batch_norm_backward_outf(const at::Tensor & input, const at::Tensor & grad_output, const 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, double epsilon, const at::Tensor & reserveSpace, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2) {
+    return at::_ops::cudnn_batch_norm_backward_out::call(input, grad_output, weight, running_mean, running_var, save_mean, save_var, epsilon, reserveSpace, out0, out1, out2);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_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 <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 &> cudnn_batch_norm_backward_out(at::Tensor & out0, at::Tensor & out1, at::Tensor & out2, const at::Tensor & input, const at::Tensor & grad_output, const 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, double epsilon, const at::Tensor & reserveSpace);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> cudnn_batch_norm_backward_outf(const at::Tensor & input, const at::Tensor & grad_output, const 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, double epsilon, const at::Tensor & reserveSpace, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/det_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 det(const at::Tensor & self);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach_copy_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 detach_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::detach_copy")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "detach_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 detach_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::detach_copy")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "detach_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

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

@@ -0,0 +1,35 @@
+#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/dsplit_ops.h>
+
+namespace at {
+
+
+// aten::dsplit.int(Tensor(a -> *) self, int sections) -> Tensor(a)[]
+inline ::std::vector<at::Tensor> dsplit(const at::Tensor & self, int64_t sections) {
+    return at::_ops::dsplit_int::call(self, sections);
+}
+
+// aten::dsplit.array(Tensor(a -> *) self, int[] indices) -> Tensor(a)[]
+inline ::std::vector<at::Tensor> dsplit(const at::Tensor & self, at::IntArrayRef indices) {
+    return at::_ops::dsplit_array::call(self, indices);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/empty_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>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor empty_names(at::IntArrayRef size, ::std::optional<at::DimnameList> names, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor & empty_names_out(at::IntArrayRef size, ::std::optional<at::DimnameList> names, ::std::optional<at::MemoryFormat> memory_format, at::Tensor & out);
+TORCH_API at::Tensor & empty_out(at::IntArrayRef size, ::std::optional<at::MemoryFormat> memory_format, at::Tensor & out);
+TORCH_API at::Tensor empty_cpu(at::IntArrayRef size, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_cuda(at::IntArrayRef size, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_sparse(at::IntArrayRef size, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_sparse_compressed(at::IntArrayRef size, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_sparse_compressed_symint(c10::SymIntArrayRef size, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_meta_symint(c10::SymIntArrayRef size, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_sparse_symint(c10::SymIntArrayRef size, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_mkldnn(at::IntArrayRef size, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_unknown_quantized(at::IntArrayRef size, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/erfc_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 <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 erfc(const at::Tensor & self);
+TORCH_API at::Tensor & erfc_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & erfc_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & erfc_(at::Tensor & self);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_linear_fp16_weight_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(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/fbgemm_linear_fp16_weight_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 fbgemm_linear_fp16_weight(const at::Tensor & input, const at::Tensor & packed_weight, const at::Tensor & bias);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fbgemm_pack_gemm_matrix_fp16_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_pack_gemm_matrix_fp16(const at::Tensor & input);
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/floor_divide_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 <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 & floor_divide_(at::Tensor & self, const at::Tensor & other);
+
+} // namespace meta
+} // namespace at

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

@@ -0,0 +1,50 @@
+#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 floor {
+  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::floor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "floor(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 floor_ {
+  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::floor_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "floor_(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 floor_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::floor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "floor.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/greater.h

@@ -0,0 +1,53 @@
+#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/greater_ops.h>
+
+namespace at {
+
+
+// aten::greater.Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & greater_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other) {
+    return at::_ops::greater_Scalar_out::call(self, other, out);
+}
+// aten::greater.Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & greater_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out) {
+    return at::_ops::greater_Scalar_out::call(self, other, out);
+}
+
+// aten::greater.Scalar(Tensor self, Scalar other) -> Tensor
+inline at::Tensor greater(const at::Tensor & self, const at::Scalar & other) {
+    return at::_ops::greater_Scalar::call(self, other);
+}
+
+// aten::greater.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & greater_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::greater_Tensor_out::call(self, other, out);
+}
+// aten::greater.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & greater_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) {
+    return at::_ops::greater_Tensor_out::call(self, other, out);
+}
+
+// aten::greater.Tensor(Tensor self, Tensor other) -> Tensor
+inline at::Tensor greater(const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::greater_Tensor::call(self, other);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_2d_cuda_dispatch.h

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 grid_sampler_2d(const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_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 grid_sampler(const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners);
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardsigmoid_backward_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 hardsigmoid_backward_grad_input {
+  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::hardsigmoid_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "grad_input")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "hardsigmoid_backward.grad_input(Tensor grad_output, Tensor self, *, Tensor(a!) grad_input) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & grad_input);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & grad_input);
+};
+
+struct TORCH_API hardsigmoid_backward {
+  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::hardsigmoid_backward")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "hardsigmoid_backward(Tensor grad_output, Tensor self) -> Tensor")
+  static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & self);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & self);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/igammac_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 <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 igammac(const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & igammac_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other);
+TORCH_API at::Tensor & igammac_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out);
+TORCH_API at::Tensor & igammac_(at::Tensor & self, const at::Tensor & other);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/index_reduce.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/index_reduce_ops.h>
+
+namespace at {
+
+
+// aten::index_reduce.out(Tensor self, int dim, Tensor index, Tensor source, str reduce, *, bool include_self=True, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & index_reduce_out(at::Tensor & out, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, c10::string_view reduce, bool include_self=true) {
+    return at::_ops::index_reduce_out::call(self, dim, index, source, reduce, include_self, out);
+}
+// aten::index_reduce.out(Tensor self, int dim, Tensor index, Tensor source, str reduce, *, bool include_self=True, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & index_reduce_outf(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, c10::string_view reduce, bool include_self, at::Tensor & out) {
+    return at::_ops::index_reduce_out::call(self, dim, index, source, reduce, include_self, out);
+}
+
+// aten::index_reduce(Tensor self, int dim, Tensor index, Tensor source, str reduce, *, bool include_self=True) -> Tensor
+inline at::Tensor index_reduce(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & source, c10::string_view reduce, bool include_self=true) {
+    return at::_ops::index_reduce::call(self, dim, index, source, reduce, include_self);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_floating_point_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 is_floating_point {
+  using schema = bool (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::is_floating_point")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "is_floating_point(Tensor self) -> bool")
+  static bool call(const at::Tensor & self);
+  static bool redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self);
+};
+
+}} // namespace at::_ops