Add files using upload-large-folder tool

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_aminmax_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::tuple<at::Tensor &,at::Tensor &> _aminmax_out(at::Tensor & out0, at::Tensor & out1, const at::Tensor & self);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> _aminmax_outf(const at::Tensor & self, at::Tensor & out0, at::Tensor & out1);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> _aminmax_out(at::Tensor & out0, at::Tensor & out1, const at::Tensor & self, int64_t dim, bool keepdim=false);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> _aminmax_outf(const at::Tensor & self, int64_t dim, bool keepdim, at::Tensor & out0, at::Tensor & out1);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_efficient_attention_forward_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::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor,c10::SymInt,c10::SymInt> _efficient_attention_forward(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const ::std::optional<at::Tensor> & bias, const ::std::optional<at::Tensor> & cu_seqlens_q, const ::std::optional<at::Tensor> & cu_seqlens_k, ::std::optional<int64_t> max_seqlen_q, ::std::optional<int64_t> max_seqlen_k, double dropout_p, int64_t custom_mask_type, bool compute_log_sumexp=false, ::std::optional<double> scale=::std::nullopt, const ::std::optional<at::Tensor> & seqlen_k={}, ::std::optional<int64_t> window_size=::std::nullopt);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor,c10::SymInt,c10::SymInt> _efficient_attention_forward_symint(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const ::std::optional<at::Tensor> & bias, const ::std::optional<at::Tensor> & cu_seqlens_q, const ::std::optional<at::Tensor> & cu_seqlens_k, ::std::optional<c10::SymInt> max_seqlen_q, ::std::optional<c10::SymInt> max_seqlen_k, double dropout_p, int64_t custom_mask_type, bool compute_log_sumexp=false, ::std::optional<double> scale=::std::nullopt, const ::std::optional<at::Tensor> & seqlen_k={}, ::std::optional<int64_t> window_size=::std::nullopt);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_ceil.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/_foreach_ceil_ops.h>
+
+namespace at {
+
+
+// aten::_foreach_ceil(Tensor[] self) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_ceil(at::TensorList self) {
+    return at::_ops::_foreach_ceil::call(self);
+}
+
+// aten::_foreach_ceil_(Tensor(a!)[] self) -> ()
+inline void _foreach_ceil_(at::TensorList self) {
+    return at::_ops::_foreach_ceil_::call(self);
+}
+
+// aten::_foreach_ceil.out(Tensor[] self, *, Tensor(a!)[] out) -> ()
+inline void _foreach_ceil_out(at::TensorList out, at::TensorList self) {
+    return at::_ops::_foreach_ceil_out::call(self, out);
+}
+// aten::_foreach_ceil.out(Tensor[] self, *, Tensor(a!)[] out) -> ()
+inline void _foreach_ceil_outf(at::TensorList self, at::TensorList out) {
+    return at::_ops::_foreach_ceil_out::call(self, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_norm_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 ::std::vector<at::Tensor> _foreach_norm(at::TensorList self, const at::Scalar & ord=2, ::std::optional<at::ScalarType> dtype=::std::nullopt);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fused_adamw_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 void _fused_adamw_(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale={}, const ::std::optional<at::Tensor> & found_inf={});
+TORCH_API void _fused_adamw_(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale={}, const ::std::optional<at::Tensor> & found_inf={});
+
+} // namespace cuda
+} // namespace at

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

@@ -0,0 +1,83 @@
+#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 _fused_adamw_ {
+  using schema = void (at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, double, double, double, double, double, bool, bool, const ::std::optional<at::Tensor> &, const ::std::optional<at::Tensor> &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_fused_adamw_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_fused_adamw_(Tensor(a!)[] self, Tensor(b!)[] grads, Tensor(c!)[] exp_avgs, Tensor(d!)[] exp_avg_sqs, Tensor(e!)[] max_exp_avg_sqs, Tensor[] state_steps, *, float lr, float beta1, float beta2, float weight_decay, float eps, bool amsgrad, bool maximize, Tensor? grad_scale=None, Tensor? found_inf=None) -> ()")
+  static void call(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf);
+  static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf);
+};
+
+struct TORCH_API _fused_adamw__tensor_lr {
+  using schema = void (at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, const at::Tensor &, double, double, double, double, bool, bool, const ::std::optional<at::Tensor> &, const ::std::optional<at::Tensor> &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_fused_adamw_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "tensor_lr")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_fused_adamw_.tensor_lr(Tensor(a!)[] self, Tensor(b!)[] grads, Tensor(c!)[] exp_avgs, Tensor(d!)[] exp_avg_sqs, Tensor(e!)[] max_exp_avg_sqs, Tensor[] state_steps, *, Tensor lr, float beta1, float beta2, float weight_decay, float eps, bool amsgrad, bool maximize, Tensor? grad_scale=None, Tensor? found_inf=None) -> ()")
+  static void call(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf);
+  static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf);
+};
+
+struct TORCH_API _fused_adamw_out {
+  using schema = void (at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, double, double, double, double, double, bool, bool, const ::std::optional<at::Tensor> &, const ::std::optional<at::Tensor> &, at::TensorList);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_fused_adamw")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_fused_adamw.out(Tensor[] self, Tensor(b!)[] grads, Tensor(c!)[] exp_avgs, Tensor(d!)[] exp_avg_sqs, Tensor(e!)[] max_exp_avg_sqs, Tensor[] state_steps, *, float lr, float beta1, float beta2, float weight_decay, float eps, bool amsgrad, bool maximize, Tensor? grad_scale=None, Tensor? found_inf=None, Tensor(a!)[] out) -> ()")
+  static void call(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf, at::TensorList out);
+  static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf, at::TensorList out);
+};
+
+struct TORCH_API _fused_adamw {
+  using schema = ::std::tuple<::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>> (at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, double, double, double, double, double, bool, bool, const ::std::optional<at::Tensor> &, const ::std::optional<at::Tensor> &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_fused_adamw")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_fused_adamw(Tensor[] self, Tensor[] grads, Tensor[] exp_avgs, Tensor[] exp_avg_sqs, Tensor[] max_exp_avg_sqs, Tensor[] state_steps, *, float lr, float beta1, float beta2, float weight_decay, float eps, bool amsgrad, bool maximize, Tensor? grad_scale=None, Tensor? found_inf=None) -> (Tensor[] self_out, Tensor[] grads_out, Tensor[] exp_avgs_out, Tensor[] exp_avg_sqs_out, Tensor[] max_exp_avg_sqs_out)")
+  static ::std::tuple<::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>> call(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf);
+  static ::std::tuple<::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>> redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, double lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf);
+};
+
+struct TORCH_API _fused_adamw_tensor_lr_out {
+  using schema = void (at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, const at::Tensor &, double, double, double, double, bool, bool, const ::std::optional<at::Tensor> &, const ::std::optional<at::Tensor> &, at::TensorList);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_fused_adamw")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "tensor_lr_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_fused_adamw.tensor_lr_out(Tensor[] self, Tensor(b!)[] grads, Tensor(c!)[] exp_avgs, Tensor(d!)[] exp_avg_sqs, Tensor(e!)[] max_exp_avg_sqs, Tensor[] state_steps, *, Tensor lr, float beta1, float beta2, float weight_decay, float eps, bool amsgrad, bool maximize, Tensor? grad_scale=None, Tensor? found_inf=None, Tensor(a!)[] out) -> ()")
+  static void call(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf, at::TensorList out);
+  static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf, at::TensorList out);
+};
+
+struct TORCH_API _fused_adamw_tensor_lr {
+  using schema = ::std::tuple<::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>> (at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, at::TensorList, const at::Tensor &, double, double, double, double, bool, bool, const ::std::optional<at::Tensor> &, const ::std::optional<at::Tensor> &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_fused_adamw")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "tensor_lr")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_fused_adamw.tensor_lr(Tensor[] self, Tensor[] grads, Tensor[] exp_avgs, Tensor[] exp_avg_sqs, Tensor[] max_exp_avg_sqs, Tensor[] state_steps, *, Tensor lr, float beta1, float beta2, float weight_decay, float eps, bool amsgrad, bool maximize, Tensor? grad_scale=None, Tensor? found_inf=None) -> (Tensor[] self_out, Tensor[] grads_out, Tensor[] exp_avgs_out, Tensor[] exp_avg_sqs_out, Tensor[] max_exp_avg_sqs_out)")
+  static ::std::tuple<::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>> call(at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf);
+  static ::std::tuple<::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>,::std::vector<at::Tensor>> redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList grads, at::TensorList exp_avgs, at::TensorList exp_avg_sqs, at::TensorList max_exp_avg_sqs, at::TensorList state_steps, const at::Tensor & lr, double beta1, double beta2, double weight_decay, double eps, bool amsgrad, bool maximize, const ::std::optional<at::Tensor> & grad_scale, const ::std::optional<at::Tensor> & found_inf);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_int_mm.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/_int_mm_ops.h>
+
+namespace at {
+
+
+// aten::_int_mm(Tensor self, Tensor mat2) -> Tensor
+inline at::Tensor _int_mm(const at::Tensor & self, const at::Tensor & mat2) {
+    return at::_ops::_int_mm::call(self, mat2);
+}
+
+// aten::_int_mm.out(Tensor self, Tensor mat2, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _int_mm_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & mat2) {
+    return at::_ops::_int_mm_out::call(self, mat2, out);
+}
+// aten::_int_mm.out(Tensor self, Tensor mat2, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _int_mm_outf(const at::Tensor & self, const at::Tensor & mat2, at::Tensor & out) {
+    return at::_ops::_int_mm_out::call(self, mat2, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_check_errors_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 _linalg_check_errors {
+  using schema = void (const at::Tensor &, c10::string_view, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_linalg_check_errors")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_linalg_check_errors(Tensor info, str api_name, *, bool is_matrix) -> ()")
+  static void call(const at::Tensor & info, c10::string_view api_name, bool is_matrix);
+  static void redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & info, c10::string_view api_name, bool is_matrix);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_det_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__linalg_det : public at::impl::MetaBase {
+    
+    
+    void meta(const at::Tensor & A);
+};
+
+} // namespace native
+} // namespace at

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_view_from_buffer.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/_nested_view_from_buffer_ops.h>
+
+namespace at {
+
+
+// aten::_nested_view_from_buffer(Tensor(a) self, Tensor nested_size, Tensor nested_strides, Tensor offsets) -> Tensor(a)
+inline at::Tensor _nested_view_from_buffer(const at::Tensor & self, const at::Tensor & nested_size, const at::Tensor & nested_strides, const at::Tensor & offsets) {
+    return at::_ops::_nested_view_from_buffer::call(self, nested_size, nested_strides, offsets);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nnpack_spatial_convolution_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 _nnpack_spatial_convolution {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const ::std::optional<at::Tensor> &, c10::SymIntArrayRef, c10::SymIntArrayRef);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_nnpack_spatial_convolution")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nnpack_spatial_convolution(Tensor input, Tensor weight, Tensor? bias, SymInt[2] padding, SymInt[2] stride=1) -> Tensor")
+  static at::Tensor call(const at::Tensor & input, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride);
+};
+
+struct TORCH_API _nnpack_spatial_convolution_out {
+  using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const ::std::optional<at::Tensor> &, c10::SymIntArrayRef, 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::_nnpack_spatial_convolution")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_nnpack_spatial_convolution.out(Tensor input, Tensor weight, Tensor? bias, SymInt[2] padding, SymInt[2] stride=1, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & input, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, at::Tensor & out);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_pin_memory_compositeexplicitautograd_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 compositeexplicitautograd {
+
+TORCH_API at::Tensor _pin_memory(const at::Tensor & self, ::std::optional<at::Device> device=::std::nullopt);
+TORCH_API at::Tensor & _pin_memory_out(at::Tensor & out, const at::Tensor & self, ::std::optional<at::Device> device=::std::nullopt);
+TORCH_API at::Tensor & _pin_memory_outf(const at::Tensor & self, ::std::optional<at::Device> device, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_prelu_kernel_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 ::std::tuple<at::Tensor,at::Tensor> _prelu_kernel_backward(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> mkldnn_prelu_backward(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & weight);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_remove_batch_dim_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 _remove_batch_dim(const at::Tensor & self, int64_t level, int64_t batch_size, int64_t out_dim);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,24 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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::tuple<at::Tensor,at::Tensor,at::Tensor> _scaled_dot_product_cudnn_attention_backward(const at::Tensor & grad_out, const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const at::Tensor & out, const at::Tensor & logsumexp, const at::Tensor & philox_seed, const at::Tensor & philox_offset, const at::Tensor & attn_bias, const at::Tensor & cum_seq_q, const at::Tensor & cum_seq_k, int64_t max_q, int64_t max_k, double dropout_p, bool is_causal, ::std::optional<double> scale=::std::nullopt);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor> _scaled_dot_product_cudnn_attention_backward_symint(const at::Tensor & grad_out, const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const at::Tensor & out, const at::Tensor & logsumexp, const at::Tensor & philox_seed, const at::Tensor & philox_offset, const at::Tensor & attn_bias, const at::Tensor & cum_seq_q, const at::Tensor & cum_seq_k, c10::SymInt max_q, c10::SymInt max_k, double dropout_p, bool is_causal, ::std::optional<double> scale=::std::nullopt);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data_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__softmax_backward_data : public at::impl::MetaBase {
+    
+    
+    void meta(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype);
+};
+
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_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 _softmax {
+  using schema = at::Tensor (const at::Tensor &, int64_t, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_softmax")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_softmax(Tensor self, int dim, bool half_to_float) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, int64_t dim, bool half_to_float);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, bool half_to_float);
+};
+
+struct TORCH_API _softmax_out {
+  using schema = at::Tensor & (const 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::_softmax")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_softmax.out(Tensor self, int dim, bool half_to_float, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, int64_t dim, bool half_to_float, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, bool half_to_float, at::Tensor & out);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_serialization_subcmul_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 _test_serialization_subcmul {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Scalar &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_test_serialization_subcmul")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_test_serialization_subcmul(Tensor self, Tensor other, Scalar alpha=1) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, const at::Scalar & alpha);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact1d_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 _upsample_nearest_exact1d_vec {
+  using schema = at::Tensor (const at::Tensor &, at::OptionalSymIntArrayRef, ::std::optional<at::ArrayRef<double>>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_upsample_nearest_exact1d")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "vec")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_upsample_nearest_exact1d.vec(Tensor input, SymInt[]? output_size, float[]? scale_factors) -> Tensor")
+  static at::Tensor call(const at::Tensor & input, at::OptionalSymIntArrayRef output_size, ::std::optional<at::ArrayRef<double>> scale_factors);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & input, at::OptionalSymIntArrayRef output_size, ::std::optional<at::ArrayRef<double>> scale_factors);
+};
+
+struct TORCH_API _upsample_nearest_exact1d_out {
+  using schema = at::Tensor & (const at::Tensor &, c10::SymIntArrayRef, ::std::optional<double>, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_upsample_nearest_exact1d")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_upsample_nearest_exact1d.out(Tensor self, SymInt[1] output_size, float? scales=None, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales, at::Tensor & out);
+};
+
+struct TORCH_API _upsample_nearest_exact1d {
+  using schema = at::Tensor (const at::Tensor &, c10::SymIntArrayRef, ::std::optional<double>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_upsample_nearest_exact1d")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_upsample_nearest_exact1d(Tensor self, SymInt[1] output_size, float? scales=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact2d.h

@@ -0,0 +1,113 @@
+#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_exact2d_ops.h>
+
+namespace at {
+
+
+// aten::_upsample_nearest_exact2d.vec(Tensor input, SymInt[]? output_size, float[]? scale_factors) -> Tensor
+inline at::Tensor _upsample_nearest_exact2d(const at::Tensor & input, at::OptionalIntArrayRef output_size, ::std::optional<at::ArrayRef<double>> scale_factors) {
+    return at::_ops::_upsample_nearest_exact2d_vec::call(input, output_size.has_value() ? ::std::make_optional(c10::fromIntArrayRefSlow(*output_size)) : ::std::nullopt, scale_factors);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor _upsample_nearest_exact2d(const at::Tensor & input, at::OptionalIntArrayRef output_size, ::std::optional<at::ArrayRef<double>> scale_factors) {
+    return at::_ops::_upsample_nearest_exact2d_vec::call(input, output_size.has_value() ? ::std::make_optional(c10::fromIntArrayRefSlow(*output_size)) : ::std::nullopt, scale_factors);
+  }
+}
+
+// aten::_upsample_nearest_exact2d.vec(Tensor input, SymInt[]? output_size, float[]? scale_factors) -> Tensor
+inline at::Tensor _upsample_nearest_exact2d_symint(const at::Tensor & input, at::OptionalSymIntArrayRef output_size, ::std::optional<at::ArrayRef<double>> scale_factors) {
+    return at::_ops::_upsample_nearest_exact2d_vec::call(input, output_size, scale_factors);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor _upsample_nearest_exact2d(const at::Tensor & input, at::OptionalSymIntArrayRef output_size, ::std::optional<at::ArrayRef<double>> scale_factors) {
+    return at::_ops::_upsample_nearest_exact2d_vec::call(input, output_size, scale_factors);
+  }
+}
+
+// aten::_upsample_nearest_exact2d.out(Tensor self, SymInt[2] output_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact2d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_out::call(self, c10::fromIntArrayRefSlow(output_size), scales_h, scales_w, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & _upsample_nearest_exact2d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef output_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_out::call(self, c10::fromIntArrayRefSlow(output_size), scales_h, scales_w, out);
+  }
+}
+
+// aten::_upsample_nearest_exact2d.out(Tensor self, SymInt[2] output_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact2d_outf(const at::Tensor & self, at::IntArrayRef output_size, ::std::optional<double> scales_h, ::std::optional<double> scales_w, at::Tensor & out) {
+    return at::_ops::_upsample_nearest_exact2d_out::call(self, c10::fromIntArrayRefSlow(output_size), scales_h, scales_w, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & _upsample_nearest_exact2d_outf(const at::Tensor & self, at::IntArrayRef output_size, ::std::optional<double> scales_h, ::std::optional<double> scales_w, at::Tensor & out) {
+    return at::_ops::_upsample_nearest_exact2d_out::call(self, c10::fromIntArrayRefSlow(output_size), scales_h, scales_w, out);
+  }
+}
+
+// aten::_upsample_nearest_exact2d.out(Tensor self, SymInt[2] output_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact2d_symint_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_out::call(self, output_size, scales_h, scales_w, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & _upsample_nearest_exact2d_out(at::Tensor & out, const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_out::call(self, output_size, scales_h, scales_w, out);
+  }
+}
+
+// aten::_upsample_nearest_exact2d.out(Tensor self, SymInt[2] output_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact2d_symint_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales_h, ::std::optional<double> scales_w, at::Tensor & out) {
+    return at::_ops::_upsample_nearest_exact2d_out::call(self, output_size, scales_h, scales_w, out);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & _upsample_nearest_exact2d_outf(const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales_h, ::std::optional<double> scales_w, at::Tensor & out) {
+    return at::_ops::_upsample_nearest_exact2d_out::call(self, output_size, scales_h, scales_w, out);
+  }
+}
+
+// aten::_upsample_nearest_exact2d(Tensor self, SymInt[2] output_size, float? scales_h=None, float? scales_w=None) -> Tensor
+inline at::Tensor _upsample_nearest_exact2d(const at::Tensor & self, at::IntArrayRef output_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d::call(self, c10::fromIntArrayRefSlow(output_size), scales_h, scales_w);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor _upsample_nearest_exact2d(const at::Tensor & self, at::IntArrayRef output_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d::call(self, c10::fromIntArrayRefSlow(output_size), scales_h, scales_w);
+  }
+}
+
+// aten::_upsample_nearest_exact2d(Tensor self, SymInt[2] output_size, float? scales_h=None, float? scales_w=None) -> Tensor
+inline at::Tensor _upsample_nearest_exact2d_symint(const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d::call(self, output_size, scales_h, scales_w);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor _upsample_nearest_exact2d(const at::Tensor & self, c10::SymIntArrayRef output_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d::call(self, output_size, scales_h, scales_w);
+  }
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_validate_sparse_csr_tensor_args_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 _validate_sparse_csr_tensor_args(const at::Tensor & crow_indices, const at::Tensor & col_indices, const at::Tensor & values, at::IntArrayRef size);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool1d_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 adaptive_avg_pool1d(const at::Tensor & self, at::IntArrayRef output_size);
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool3d_backward_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/adaptive_max_pool3d_backward_meta.h>
+
+namespace at {
+namespace native {
+struct TORCH_API structured_adaptive_max_pool3d_backward_out_cpu : public at::meta::structured_adaptive_max_pool3d_backward {
+void impl(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & indices, const at::Tensor & grad_input);
+};
+struct TORCH_API structured_adaptive_max_pool3d_backward_out_cuda : public at::meta::structured_adaptive_max_pool3d_backward {
+void impl(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & indices, const at::Tensor & grad_input);
+};
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 amax(const at::Tensor & self, at::IntArrayRef dim={}, bool keepdim=false);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/any_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 any(const at::Tensor & self, int64_t dim, bool keepdim=false);
+TORCH_API at::Tensor & any_out(at::Tensor & out, const at::Tensor & self, int64_t dim, bool keepdim=false);
+TORCH_API at::Tensor & any_outf(const at::Tensor & self, int64_t dim, bool keepdim, at::Tensor & out);
+TORCH_API at::Tensor any(const at::Tensor & self, at::OptionalIntArrayRef dim, bool keepdim=false);
+TORCH_API at::Tensor & any_out(at::Tensor & out, const at::Tensor & self, at::OptionalIntArrayRef dim, bool keepdim=false);
+TORCH_API at::Tensor & any_outf(const at::Tensor & self, at::OptionalIntArrayRef dim, bool keepdim, at::Tensor & out);
+TORCH_API at::Tensor any(const at::Tensor & self);
+TORCH_API at::Tensor & any_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & any_outf(const at::Tensor & self, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/arccosh_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 arccosh(const at::Tensor & self);
+TORCH_API at::Tensor & arccosh_out(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & arccosh_(at::Tensor & self);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/arcsinh_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 arcsinh(const at::Tensor & self);
+TORCH_API at::Tensor & arcsinh_out(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & arcsinh_(at::Tensor & self);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,24 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 asinh(const at::Tensor & self);
+TORCH_API at::Tensor & asinh_(at::Tensor & self);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/atleast_2d_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 atleast_2d(const at::Tensor & self);
+TORCH_API ::std::vector<at::Tensor> atleast_2d(at::TensorList tensors);
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/convolution_backward_overrideable_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 convolution_backward_overrideable {
+  using schema = ::std::tuple<at::Tensor,at::Tensor,at::Tensor> (const at::Tensor &, const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymIntArrayRef, bool, c10::SymIntArrayRef, c10::SymInt, ::std::array<bool,3>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::convolution_backward_overrideable")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "convolution_backward_overrideable(Tensor grad_output, Tensor input, Tensor weight, SymInt[] stride, SymInt[] padding, SymInt[] dilation, bool transposed, SymInt[] output_padding, SymInt groups, bool[3] output_mask) -> (Tensor grad_input, Tensor grad_weight, Tensor grad_bias)")
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor> call(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & weight, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, bool transposed, c10::SymIntArrayRef output_padding, c10::SymInt groups, ::std::array<bool,3> output_mask);
+  static ::std::tuple<at::Tensor,at::Tensor,at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & weight, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, bool transposed, c10::SymIntArrayRef output_padding, c10::SymInt groups, ::std::array<bool,3> output_mask);
+};
+
+struct TORCH_API convolution_backward_overrideable_out {
+  using schema = ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> (const at::Tensor &, const at::Tensor &, const at::Tensor &, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymIntArrayRef, bool, c10::SymIntArrayRef, c10::SymInt, ::std::array<bool,3>, at::Tensor &, at::Tensor &, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::convolution_backward_overrideable")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "convolution_backward_overrideable.out(Tensor grad_output, Tensor input, Tensor weight, SymInt[] stride, SymInt[] padding, SymInt[] dilation, bool transposed, SymInt[] output_padding, SymInt groups, bool[3] output_mask, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!))")
+  static ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> call(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & weight, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, bool transposed, c10::SymIntArrayRef output_padding, c10::SymInt groups, ::std::array<bool,3> output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2);
+  static ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & weight, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, bool transposed, c10::SymIntArrayRef output_padding, c10::SymInt groups, ::std::array<bool,3> output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_batch_norm_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 ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor> cudnn_batch_norm(const at::Tensor & input, const at::Tensor & weight, const ::std::optional<at::Tensor> & bias, const ::std::optional<at::Tensor> & running_mean, const ::std::optional<at::Tensor> & running_var, bool training, double exponential_average_factor, double epsilon);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cudnn_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 & cudnn_convolution_transpose_out_symint(const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef padding, c10::SymIntArrayRef output_padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, bool allow_tf32, at::Tensor & out);
+TORCH_API at::Tensor cudnn_convolution_transpose(const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef padding, at::IntArrayRef output_padding, at::IntArrayRef stride, at::IntArrayRef dilation, int64_t groups, bool benchmark, bool deterministic, bool allow_tf32);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/dstack_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 dstack(at::TensorList tensors);
+TORCH_API at::Tensor & dstack_out(at::TensorList tensors, at::Tensor & out);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/exp2_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 exp2(const at::Tensor & self);
+TORCH_API at::Tensor & exp2_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & exp2_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & exp2_(at::Tensor & self);
+
+} // namespace meta
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fft_hfftn_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 fft_hfftn {
+  using schema = at::Tensor (const at::Tensor &, at::OptionalSymIntArrayRef, at::OptionalIntArrayRef, ::std::optional<c10::string_view>);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::fft_hfftn")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fft_hfftn(Tensor self, SymInt[1]? s=None, int[1]? dim=None, str? norm=None) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, at::OptionalSymIntArrayRef s, at::OptionalIntArrayRef dim, ::std::optional<c10::string_view> norm);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::OptionalSymIntArrayRef s, at::OptionalIntArrayRef dim, ::std::optional<c10::string_view> norm);
+};
+
+struct TORCH_API fft_hfftn_out {
+  using schema = const at::Tensor & (const at::Tensor &, at::OptionalSymIntArrayRef, at::OptionalIntArrayRef, ::std::optional<c10::string_view>, 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::fft_hfftn")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "fft_hfftn.out(Tensor self, SymInt[1]? s=None, int[1]? dim=None, str? norm=None, *, Tensor(a!) out) -> Tensor(a!)")
+  static const at::Tensor & call(const at::Tensor & self, at::OptionalSymIntArrayRef s, at::OptionalIntArrayRef dim, ::std::optional<c10::string_view> norm, const at::Tensor & out);
+  static const at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::OptionalSymIntArrayRef s, at::OptionalIntArrayRef dim, ::std::optional<c10::string_view> norm, const at::Tensor & out);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/frobenius_norm_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 frobenius_norm(const at::Tensor & self, at::IntArrayRef dim, bool keepdim=false);
+TORCH_API at::Tensor & frobenius_norm_out(const at::Tensor & self, at::IntArrayRef dim, bool keepdim, at::Tensor & out);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/histc.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/histc_ops.h>
+
+namespace at {
+
+
+// aten::histc.out(Tensor self, int bins=100, Scalar min=0, Scalar max=0, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & histc_out(at::Tensor & out, const at::Tensor & self, int64_t bins=100, const at::Scalar & min=0, const at::Scalar & max=0) {
+    return at::_ops::histc_out::call(self, bins, min, max, out);
+}
+// aten::histc.out(Tensor self, int bins=100, Scalar min=0, Scalar max=0, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & histc_outf(const at::Tensor & self, int64_t bins, const at::Scalar & min, const at::Scalar & max, at::Tensor & out) {
+    return at::_ops::histc_out::call(self, bins, min, max, out);
+}
+
+// aten::histc(Tensor self, int bins=100, Scalar min=0, Scalar max=0) -> Tensor
+inline at::Tensor histc(const at::Tensor & self, int64_t bins=100, const at::Scalar & min=0, const at::Scalar & max=0) {
+    return at::_ops::histc::call(self, bins, min, max);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/im2col_cpu_dispatch.h

@@ -0,0 +1,25 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 im2col(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride);
+TORCH_API at::Tensor & im2col_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride);
+TORCH_API at::Tensor & im2col_outf(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef dilation, at::IntArrayRef padding, at::IntArrayRef stride, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/int_repr_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 at::Tensor & int_repr_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & int_repr_outf(const at::Tensor & self, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_complex.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/is_complex_ops.h>
+
+namespace at {
+
+
+// aten::is_complex(Tensor self) -> bool
+inline bool __dispatch_is_complex(const at::Tensor & self) {
+    return at::_ops::is_complex::call(self);
+}
+
+}

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

@@ -0,0 +1,83 @@
+#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 isin_Tensor_Tensor_out {
+  using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, bool, 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::isin")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Tensor_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "isin.Tensor_Tensor_out(Tensor elements, Tensor test_elements, *, bool assume_unique=False, bool invert=False, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out);
+};
+
+struct TORCH_API isin_Tensor_Tensor {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, bool, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::isin")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "isin.Tensor_Tensor(Tensor elements, Tensor test_elements, *, bool assume_unique=False, bool invert=False) -> Tensor")
+  static at::Tensor call(const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique, bool invert);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & elements, const at::Tensor & test_elements, bool assume_unique, bool invert);
+};
+
+struct TORCH_API isin_Tensor_Scalar_out {
+  using schema = at::Tensor & (const at::Tensor &, const at::Scalar &, bool, 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::isin")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Scalar_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "isin.Tensor_Scalar_out(Tensor elements, Scalar test_element, *, bool assume_unique=False, bool invert=False, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique, bool invert, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique, bool invert, at::Tensor & out);
+};
+
+struct TORCH_API isin_Tensor_Scalar {
+  using schema = at::Tensor (const at::Tensor &, const at::Scalar &, bool, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::isin")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Tensor_Scalar")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "isin.Tensor_Scalar(Tensor elements, Scalar test_element, *, bool assume_unique=False, bool invert=False) -> Tensor")
+  static at::Tensor call(const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique, bool invert);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & elements, const at::Scalar & test_element, bool assume_unique, bool invert);
+};
+
+struct TORCH_API isin_Scalar_Tensor_out {
+  using schema = at::Tensor & (const at::Scalar &, const at::Tensor &, bool, 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::isin")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_Tensor_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "isin.Scalar_Tensor_out(Scalar element, Tensor test_elements, *, bool assume_unique=False, bool invert=False, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique, bool invert, at::Tensor & out);
+};
+
+struct TORCH_API isin_Scalar_Tensor {
+  using schema = at::Tensor (const at::Scalar &, const at::Tensor &, bool, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::isin")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Scalar_Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "isin.Scalar_Tensor(Scalar element, Tensor test_elements, *, bool assume_unique=False, bool invert=False) -> Tensor")
+  static at::Tensor call(const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique, bool invert);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Scalar & element, const at::Tensor & test_elements, bool assume_unique, bool invert);
+};
+
+}} // namespace at::_ops

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

@@ -0,0 +1,79 @@
+#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/kaiser_window_ops.h>
+
+namespace at {
+
+
+// aten::kaiser_window(int window_length, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor kaiser_window(int64_t window_length, at::TensorOptions options={}) {
+    return at::_ops::kaiser_window::call(window_length, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+}
+// aten::kaiser_window(int window_length, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor kaiser_window(int64_t window_length, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory) {
+    return at::_ops::kaiser_window::call(window_length, dtype, layout, device, pin_memory);
+}
+
+// aten::kaiser_window.periodic(int window_length, bool periodic, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor kaiser_window(int64_t window_length, bool periodic, at::TensorOptions options={}) {
+    return at::_ops::kaiser_window_periodic::call(window_length, periodic, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+}
+// aten::kaiser_window.periodic(int window_length, bool periodic, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor kaiser_window(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) {
+    return at::_ops::kaiser_window_periodic::call(window_length, periodic, dtype, layout, device, pin_memory);
+}
+
+// aten::kaiser_window.beta(int window_length, bool periodic, float beta, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor kaiser_window(int64_t window_length, bool periodic, double beta, at::TensorOptions options={}) {
+    return at::_ops::kaiser_window_beta::call(window_length, periodic, beta, c10::optTypeMetaToScalarType(options.dtype_opt()), options.layout_opt(), options.device_opt(), options.pinned_memory_opt());
+}
+// aten::kaiser_window.beta(int window_length, bool periodic, float beta, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
+inline at::Tensor kaiser_window(int64_t window_length, bool periodic, double beta, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory) {
+    return at::_ops::kaiser_window_beta::call(window_length, periodic, beta, dtype, layout, device, pin_memory);
+}
+
+// aten::kaiser_window.out(int window_length, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & kaiser_window_out(at::Tensor & out, int64_t window_length) {
+    return at::_ops::kaiser_window_out::call(window_length, out);
+}
+// aten::kaiser_window.out(int window_length, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & kaiser_window_outf(int64_t window_length, at::Tensor & out) {
+    return at::_ops::kaiser_window_out::call(window_length, out);
+}
+
+// aten::kaiser_window.periodic_out(int window_length, bool periodic, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & kaiser_window_out(at::Tensor & out, int64_t window_length, bool periodic) {
+    return at::_ops::kaiser_window_periodic_out::call(window_length, periodic, out);
+}
+// aten::kaiser_window.periodic_out(int window_length, bool periodic, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & kaiser_window_outf(int64_t window_length, bool periodic, at::Tensor & out) {
+    return at::_ops::kaiser_window_periodic_out::call(window_length, periodic, out);
+}
+
+// aten::kaiser_window.beta_out(int window_length, bool periodic, float beta, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & kaiser_window_out(at::Tensor & out, int64_t window_length, bool periodic, double beta) {
+    return at::_ops::kaiser_window_beta_out::call(window_length, periodic, beta, out);
+}
+// aten::kaiser_window.beta_out(int window_length, bool periodic, float beta, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & kaiser_window_outf(int64_t window_length, bool periodic, double beta, at::Tensor & out) {
+    return at::_ops::kaiser_window_beta_out::call(window_length, periodic, beta, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_inv_ex.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/linalg_inv_ex_ops.h>
+
+namespace at {
+
+
+// aten::linalg_inv_ex(Tensor A, *, bool check_errors=False) -> (Tensor inverse, Tensor info)
+inline ::std::tuple<at::Tensor,at::Tensor> linalg_inv_ex(const at::Tensor & A, bool check_errors=false) {
+    return at::_ops::linalg_inv_ex::call(A, check_errors);
+}
+
+// aten::linalg_inv_ex.inverse(Tensor A, *, bool check_errors=False, Tensor(a!) inverse, Tensor(b!) info) -> (Tensor(a!) inverse, Tensor(b!) info)
+inline ::std::tuple<at::Tensor &,at::Tensor &> linalg_inv_ex_out(at::Tensor & inverse, at::Tensor & info, const at::Tensor & A, bool check_errors=false) {
+    return at::_ops::linalg_inv_ex_inverse::call(A, check_errors, inverse, info);
+}
+// aten::linalg_inv_ex.inverse(Tensor A, *, bool check_errors=False, Tensor(a!) inverse, Tensor(b!) info) -> (Tensor(a!) inverse, Tensor(b!) info)
+inline ::std::tuple<at::Tensor &,at::Tensor &> linalg_inv_ex_outf(const at::Tensor & A, bool check_errors, at::Tensor & inverse, at::Tensor & info) {
+    return at::_ops::linalg_inv_ex_inverse::call(A, check_errors, inverse, info);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_slogdet_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 ::std::tuple<at::Tensor,at::Tensor> linalg_slogdet(const at::Tensor & A);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> linalg_slogdet_out(const at::Tensor & A, at::Tensor & sign, at::Tensor & logabsdet);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/lu_solve_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 lu_solve(const at::Tensor & self, const at::Tensor & LU_data, const at::Tensor & LU_pivots);
+TORCH_API at::Tensor & lu_solve_out(const at::Tensor & self, const at::Tensor & LU_data, const at::Tensor & LU_pivots, at::Tensor & out);
+} // namespace native
+} // namespace at

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/median_compositeexplicitautograd_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 compositeexplicitautograd {
+
+TORCH_API at::Tensor & median_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & median_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> median(const at::Tensor & self, int64_t dim, bool keepdim=false);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mm_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 mm(const at::Tensor & self, const at::Tensor & mat2);
+TORCH_API at::Tensor & mm_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & mat2);
+TORCH_API at::Tensor & mm_outf(const at::Tensor & self, const at::Tensor & mat2, at::Tensor & out);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mode_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 ::std::tuple<at::Tensor,at::Tensor> mode(const at::Tensor & self, int64_t dim=-1, bool keepdim=false);
+
+} // namespace cuda
+} // namespace at