Add files using upload-large-folder tool

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_amp_foreach_non_finite_check_and_unscale_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 void _amp_foreach_non_finite_check_and_unscale_(at::TensorList self, at::Tensor & found_inf, const at::Tensor & inv_scale);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_batch_norm_with_update.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/_batch_norm_with_update_ops.h>
+
+namespace at {
+
+
+// aten::_batch_norm_with_update(Tensor input, Tensor? weight, Tensor? bias, Tensor(a!) running_mean, Tensor(b!) running_var, float momentum, float eps) -> (Tensor, Tensor, Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor> _batch_norm_with_update(const at::Tensor & input, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & bias, at::Tensor & running_mean, at::Tensor & running_var, double momentum, double eps) {
+    return at::_ops::_batch_norm_with_update::call(input, weight, bias, running_mean, running_var, momentum, eps);
+}
+
+// aten::_batch_norm_with_update.out(Tensor input, Tensor? weight, Tensor? bias, Tensor(a!) running_mean, Tensor(b!) running_var, float momentum, float eps, *, Tensor(d!) out, Tensor(e!) save_mean, Tensor(f!) save_invstd, Tensor(g!) reserve) -> (Tensor(d!), Tensor(e!), Tensor(f!), Tensor(g!))
+inline ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &,at::Tensor &> _batch_norm_with_update_out(at::Tensor & out, at::Tensor & save_mean, at::Tensor & save_invstd, at::Tensor & reserve, const at::Tensor & input, const ::std::optional<at::Tensor> & weight, const ::std::optional<at::Tensor> & bias, at::Tensor & running_mean, at::Tensor & running_var, double momentum, double eps) {
+    return at::_ops::_batch_norm_with_update_out::call(input, weight, bias, running_mean, running_var, momentum, eps, out, save_mean, save_invstd, reserve);
+}
+// aten::_batch_norm_with_update.out(Tensor input, Tensor? weight, Tensor? bias, Tensor(a!) running_mean, Tensor(b!) running_var, float momentum, float eps, *, Tensor(d!) out, Tensor(e!) save_mean, Tensor(f!) save_invstd, Tensor(g!) reserve) -> (Tensor(d!), Tensor(e!), Tensor(f!), Tensor(g!))
+inline ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &,at::Tensor &> _batch_norm_with_update_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, double momentum, double eps, at::Tensor & out, at::Tensor & save_mean, at::Tensor & save_invstd, at::Tensor & reserve) {
+    return at::_ops::_batch_norm_with_update_out::call(input, weight, bias, running_mean, running_var, momentum, eps, out, save_mean, save_invstd, reserve);
+}
+
+// aten::_batch_norm_with_update_functional(Tensor input, Tensor? weight, Tensor? bias, Tensor running_mean, Tensor running_var, float momentum, float eps) -> (Tensor, Tensor, Tensor, Tensor, Tensor running_mean_out, Tensor running_var_out)
+inline ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor,at::Tensor,at::Tensor> _batch_norm_with_update_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, double momentum, double eps) {
+    return at::_ops::_batch_norm_with_update_functional::call(input, weight, bias, running_mean, running_var, momentum, eps);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cufft_set_plan_cache_max_size_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 void _cufft_set_plan_cache_max_size(at::DeviceIndex device_index, int64_t max_size);
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

@@ -0,0 +1,26 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 & _embedding_bag_dense_backward_out(at::Tensor & out, const at::Tensor & grad, const at::Tensor & indices, const at::Tensor & offset2bag, const at::Tensor & bag_size, const at::Tensor & maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode, const ::std::optional<at::Tensor> & per_sample_weights, int64_t padding_idx=-1);
+TORCH_API at::Tensor & _embedding_bag_dense_backward_outf(const at::Tensor & grad, const at::Tensor & indices, const at::Tensor & offset2bag, const at::Tensor & bag_size, const at::Tensor & maximum_indices, int64_t num_weights, bool scale_grad_by_freq, int64_t mode, const ::std::optional<at::Tensor> & per_sample_weights, int64_t padding_idx, at::Tensor & out);
+TORCH_API at::Tensor & _embedding_bag_dense_backward_symint_out(at::Tensor & out, const at::Tensor & grad, const at::Tensor & indices, const at::Tensor & offset2bag, const at::Tensor & bag_size, const at::Tensor & maximum_indices, 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);
+TORCH_API at::Tensor & _embedding_bag_dense_backward_symint_outf(const at::Tensor & grad, const at::Tensor & indices, const at::Tensor & offset2bag, const at::Tensor & bag_size, const at::Tensor & maximum_indices, c10::SymInt num_weights, bool scale_grad_by_freq, int64_t mode, const ::std::optional<at::Tensor> & per_sample_weights, int64_t padding_idx, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fill_mem_eff_dropout_mask_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 _fill_mem_eff_dropout_mask_ {
+  using schema = at::Tensor & (at::Tensor &, double, int64_t, int64_t);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_fill_mem_eff_dropout_mask_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_fill_mem_eff_dropout_mask_(Tensor(a!) self, float dropout_p, int seed, int offset) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, double dropout_p, int64_t seed, int64_t offset);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, double dropout_p, int64_t seed, int64_t offset);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foobar_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 & _foobar_out(at::Tensor & out, const at::Tensor & self, bool arg1=true, bool arg2=true, bool arg3=true);
+TORCH_API at::Tensor & _foobar_outf(const at::Tensor & self, bool arg1, bool arg2, bool arg3, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

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

@@ -0,0 +1,24 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cuda {
+
+TORCH_API ::std::vector<at::Tensor> _foreach_cosh(at::TensorList self);
+TORCH_API void _foreach_cosh_(at::TensorList self);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log1p_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_log1p(at::TensorList self);
+TORCH_API void _foreach_log1p_out(at::TensorList out, at::TensorList self);
+TORCH_API void _foreach_log1p_outf(at::TensorList self, at::TensorList out);
+TORCH_API void _foreach_log1p_(at::TensorList self);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_log_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 _foreach_log {
+  using schema = ::std::vector<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::_foreach_log")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_log(Tensor[] self) -> Tensor[]")
+  static ::std::vector<at::Tensor> call(at::TensorList self);
+  static ::std::vector<at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self);
+};
+
+struct TORCH_API _foreach_log_ {
+  using schema = void (at::TensorList);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_log_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_log_(Tensor(a!)[] self) -> ()")
+  static void call(at::TensorList self);
+  static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self);
+};
+
+struct TORCH_API _foreach_log_out {
+  using schema = void (at::TensorList, at::TensorList);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_foreach_log")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_foreach_log.out(Tensor[] self, *, Tensor(a!)[] out) -> ()")
+  static void call(at::TensorList self, at::TensorList out);
+  static void redispatch(c10::DispatchKeySet dispatchKeySet, at::TensorList self, at::TensorList out);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_maximum.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_maximum_ops.h>
+
+namespace at {
+
+
+// aten::_foreach_maximum.Scalar(Tensor[] self, Scalar scalar) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_maximum(at::TensorList self, const at::Scalar & scalar) {
+    return at::_ops::_foreach_maximum_Scalar::call(self, scalar);
+}
+
+// aten::_foreach_maximum_.Scalar(Tensor(a!)[] self, Scalar scalar) -> ()
+inline void _foreach_maximum_(at::TensorList self, const at::Scalar & scalar) {
+    return at::_ops::_foreach_maximum__Scalar::call(self, scalar);
+}
+
+// aten::_foreach_maximum.List(Tensor[] self, Tensor[] other) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_maximum(at::TensorList self, at::TensorList other) {
+    return at::_ops::_foreach_maximum_List::call(self, other);
+}
+
+// aten::_foreach_maximum_.List(Tensor(a!)[] self, Tensor[] other) -> ()
+inline void _foreach_maximum_(at::TensorList self, at::TensorList other) {
+    return at::_ops::_foreach_maximum__List::call(self, other);
+}
+
+// aten::_foreach_maximum.ScalarList(Tensor[] self, Scalar[] scalars) -> Tensor[]
+inline ::std::vector<at::Tensor> _foreach_maximum(at::TensorList self, at::ArrayRef<at::Scalar> scalars) {
+    return at::_ops::_foreach_maximum_ScalarList::call(self, scalars);
+}
+
+// aten::_foreach_maximum_.ScalarList(Tensor(a!)[] self, Scalar[] scalars) -> ()
+inline void _foreach_maximum_(at::TensorList self, at::ArrayRef<at::Scalar> scalars) {
+    return at::_ops::_foreach_maximum__ScalarList::call(self, scalars);
+}
+
+// aten::_foreach_maximum.Scalar_out(Tensor[] self, Scalar scalar, *, Tensor(a!)[] out) -> ()
+inline void _foreach_maximum_out(at::TensorList out, at::TensorList self, const at::Scalar & scalar) {
+    return at::_ops::_foreach_maximum_Scalar_out::call(self, scalar, out);
+}
+// aten::_foreach_maximum.Scalar_out(Tensor[] self, Scalar scalar, *, Tensor(a!)[] out) -> ()
+inline void _foreach_maximum_outf(at::TensorList self, const at::Scalar & scalar, at::TensorList out) {
+    return at::_ops::_foreach_maximum_Scalar_out::call(self, scalar, out);
+}
+
+// aten::_foreach_maximum.List_out(Tensor[] self, Tensor[] other, *, Tensor(a!)[] out) -> ()
+inline void _foreach_maximum_out(at::TensorList out, at::TensorList self, at::TensorList other) {
+    return at::_ops::_foreach_maximum_List_out::call(self, other, out);
+}
+// aten::_foreach_maximum.List_out(Tensor[] self, Tensor[] other, *, Tensor(a!)[] out) -> ()
+inline void _foreach_maximum_outf(at::TensorList self, at::TensorList other, at::TensorList out) {
+    return at::_ops::_foreach_maximum_List_out::call(self, other, out);
+}
+
+// aten::_foreach_maximum.ScalarList_out(Tensor[] self, Scalar[] scalars, *, Tensor(a!)[] out) -> ()
+inline void _foreach_maximum_out(at::TensorList out, at::TensorList self, at::ArrayRef<at::Scalar> scalars) {
+    return at::_ops::_foreach_maximum_ScalarList_out::call(self, scalars, out);
+}
+// aten::_foreach_maximum.ScalarList_out(Tensor[] self, Scalar[] scalars, *, Tensor(a!)[] out) -> ()
+inline void _foreach_maximum_outf(at::TensorList self, at::ArrayRef<at::Scalar> scalars, at::TensorList out) {
+    return at::_ops::_foreach_maximum_ScalarList_out::call(self, scalars, out);
+}
+
+}

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

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 _functional_assert_scalar(const at::Scalar & self, c10::string_view assert_msg, const at::Tensor & dep_token);
+
+} // namespace compositeexplicitautograd
+} // namespace at

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_mkldnn_transpose_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 & _mkldnn_transpose_out(at::Tensor & out, const at::Tensor & self, int64_t dim0, int64_t dim1);
+TORCH_API at::Tensor & _mkldnn_transpose_outf(const at::Tensor & self, int64_t dim0, int64_t dim1, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_new_zeros_with_same_feature_meta_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 _new_zeros_with_same_feature_meta(const at::Tensor & self, const at::Tensor & other, int64_t self_num_batch_dims=0);
+TORCH_API at::Tensor & _new_zeros_with_same_feature_meta_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other, int64_t self_num_batch_dims=0);
+TORCH_API at::Tensor & _new_zeros_with_same_feature_meta_outf(const at::Tensor & self, const at::Tensor & other, int64_t self_num_batch_dims, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_saturate_weight_to_fp16_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 _saturate_weight_to_fp16(const at::Tensor & weight);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_attention_math.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/_scaled_dot_product_attention_math_ops.h>
+
+namespace at {
+
+
+// aten::_scaled_dot_product_attention_math(Tensor query, Tensor key, Tensor value, Tensor? attn_mask=None, float dropout_p=0.0, bool is_causal=False, Tensor? dropout_mask=None, *, float? scale=None, bool enable_gqa=False) -> (Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor> _scaled_dot_product_attention_math(const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const ::std::optional<at::Tensor> & attn_mask={}, double dropout_p=0.0, bool is_causal=false, const ::std::optional<at::Tensor> & dropout_mask={}, ::std::optional<double> scale=::std::nullopt, bool enable_gqa=false) {
+    return at::_ops::_scaled_dot_product_attention_math::call(query, key, value, attn_mask, dropout_p, is_causal, dropout_mask, scale, enable_gqa);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact3d_backward_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__upsample_nearest_exact3d_backward : public at::impl::MetaBase {
+    
+    
+    void meta(const at::Tensor & grad_output, at::ArrayRef<int64_t> output_size, at::ArrayRef<int64_t> input_size, ::std::optional<double> scales_d, ::std::optional<double> scales_h, ::std::optional<double> scales_w);
+};
+
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_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 ::std::tuple<at::Tensor,at::Tensor> adaptive_max_pool2d(const at::Tensor & self, at::IntArrayRef output_size);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/amax.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/amax_ops.h>
+
+namespace at {
+
+
+// aten::amax(Tensor self, int[1] dim=[], bool keepdim=False) -> Tensor
+inline at::Tensor amax(const at::Tensor & self, at::IntArrayRef dim={}, bool keepdim=false) {
+    return at::_ops::amax::call(self, dim, keepdim);
+}
+
+// aten::amax.out(Tensor self, int[1] dim=[], bool keepdim=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & amax_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef dim={}, bool keepdim=false) {
+    return at::_ops::amax_out::call(self, dim, keepdim, out);
+}
+// aten::amax.out(Tensor self, int[1] dim=[], bool keepdim=False, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & amax_outf(const at::Tensor & self, at::IntArrayRef dim, bool keepdim, at::Tensor & out) {
+    return at::_ops::amax_out::call(self, dim, keepdim, out);
+}
+
+}

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

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> batch_norm_update_stats(const at::Tensor & input, const ::std::optional<at::Tensor> & running_mean, const ::std::optional<at::Tensor> & running_var, double momentum);
+
+} // namespace cpu
+} // namespace at

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

@@ -0,0 +1,67 @@
+#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/bitwise_left_shift_ops.h>
+
+namespace at {
+
+
+// aten::bitwise_left_shift.Tensor(Tensor self, Tensor other) -> Tensor
+inline at::Tensor bitwise_left_shift(const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::bitwise_left_shift_Tensor::call(self, other);
+}
+
+// aten::bitwise_left_shift.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_left_shift_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::bitwise_left_shift_Tensor_out::call(self, other, out);
+}
+// aten::bitwise_left_shift.Tensor_out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_left_shift_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) {
+    return at::_ops::bitwise_left_shift_Tensor_out::call(self, other, out);
+}
+
+// aten::bitwise_left_shift.Tensor_Scalar(Tensor self, Scalar other) -> Tensor
+inline at::Tensor bitwise_left_shift(const at::Tensor & self, const at::Scalar & other) {
+    return at::_ops::bitwise_left_shift_Tensor_Scalar::call(self, other);
+}
+
+// aten::bitwise_left_shift.Tensor_Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_left_shift_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other) {
+    return at::_ops::bitwise_left_shift_Tensor_Scalar_out::call(self, other, out);
+}
+// aten::bitwise_left_shift.Tensor_Scalar_out(Tensor self, Scalar other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_left_shift_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out) {
+    return at::_ops::bitwise_left_shift_Tensor_Scalar_out::call(self, other, out);
+}
+
+// aten::bitwise_left_shift.Scalar_Tensor(Scalar self, Tensor other) -> Tensor
+inline at::Tensor bitwise_left_shift(const at::Scalar & self, const at::Tensor & other) {
+    return at::_ops::bitwise_left_shift_Scalar_Tensor::call(self, other);
+}
+
+// aten::bitwise_left_shift.Scalar_Tensor_out(Scalar self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_left_shift_out(at::Tensor & out, const at::Scalar & self, const at::Tensor & other) {
+    return at::_ops::bitwise_left_shift_Scalar_Tensor_out::call(self, other, out);
+}
+// aten::bitwise_left_shift.Scalar_Tensor_out(Scalar self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & bitwise_left_shift_outf(const at::Scalar & self, const at::Tensor & other, at::Tensor & out) {
+    return at::_ops::bitwise_left_shift_Scalar_Tensor_out::call(self, other, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/ceil_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 ceil(const at::Tensor & self);
+TORCH_API at::Tensor & ceil_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & ceil_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & ceil_(at::Tensor & self);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/conv_tbc_backward.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/conv_tbc_backward_ops.h>
+
+namespace at {
+
+
+// aten::conv_tbc_backward(Tensor self, Tensor input, Tensor weight, Tensor bias, int pad) -> (Tensor, Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor,at::Tensor> conv_tbc_backward(const at::Tensor & self, const at::Tensor & input, const at::Tensor & weight, const at::Tensor & bias, int64_t pad) {
+    return at::_ops::conv_tbc_backward::call(self, input, weight, bias, pad);
+}
+
+}

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagonal_scatter_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 & diagonal_scatter_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & src, int64_t offset=0, int64_t dim1=0, int64_t dim2=1);
+TORCH_API at::Tensor & diagonal_scatter_outf(const at::Tensor & self, const at::Tensor & src, int64_t offset, int64_t dim1, int64_t dim2, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/empty_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 at::Tensor empty(at::IntArrayRef size, ::std::optional<at::DimnameList> names, at::TensorOptions options={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty(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);
+TORCH_API at::Tensor & empty_out(at::Tensor & out, at::IntArrayRef size, ::std::optional<at::DimnameList> names, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor & empty_outf(at::IntArrayRef size, ::std::optional<at::DimnameList> names, ::std::optional<at::MemoryFormat> memory_format, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/expm1_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 expm1(const at::Tensor & self);
+TORCH_API at::Tensor & expm1_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & expm1_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & expm1_(at::Tensor & self);
+
+} // namespace cuda
+} // namespace at

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

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

@@ -0,0 +1,28 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 fft_irfft(const at::Tensor & self, ::std::optional<int64_t> n=::std::nullopt, int64_t dim=-1, ::std::optional<c10::string_view> norm=::std::nullopt);
+TORCH_API at::Tensor fft_irfft_symint(const at::Tensor & self, ::std::optional<c10::SymInt> n=::std::nullopt, int64_t dim=-1, ::std::optional<c10::string_view> norm=::std::nullopt);
+TORCH_API at::Tensor & fft_irfft_out(at::Tensor & out, const at::Tensor & self, ::std::optional<int64_t> n=::std::nullopt, int64_t dim=-1, ::std::optional<c10::string_view> norm=::std::nullopt);
+TORCH_API at::Tensor & fft_irfft_outf(const at::Tensor & self, ::std::optional<int64_t> n, int64_t dim, ::std::optional<c10::string_view> norm, at::Tensor & out);
+TORCH_API at::Tensor & fft_irfft_symint_out(at::Tensor & out, const at::Tensor & self, ::std::optional<c10::SymInt> n=::std::nullopt, int64_t dim=-1, ::std::optional<c10::string_view> norm=::std::nullopt);
+TORCH_API at::Tensor & fft_irfft_symint_outf(const at::Tensor & self, ::std::optional<c10::SymInt> n, int64_t dim, ::std::optional<c10::string_view> norm, at::Tensor & out);
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

@@ -0,0 +1,26 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 flatten(const at::Tensor & self, int64_t start_dim=0, int64_t end_dim=-1);
+TORCH_API at::Tensor flatten(const at::Tensor & self, int64_t start_dim, int64_t end_dim, at::Dimname out_dim);
+TORCH_API at::Tensor flatten(const at::Tensor & self, at::Dimname start_dim, at::Dimname end_dim, at::Dimname out_dim);
+TORCH_API at::Tensor flatten(const at::Tensor & self, at::DimnameList dims, at::Dimname out_dim);
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fmod_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 at::Tensor fmod(const at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor & fmod_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & other);
+TORCH_API at::Tensor & fmod_outf(const at::Tensor & self, const at::Scalar & other, at::Tensor & out);
+TORCH_API at::Tensor & fmod_(at::Tensor & self, const at::Scalar & other);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/full_like_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 full_like(const at::Tensor & self, const at::Scalar & fill_value, ::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 & full_like_out(const at::Tensor & self, const at::Scalar & fill_value, ::std::optional<at::MemoryFormat> memory_format, at::Tensor & out);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gelu.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/gelu_ops.h>
+
+namespace at {
+
+
+// aten::gelu.out(Tensor self, *, str approximate='none', Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & gelu_out(at::Tensor & out, const at::Tensor & self, c10::string_view approximate="none") {
+    return at::_ops::gelu_out::call(self, approximate, out);
+}
+// aten::gelu.out(Tensor self, *, str approximate='none', Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & gelu_outf(const at::Tensor & self, c10::string_view approximate, at::Tensor & out) {
+    return at::_ops::gelu_out::call(self, approximate, out);
+}
+
+// aten::gelu_(Tensor(a!) self, *, str approximate='none') -> Tensor(a!)
+inline at::Tensor & gelu_(at::Tensor & self, c10::string_view approximate="none") {
+    return at::_ops::gelu_::call(self, approximate);
+}
+
+// aten::gelu(Tensor self, *, str approximate='none') -> Tensor
+inline at::Tensor gelu(const at::Tensor & self, c10::string_view approximate="none") {
+    return at::_ops::gelu::call(self, approximate);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gelu_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 gelu_backward(const at::Tensor & grad_output, const at::Tensor & self, c10::string_view approximate="none");
+TORCH_API at::Tensor & gelu_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, c10::string_view approximate="none");
+TORCH_API at::Tensor & gelu_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, c10::string_view approximate, at::Tensor & grad_input);
+
+} // namespace cuda
+} // namespace at

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

@@ -0,0 +1,63 @@
+#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_fill_ops.h>
+
+namespace at {
+
+
+// aten::index_fill.int_Scalar(Tensor self, int dim, Tensor index, Scalar value) -> Tensor
+inline at::Tensor index_fill(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value) {
+    return at::_ops::index_fill_int_Scalar::call(self, dim, index, value);
+}
+
+// aten::index_fill.int_Tensor(Tensor self, int dim, Tensor index, Tensor value) -> Tensor
+inline at::Tensor index_fill(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & value) {
+    return at::_ops::index_fill_int_Tensor::call(self, dim, index, value);
+}
+
+// aten::index_fill.Dimname_Scalar(Tensor self, Dimname dim, Tensor index, Scalar value) -> Tensor
+inline at::Tensor index_fill(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Scalar & value) {
+    return at::_ops::index_fill_Dimname_Scalar::call(self, dim, index, value);
+}
+
+// aten::index_fill.Dimname_Tensor(Tensor self, Dimname dim, Tensor index, Tensor value) -> Tensor
+inline at::Tensor index_fill(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & value) {
+    return at::_ops::index_fill_Dimname_Tensor::call(self, dim, index, value);
+}
+
+// aten::index_fill.int_Scalar_out(Tensor self, int dim, Tensor index, Scalar value, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & index_fill_out(at::Tensor & out, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value) {
+    return at::_ops::index_fill_int_Scalar_out::call(self, dim, index, value, out);
+}
+// aten::index_fill.int_Scalar_out(Tensor self, int dim, Tensor index, Scalar value, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & index_fill_outf(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, at::Tensor & out) {
+    return at::_ops::index_fill_int_Scalar_out::call(self, dim, index, value, out);
+}
+
+// aten::index_fill.int_Tensor_out(Tensor self, int dim, Tensor index, Tensor value, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & index_fill_out(at::Tensor & out, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & value) {
+    return at::_ops::index_fill_int_Tensor_out::call(self, dim, index, value, out);
+}
+// aten::index_fill.int_Tensor_out(Tensor self, int dim, Tensor index, Tensor value, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & index_fill_outf(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & value, at::Tensor & out) {
+    return at::_ops::index_fill_int_Tensor_out::call(self, dim, index, value, out);
+}
+
+}

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

@@ -0,0 +1,127 @@
+#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 index_fill__int_Scalar {
+  using schema = at::Tensor & (at::Tensor &, int64_t, 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::index_fill_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "int_Scalar")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill_.int_Scalar(Tensor(a!) self, int dim, Tensor index, Scalar value) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value);
+};
+
+struct TORCH_API index_fill_int_Scalar {
+  using schema = at::Tensor (const at::Tensor &, int64_t, 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::index_fill")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "int_Scalar")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill.int_Scalar(Tensor self, int dim, Tensor index, Scalar value) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value);
+};
+
+struct TORCH_API index_fill__int_Tensor {
+  using schema = at::Tensor & (at::Tensor &, int64_t, 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::index_fill_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "int_Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill_.int_Tensor(Tensor(a!) self, int dim, Tensor index, Tensor value) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & value);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & value);
+};
+
+struct TORCH_API index_fill_int_Tensor {
+  using schema = at::Tensor (const at::Tensor &, int64_t, 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::index_fill")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "int_Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill.int_Tensor(Tensor self, int dim, Tensor index, Tensor value) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & value);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & value);
+};
+
+struct TORCH_API index_fill__Dimname_Scalar {
+  using schema = at::Tensor & (at::Tensor &, at::Dimname, 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::index_fill_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Dimname_Scalar")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill_.Dimname_Scalar(Tensor(a!) self, Dimname dim, Tensor index, Scalar value) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Scalar & value);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Scalar & value);
+};
+
+struct TORCH_API index_fill__Dimname_Tensor {
+  using schema = at::Tensor & (at::Tensor &, at::Dimname, 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::index_fill_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Dimname_Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill_.Dimname_Tensor(Tensor(a!) self, Dimname dim, Tensor index, Tensor value) -> Tensor(a!)")
+  static at::Tensor & call(at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & value);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & value);
+};
+
+struct TORCH_API index_fill_Dimname_Scalar {
+  using schema = at::Tensor (const at::Tensor &, at::Dimname, 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::index_fill")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Dimname_Scalar")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill.Dimname_Scalar(Tensor self, Dimname dim, Tensor index, Scalar value) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Scalar & value);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Scalar & value);
+};
+
+struct TORCH_API index_fill_Dimname_Tensor {
+  using schema = at::Tensor (const at::Tensor &, at::Dimname, 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::index_fill")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "Dimname_Tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill.Dimname_Tensor(Tensor self, Dimname dim, Tensor index, Tensor value) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & value);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim, const at::Tensor & index, const at::Tensor & value);
+};
+
+struct TORCH_API index_fill_int_Scalar_out {
+  using schema = at::Tensor & (const at::Tensor &, int64_t, const at::Tensor &, const at::Scalar &, at::Tensor &);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::index_fill")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "int_Scalar_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill.int_Scalar_out(Tensor self, int dim, Tensor index, Scalar value, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Scalar & value, at::Tensor & out);
+};
+
+struct TORCH_API index_fill_int_Tensor_out {
+  using schema = at::Tensor & (const at::Tensor &, int64_t, 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::index_fill")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "int_Tensor_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "index_fill.int_Tensor_out(Tensor self, int dim, Tensor index, Tensor value, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & value, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, const at::Tensor & index, const at::Tensor & value, at::Tensor & out);
+};
+
+}} // namespace at::_ops

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

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

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_qr_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 linalg_qr {
+  using schema = ::std::tuple<at::Tensor,at::Tensor> (const at::Tensor &, 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::linalg_qr")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_qr(Tensor A, str mode='reduced') -> (Tensor Q, Tensor R)")
+  static ::std::tuple<at::Tensor,at::Tensor> call(const at::Tensor & A, c10::string_view mode);
+  static ::std::tuple<at::Tensor,at::Tensor> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & A, c10::string_view mode);
+};
+
+struct TORCH_API linalg_qr_out {
+  using schema = ::std::tuple<at::Tensor &,at::Tensor &> (const at::Tensor &, c10::string_view, 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::linalg_qr")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_qr.out(Tensor A, str mode='reduced', *, Tensor(a!) Q, Tensor(b!) R) -> (Tensor(a!) Q, Tensor(b!) R)")
+  static ::std::tuple<at::Tensor &,at::Tensor &> call(const at::Tensor & A, c10::string_view mode, at::Tensor & Q, at::Tensor & R);
+  static ::std::tuple<at::Tensor &,at::Tensor &> redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & A, c10::string_view mode, at::Tensor & Q, at::Tensor & R);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log1p_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 log1p(const at::Tensor & self);
+TORCH_API at::Tensor & log1p_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & log1p_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & log1p_(at::Tensor & self);
+
+} // namespace meta
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logaddexp2.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/logaddexp2_ops.h>
+
+namespace at {
+
+
+// aten::logaddexp2.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & logaddexp2_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::logaddexp2_out::call(self, other, out);
+}
+// aten::logaddexp2.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & logaddexp2_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out) {
+    return at::_ops::logaddexp2_out::call(self, other, out);
+}
+
+// aten::logaddexp2(Tensor self, Tensor other) -> Tensor
+inline at::Tensor logaddexp2(const at::Tensor & self, const at::Tensor & other) {
+    return at::_ops::logaddexp2::call(self, other);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/logcumsumexp.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/logcumsumexp_ops.h>
+
+namespace at {
+
+
+// aten::logcumsumexp(Tensor self, int dim) -> Tensor
+inline at::Tensor logcumsumexp(const at::Tensor & self, int64_t dim) {
+    return at::_ops::logcumsumexp::call(self, dim);
+}
+
+// aten::logcumsumexp.out(Tensor self, int dim, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & logcumsumexp_out(at::Tensor & out, const at::Tensor & self, int64_t dim) {
+    return at::_ops::logcumsumexp_out::call(self, dim, out);
+}
+// aten::logcumsumexp.out(Tensor self, int dim, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & logcumsumexp_outf(const at::Tensor & self, int64_t dim, at::Tensor & out) {
+    return at::_ops::logcumsumexp_out::call(self, dim, out);
+}
+
+// aten::logcumsumexp.dimname(Tensor self, Dimname dim) -> Tensor
+inline at::Tensor logcumsumexp(const at::Tensor & self, at::Dimname dim) {
+    return at::_ops::logcumsumexp_dimname::call(self, dim);
+}
+
+// aten::logcumsumexp.dimname_out(Tensor self, Dimname dim, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & logcumsumexp_out(at::Tensor & out, const at::Tensor & self, at::Dimname dim) {
+    return at::_ops::logcumsumexp_dimname_out::call(self, dim, out);
+}
+// aten::logcumsumexp.dimname_out(Tensor self, Dimname dim, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & logcumsumexp_outf(const at::Tensor & self, at::Dimname dim, at::Tensor & out) {
+    return at::_ops::logcumsumexp_dimname_out::call(self, dim, out);
+}
+
+}

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

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

@@ -0,0 +1,25 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace meta {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor> lu_unpack(const at::Tensor & LU_data, const at::Tensor & LU_pivots, bool unpack_data=true, bool unpack_pivots=true);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> lu_unpack_out(at::Tensor & P, at::Tensor & L, at::Tensor & U, const at::Tensor & LU_data, const at::Tensor & LU_pivots, bool unpack_data=true, bool unpack_pivots=true);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &,at::Tensor &> lu_unpack_outf(const at::Tensor & LU_data, const at::Tensor & LU_pivots, bool unpack_data, bool unpack_pivots, at::Tensor & P, at::Tensor & L, at::Tensor & U);
+
+} // namespace meta
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/matmul_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/matmul_backward_ops.h>
+
+namespace at {
+
+
+// aten::matmul_backward(Tensor grad, Tensor self, Tensor other, bool[2] mask) -> (Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor> matmul_backward(const at::Tensor & grad, const at::Tensor & self, const at::Tensor & other, ::std::array<bool,2> mask) {
+    return at::_ops::matmul_backward::call(grad, self, other, mask);
+}
+
+// aten::matmul_backward.out(Tensor grad, Tensor self, Tensor other, bool[2] mask, *, Tensor(a!) out0, Tensor(b!) out1) -> (Tensor(a!), Tensor(b!))
+inline ::std::tuple<at::Tensor &,at::Tensor &> matmul_backward_out(at::Tensor & out0, at::Tensor & out1, const at::Tensor & grad, const at::Tensor & self, const at::Tensor & other, ::std::array<bool,2> mask) {
+    return at::_ops::matmul_backward_out::call(grad, self, other, mask, out0, out1);
+}
+// aten::matmul_backward.out(Tensor grad, Tensor self, Tensor other, bool[2] mask, *, Tensor(a!) out0, Tensor(b!) out1) -> (Tensor(a!), Tensor(b!))
+inline ::std::tuple<at::Tensor &,at::Tensor &> matmul_backward_outf(const at::Tensor & grad, const at::Tensor & self, const at::Tensor & other, ::std::array<bool,2> mask, at::Tensor & out0, at::Tensor & out1) {
+    return at::_ops::matmul_backward_out::call(grad, self, other, mask, out0, out1);
+}
+
+}

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

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/mish_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 mish {
+  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::mish")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mish(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 mish_ {
+  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::mish_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mish_(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 mish_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::mish")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "mish.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