Add files using upload-large-folder tool

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_amp_foreach_non_finite_check_and_unscale.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/_amp_foreach_non_finite_check_and_unscale_ops.h>
+
+namespace at {
+
+
+// aten::_amp_foreach_non_finite_check_and_unscale_(Tensor(a!)[] self, Tensor(b!) found_inf, Tensor inv_scale) -> ()
+inline void _amp_foreach_non_finite_check_and_unscale_(at::TensorList self, at::Tensor & found_inf, const at::Tensor & inv_scale) {
+    return at::_ops::_amp_foreach_non_finite_check_and_unscale_::call(self, found_inf, inv_scale);
+}
+
+// aten::_amp_foreach_non_finite_check_and_unscale.out(Tensor[] self, Tensor(b!) found_inf, Tensor inv_scale, *, Tensor(a!)[] out) -> ()
+inline void _amp_foreach_non_finite_check_and_unscale_out(at::TensorList out, at::TensorList self, at::Tensor & found_inf, const at::Tensor & inv_scale) {
+    return at::_ops::_amp_foreach_non_finite_check_and_unscale_out::call(self, found_inf, inv_scale, out);
+}
+// aten::_amp_foreach_non_finite_check_and_unscale.out(Tensor[] self, Tensor(b!) found_inf, Tensor inv_scale, *, Tensor(a!)[] out) -> ()
+inline void _amp_foreach_non_finite_check_and_unscale_outf(at::TensorList self, at::Tensor & found_inf, const at::Tensor & inv_scale, at::TensorList out) {
+    return at::_ops::_amp_foreach_non_finite_check_and_unscale_out::call(self, found_inf, inv_scale, out);
+}
+
+// aten::_amp_foreach_non_finite_check_and_unscale(Tensor[] self, Tensor found_inf, Tensor inv_scale) -> (Tensor[] self_out, Tensor found_inf_out)
+inline ::std::tuple<::std::vector<at::Tensor>,at::Tensor> _amp_foreach_non_finite_check_and_unscale(at::TensorList self, const at::Tensor & found_inf, const at::Tensor & inv_scale) {
+    return at::_ops::_amp_foreach_non_finite_check_and_unscale::call(self, found_inf, inv_scale);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_assert_async_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 _assert_async(const at::Tensor & self);
+TORCH_API void _assert_async(const at::Tensor & self, c10::string_view assert_msg);
+
+} // namespace cuda
+} // namespace at

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cudnn_init_dropout_state_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_init_dropout_state_out(double dropout, bool train, int64_t dropout_seed, at::Tensor & out);
+TORCH_API at::Tensor _cudnn_init_dropout_state(double dropout, bool train, int64_t dropout_seed, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={});
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,23 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <c10/core/MemoryFormat.h>
+#include <c10/core/Scalar.h>
+#include <ATen/core/Reduction.h>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+
+namespace cpu {
+
+TORCH_API ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor> _embedding_bag_forward_only(const at::Tensor & weight, const at::Tensor & indices, const at::Tensor & offsets, bool scale_grad_by_freq=false, int64_t mode=0, bool sparse=false, const ::std::optional<at::Tensor> & per_sample_weights={}, bool include_last_offset=false, int64_t padding_idx=-1);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_fake_quantize_per_tensor_affine_cachemask_tensor_qparams_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 &> _fake_quantize_per_tensor_affine_cachemask_tensor_qparams_out(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, const at::Tensor & fake_quant_enabled, int64_t quant_min, int64_t quant_max, at::Tensor & out0, at::Tensor & out1);
+TORCH_API ::std::tuple<at::Tensor,at::Tensor> _fake_quantize_per_tensor_affine_cachemask_tensor_qparams(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, const at::Tensor & fake_quant_enabled, int64_t quant_min, int64_t quant_max);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_make_dual_copy.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/_make_dual_copy_ops.h>
+
+namespace at {
+
+
+// aten::_make_dual_copy(Tensor primal, Tensor tangent, int level) -> Tensor
+inline at::Tensor _make_dual_copy(const at::Tensor & primal, const at::Tensor & tangent, int64_t level) {
+    return at::_ops::_make_dual_copy::call(primal, tangent, level);
+}
+
+// aten::_make_dual_copy.out(Tensor primal, Tensor tangent, int level, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _make_dual_copy_out(at::Tensor & out, const at::Tensor & primal, const at::Tensor & tangent, int64_t level) {
+    return at::_ops::_make_dual_copy_out::call(primal, tangent, level, out);
+}
+// aten::_make_dual_copy.out(Tensor primal, Tensor tangent, int level, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _make_dual_copy_outf(const at::Tensor & primal, const at::Tensor & tangent, int64_t level, at::Tensor & out) {
+    return at::_ops::_make_dual_copy_out::call(primal, tangent, level, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_make_per_channel_quantized_tensor.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/_make_per_channel_quantized_tensor_ops.h>
+
+namespace at {
+
+
+// aten::_make_per_channel_quantized_tensor(Tensor self, Tensor scale, Tensor zero_point, int axis) -> Tensor
+inline at::Tensor _make_per_channel_quantized_tensor(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis) {
+    return at::_ops::_make_per_channel_quantized_tensor::call(self, scale, zero_point, axis);
+}
+
+// aten::_make_per_channel_quantized_tensor.out(Tensor self, Tensor scale, Tensor zero_point, int axis, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _make_per_channel_quantized_tensor_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis) {
+    return at::_ops::_make_per_channel_quantized_tensor_out::call(self, scale, zero_point, axis, out);
+}
+// aten::_make_per_channel_quantized_tensor.out(Tensor self, Tensor scale, Tensor zero_point, int axis, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _make_per_channel_quantized_tensor_outf(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, int64_t axis, at::Tensor & out) {
+    return at::_ops::_make_per_channel_quantized_tensor_out::call(self, scale, zero_point, axis, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_make_per_tensor_quantized_tensor_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 & _make_per_tensor_quantized_tensor_out(at::Tensor & out, const at::Tensor & self, double scale, int64_t zero_point);
+TORCH_API at::Tensor & _make_per_tensor_quantized_tensor_outf(const at::Tensor & self, double scale, int64_t zero_point, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

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

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

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

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_compute_contiguous_strides_offsets_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> _nested_compute_contiguous_strides_offsets(const at::Tensor & nested_size);
+
+} // namespace cpu
+} // namespace at

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

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

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_sparse_semi_structured_apply_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> _sparse_semi_structured_apply(const at::Tensor & input, const at::Tensor & thread_masks);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_ambiguous_defaults_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 _test_ambiguous_defaults_a {
+  using schema = at::Tensor (const at::Tensor &, 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::_test_ambiguous_defaults")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "a")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_test_ambiguous_defaults.a(Tensor dummy, int a=1, int b=1) -> Tensor")
+  static at::Tensor call(const at::Tensor & dummy, int64_t a, int64_t b);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & dummy, int64_t a, int64_t b);
+};
+
+struct TORCH_API _test_ambiguous_defaults_b {
+  using schema = at::Tensor (const at::Tensor &, int64_t, 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::_test_ambiguous_defaults")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "b")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_test_ambiguous_defaults.b(Tensor dummy, int a=2, str b=\"2\") -> Tensor")
+  static at::Tensor call(const at::Tensor & dummy, int64_t a, c10::string_view b);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & dummy, int64_t a, c10::string_view b);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_test_parallel_materialize_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 _test_parallel_materialize(const at::Tensor & self, int64_t num_parallel, bool skip_first=false);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_differentiable_gru_cell_backward_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 ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor,at::Tensor> _thnn_differentiable_gru_cell_backward(const at::Tensor & grad_hy, const at::Tensor & input_gates, const at::Tensor & hidden_gates, const at::Tensor & hx, const ::std::optional<at::Tensor> & input_bias, const ::std::optional<at::Tensor> & hidden_bias);
+
+} // namespace compositeimplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_thnn_fused_lstm_cell_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/_thnn_fused_lstm_cell_backward_ops.h>
+
+namespace at {
+
+
+// aten::_thnn_fused_lstm_cell_backward(Tensor? grad_hy, Tensor? grad_cy, Tensor cx, Tensor cy, Tensor workspace, bool has_bias) -> (Tensor, Tensor, Tensor, Tensor, Tensor)
+inline ::std::tuple<at::Tensor,at::Tensor,at::Tensor,at::Tensor,at::Tensor> _thnn_fused_lstm_cell_backward(const ::std::optional<at::Tensor> & grad_hy, const ::std::optional<at::Tensor> & grad_cy, const at::Tensor & cx, const at::Tensor & cy, const at::Tensor & workspace, bool has_bias) {
+    return at::_ops::_thnn_fused_lstm_cell_backward::call(grad_hy, grad_cy, cx, cy, workspace, has_bias);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_transformer_encoder_layer_fwd.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/_transformer_encoder_layer_fwd_ops.h>
+
+namespace at {
+
+
+// aten::_transformer_encoder_layer_fwd(Tensor src, int embed_dim, int num_heads, Tensor qkv_weight, Tensor qkv_bias, Tensor proj_weight, Tensor proj_bias, bool use_gelu, bool norm_first, float eps, Tensor norm_weight_1, Tensor norm_bias_1, Tensor norm_weight_2, Tensor norm_bias_2, Tensor ffn_weight_1, Tensor ffn_bias_1, Tensor ffn_weight_2, Tensor ffn_bias_2, Tensor? mask=None, int? mask_type=None) -> Tensor
+inline at::Tensor _transformer_encoder_layer_fwd(const at::Tensor & src, int64_t embed_dim, int64_t num_heads, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, bool use_gelu, bool norm_first, double eps, const at::Tensor & norm_weight_1, const at::Tensor & norm_bias_1, const at::Tensor & norm_weight_2, const at::Tensor & norm_bias_2, const at::Tensor & ffn_weight_1, const at::Tensor & ffn_bias_1, const at::Tensor & ffn_weight_2, const at::Tensor & ffn_bias_2, const ::std::optional<at::Tensor> & mask={}, ::std::optional<int64_t> mask_type=::std::nullopt) {
+    return at::_ops::_transformer_encoder_layer_fwd::call(src, embed_dim, num_heads, qkv_weight, qkv_bias, proj_weight, proj_bias, use_gelu, norm_first, eps, norm_weight_1, norm_bias_1, norm_weight_2, norm_bias_2, ffn_weight_1, ffn_bias_1, ffn_weight_2, ffn_bias_2, mask, mask_type);
+}
+
+// aten::_transformer_encoder_layer_fwd.out(Tensor src, int embed_dim, int num_heads, Tensor qkv_weight, Tensor qkv_bias, Tensor proj_weight, Tensor proj_bias, bool use_gelu, bool norm_first, float eps, Tensor norm_weight_1, Tensor norm_bias_1, Tensor norm_weight_2, Tensor norm_bias_2, Tensor ffn_weight_1, Tensor ffn_bias_1, Tensor ffn_weight_2, Tensor ffn_bias_2, Tensor? mask=None, int? mask_type=None, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _transformer_encoder_layer_fwd_out(at::Tensor & out, const at::Tensor & src, int64_t embed_dim, int64_t num_heads, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, bool use_gelu, bool norm_first, double eps, const at::Tensor & norm_weight_1, const at::Tensor & norm_bias_1, const at::Tensor & norm_weight_2, const at::Tensor & norm_bias_2, const at::Tensor & ffn_weight_1, const at::Tensor & ffn_bias_1, const at::Tensor & ffn_weight_2, const at::Tensor & ffn_bias_2, const ::std::optional<at::Tensor> & mask={}, ::std::optional<int64_t> mask_type=::std::nullopt) {
+    return at::_ops::_transformer_encoder_layer_fwd_out::call(src, embed_dim, num_heads, qkv_weight, qkv_bias, proj_weight, proj_bias, use_gelu, norm_first, eps, norm_weight_1, norm_bias_1, norm_weight_2, norm_bias_2, ffn_weight_1, ffn_bias_1, ffn_weight_2, ffn_bias_2, mask, mask_type, out);
+}
+// aten::_transformer_encoder_layer_fwd.out(Tensor src, int embed_dim, int num_heads, Tensor qkv_weight, Tensor qkv_bias, Tensor proj_weight, Tensor proj_bias, bool use_gelu, bool norm_first, float eps, Tensor norm_weight_1, Tensor norm_bias_1, Tensor norm_weight_2, Tensor norm_bias_2, Tensor ffn_weight_1, Tensor ffn_bias_1, Tensor ffn_weight_2, Tensor ffn_bias_2, Tensor? mask=None, int? mask_type=None, *, Tensor(a!) out) -> Tensor(a!)
+inline at::Tensor & _transformer_encoder_layer_fwd_outf(const at::Tensor & src, int64_t embed_dim, int64_t num_heads, const at::Tensor & qkv_weight, const at::Tensor & qkv_bias, const at::Tensor & proj_weight, const at::Tensor & proj_bias, bool use_gelu, bool norm_first, double eps, const at::Tensor & norm_weight_1, const at::Tensor & norm_bias_1, const at::Tensor & norm_weight_2, const at::Tensor & norm_bias_2, const at::Tensor & ffn_weight_1, const at::Tensor & ffn_bias_1, const at::Tensor & ffn_weight_2, const at::Tensor & ffn_bias_2, const ::std::optional<at::Tensor> & mask, ::std::optional<int64_t> mask_type, at::Tensor & out) {
+    return at::_ops::_transformer_encoder_layer_fwd_out::call(src, embed_dim, num_heads, qkv_weight, qkv_bias, proj_weight, proj_bias, use_gelu, norm_first, eps, norm_weight_1, norm_bias_1, norm_weight_2, norm_bias_2, ffn_weight_1, ffn_bias_1, ffn_weight_2, ffn_bias_2, mask, mask_type, out);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact2d_backward.h

@@ -0,0 +1,91 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/_upsample_nearest_exact2d_backward_ops.h>
+
+namespace at {
+
+
+// aten::_upsample_nearest_exact2d_backward.grad_input(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w, grad_input);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & _upsample_nearest_exact2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w, grad_input);
+  }
+}
+
+// aten::_upsample_nearest_exact2d_backward.grad_input(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact2d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales_h, ::std::optional<double> scales_w, at::Tensor & grad_input) {
+    return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w, grad_input);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, int64_t>::value>>
+  at::Tensor & _upsample_nearest_exact2d_backward_outf(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales_h, ::std::optional<double> scales_w, at::Tensor & grad_input) {
+    return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w, grad_input);
+  }
+}
+
+// aten::_upsample_nearest_exact2d_backward.grad_input(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact2d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, output_size, input_size, scales_h, scales_w, grad_input);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & _upsample_nearest_exact2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, output_size, input_size, scales_h, scales_w, grad_input);
+  }
+}
+
+// aten::_upsample_nearest_exact2d_backward.grad_input(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None, *, Tensor(a!) grad_input) -> Tensor(a!)
+inline at::Tensor & _upsample_nearest_exact2d_backward_symint_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales_h, ::std::optional<double> scales_w, at::Tensor & grad_input) {
+    return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, output_size, input_size, scales_h, scales_w, grad_input);
+}
+namespace symint {
+  template <typename T, typename = std::enable_if_t<std::is_same<T, c10::SymInt>::value>>
+  at::Tensor & _upsample_nearest_exact2d_backward_outf(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales_h, ::std::optional<double> scales_w, at::Tensor & grad_input) {
+    return at::_ops::_upsample_nearest_exact2d_backward_grad_input::call(grad_output, output_size, input_size, scales_h, scales_w, grad_input);
+  }
+}
+
+// aten::_upsample_nearest_exact2d_backward(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None) -> Tensor
+inline at::Tensor _upsample_nearest_exact2d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_backward::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_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_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_backward::call(grad_output, c10::fromIntArrayRefSlow(output_size), c10::fromIntArrayRefSlow(input_size), scales_h, scales_w);
+  }
+}
+
+// aten::_upsample_nearest_exact2d_backward(Tensor grad_output, SymInt[2] output_size, SymInt[4] input_size, float? scales_h=None, float? scales_w=None) -> Tensor
+inline at::Tensor _upsample_nearest_exact2d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_backward::call(grad_output, output_size, input_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_backward(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, ::std::optional<double> scales_h=::std::nullopt, ::std::optional<double> scales_w=::std::nullopt) {
+    return at::_ops::_upsample_nearest_exact2d_backward::call(grad_output, output_size, input_size, scales_h, scales_w);
+  }
+}
+
+}

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool1d_ops.h

@@ -0,0 +1,28 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Operator.h
+
+#include <tuple>
+#include <vector>
+
+// Forward declarations of any types needed in the operator signatures.
+// We can't directly include these classes because it will cause circular include dependencies.
+// This file is included by TensorBody.h, which defines the Tensor class.
+#include <ATen/core/ATen_fwd.h>
+
+namespace at {
+namespace _ops {
+
+
+struct TORCH_API adaptive_avg_pool1d {
+  using schema = at::Tensor (const at::Tensor &, at::IntArrayRef);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::adaptive_avg_pool1d")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "adaptive_avg_pool1d(Tensor self, int[1] output_size) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, at::IntArrayRef output_size);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::IntArrayRef output_size);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/align_to_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 align_to(const at::Tensor & self, at::DimnameList names);
+TORCH_API at::Tensor align_to(const at::Tensor & self, at::DimnameList order, int64_t ellipsis_idx);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/aminmax_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> aminmax(const at::Tensor & self, ::std::optional<int64_t> dim=::std::nullopt, bool keepdim=false);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> aminmax_out(at::Tensor & min, at::Tensor & max, const at::Tensor & self, ::std::optional<int64_t> dim=::std::nullopt, bool keepdim=false);
+TORCH_API ::std::tuple<at::Tensor &,at::Tensor &> aminmax_outf(const at::Tensor & self, ::std::optional<int64_t> dim, bool keepdim, at::Tensor & min, at::Tensor & max);
+
+} // namespace meta
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/argmax_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 argmax(const at::Tensor & self, ::std::optional<int64_t> dim=::std::nullopt, bool keepdim=false);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/argmax_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 argmax(const at::Tensor & self, ::std::optional<int64_t> dim=::std::nullopt, bool keepdim=false);
+TORCH_API at::Tensor & argmax_out(at::Tensor & out, const at::Tensor & self, ::std::optional<int64_t> dim=::std::nullopt, bool keepdim=false);
+TORCH_API at::Tensor & argmax_outf(const at::Tensor & self, ::std::optional<int64_t> dim, bool keepdim, at::Tensor & out);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/as_strided_scatter_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 & as_strided_scatter_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & src, at::IntArrayRef size, at::IntArrayRef stride, ::std::optional<int64_t> storage_offset=::std::nullopt);
+TORCH_API at::Tensor & as_strided_scatter_outf(const at::Tensor & self, const at::Tensor & src, at::IntArrayRef size, at::IntArrayRef stride, ::std::optional<int64_t> storage_offset, at::Tensor & out);
+TORCH_API at::Tensor & as_strided_scatter_symint_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & src, c10::SymIntArrayRef size, c10::SymIntArrayRef stride, ::std::optional<c10::SymInt> storage_offset=::std::nullopt);
+TORCH_API at::Tensor & as_strided_scatter_symint_outf(const at::Tensor & self, const at::Tensor & src, c10::SymIntArrayRef size, c10::SymIntArrayRef stride, ::std::optional<c10::SymInt> storage_offset, at::Tensor & out);
+
+} // namespace compositeexplicitautograd
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cdist.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/cdist_ops.h>
+
+namespace at {
+
+
+// aten::cdist(Tensor x1, Tensor x2, float p=2, int? compute_mode=None) -> Tensor
+inline at::Tensor cdist(const at::Tensor & x1, const at::Tensor & x2, double p=2, ::std::optional<int64_t> compute_mode=::std::nullopt) {
+    return at::_ops::cdist::call(x1, x2, p, compute_mode);
+}
+
+}

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cholesky_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 cholesky_solve(const at::Tensor & self, const at::Tensor & input2, bool upper=false);
+TORCH_API at::Tensor & cholesky_solve_out(const at::Tensor & self, const at::Tensor & input2, bool upper, at::Tensor & out);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,35 @@
+#pragma once
+
+// @generated by torchgen/gen.py from Function.h
+
+#include <ATen/Context.h>
+#include <ATen/DeviceGuard.h>
+#include <ATen/TensorUtils.h>
+#include <ATen/TracerMode.h>
+#include <ATen/core/Generator.h>
+#include <ATen/core/Reduction.h>
+#include <ATen/core/Tensor.h>
+#include <c10/core/Scalar.h>
+#include <c10/core/Storage.h>
+#include <c10/core/TensorOptions.h>
+#include <c10/util/Deprecated.h>
+#include <optional>
+
+
+
+#include <ATen/ops/cumulative_trapezoid_ops.h>
+
+namespace at {
+
+
+// aten::cumulative_trapezoid.x(Tensor y, Tensor x, *, int dim=-1) -> Tensor
+inline at::Tensor cumulative_trapezoid(const at::Tensor & y, const at::Tensor & x, int64_t dim=-1) {
+    return at::_ops::cumulative_trapezoid_x::call(y, x, dim);
+}
+
+// aten::cumulative_trapezoid.dx(Tensor y, *, Scalar dx=1, int dim=-1) -> Tensor
+inline at::Tensor cumulative_trapezoid(const at::Tensor & y, const at::Scalar & dx=1, int64_t dim=-1) {
+    return at::_ops::cumulative_trapezoid_dx::call(y, dx, dim);
+}
+
+}

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/empty_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 empty(at::IntArrayRef size, at::TensorOptions options={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty(at::IntArrayRef size, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory, ::std::optional<at::MemoryFormat> memory_format);
+TORCH_API at::Tensor empty_symint(c10::SymIntArrayRef size, at::TensorOptions options={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_symint(c10::SymIntArrayRef size, ::std::optional<at::ScalarType> dtype, ::std::optional<at::Layout> layout, ::std::optional<at::Device> device, ::std::optional<bool> pin_memory, ::std::optional<at::MemoryFormat> memory_format);
+
+} // namespace cuda
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/empty_like_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>
+
+
+namespace at {
+namespace native {
+TORCH_API at::Tensor empty_like(const at::Tensor & self, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor & empty_like_out(const at::Tensor & self, ::std::optional<at::MemoryFormat> memory_format, at::Tensor & out);
+TORCH_API at::Tensor empty_like_nested(const at::Tensor & self, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_like_sparse_coo(const at::Tensor & self, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_like_sparse_csr(const at::Tensor & self, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+TORCH_API at::Tensor empty_like_quantized(const at::Tensor & self, ::std::optional<at::ScalarType> dtype={}, ::std::optional<at::Layout> layout={}, ::std::optional<at::Device> device={}, ::std::optional<bool> pin_memory={}, ::std::optional<at::MemoryFormat> memory_format=::std::nullopt);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/erf_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 erf(const at::Tensor & self);
+TORCH_API at::Tensor & erf_out(at::Tensor & out, const at::Tensor & self);
+TORCH_API at::Tensor & erf_outf(const at::Tensor & self, at::Tensor & out);
+TORCH_API at::Tensor & erf_(at::Tensor & self);
+
+} // namespace meta
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/exp2_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 exp2 {
+  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::exp2")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "exp2(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 exp2_ {
+  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::exp2_")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "exp2_(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 exp2_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::exp2")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "exp2.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Tensor & out);
+};
+
+}} // namespace at::_ops

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/expm1_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 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 cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_max_pool2d_backward_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 fractional_max_pool2d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices);
+TORCH_API at::Tensor & fractional_max_pool2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices);
+TORCH_API at::Tensor & fractional_max_pool2d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & indices, at::Tensor & grad_input);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fractional_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> fractional_max_pool2d(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef output_size, const at::Tensor & random_samples);
+
+} // namespace compositeexplicitautogradnonfunctional
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/histogram.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/histogram_ops.h>
+
+namespace at {
+
+
+// aten::histogram.bins_tensor_out(Tensor self, Tensor bins, *, Tensor? weight=None, bool density=False, Tensor(a!) hist, Tensor(b!) bin_edges) -> (Tensor(a!) hist, Tensor(b!) bin_edges)
+inline ::std::tuple<at::Tensor &,at::Tensor &> histogram_out(at::Tensor & hist, at::Tensor & bin_edges, const at::Tensor & self, const at::Tensor & bins, const ::std::optional<at::Tensor> & weight={}, bool density=false) {
+    return at::_ops::histogram_bins_tensor_out::call(self, bins, weight, density, hist, bin_edges);
+}
+// aten::histogram.bins_tensor_out(Tensor self, Tensor bins, *, Tensor? weight=None, bool density=False, Tensor(a!) hist, Tensor(b!) bin_edges) -> (Tensor(a!) hist, Tensor(b!) bin_edges)
+inline ::std::tuple<at::Tensor &,at::Tensor &> histogram_outf(const at::Tensor & self, const at::Tensor & bins, const ::std::optional<at::Tensor> & weight, bool density, at::Tensor & hist, at::Tensor & bin_edges) {
+    return at::_ops::histogram_bins_tensor_out::call(self, bins, weight, density, hist, bin_edges);
+}
+
+// aten::histogram.bins_tensor(Tensor self, Tensor bins, *, Tensor? weight=None, bool density=False) -> (Tensor hist, Tensor bin_edges)
+inline ::std::tuple<at::Tensor,at::Tensor> histogram(const at::Tensor & self, const at::Tensor & bins, const ::std::optional<at::Tensor> & weight={}, bool density=false) {
+    return at::_ops::histogram_bins_tensor::call(self, bins, weight, density);
+}
+
+// aten::histogram.bin_ct_out(Tensor self, int bins=100, *, float[]? range=None, Tensor? weight=None, bool density=False, Tensor(a!) hist, Tensor(b!) bin_edges) -> (Tensor(a!) hist, Tensor(b!) bin_edges)
+inline ::std::tuple<at::Tensor &,at::Tensor &> histogram_out(at::Tensor & hist, at::Tensor & bin_edges, const at::Tensor & self, int64_t bins=100, ::std::optional<at::ArrayRef<double>> range=::std::nullopt, const ::std::optional<at::Tensor> & weight={}, bool density=false) {
+    return at::_ops::histogram_bin_ct_out::call(self, bins, range, weight, density, hist, bin_edges);
+}
+// aten::histogram.bin_ct_out(Tensor self, int bins=100, *, float[]? range=None, Tensor? weight=None, bool density=False, Tensor(a!) hist, Tensor(b!) bin_edges) -> (Tensor(a!) hist, Tensor(b!) bin_edges)
+inline ::std::tuple<at::Tensor &,at::Tensor &> histogram_outf(const at::Tensor & self, int64_t bins, ::std::optional<at::ArrayRef<double>> range, const ::std::optional<at::Tensor> & weight, bool density, at::Tensor & hist, at::Tensor & bin_edges) {
+    return at::_ops::histogram_bin_ct_out::call(self, bins, range, weight, density, hist, bin_edges);
+}
+
+// aten::histogram.bin_ct(Tensor self, int bins=100, *, float[]? range=None, Tensor? weight=None, bool density=False) -> (Tensor hist, Tensor bin_edges)
+inline ::std::tuple<at::Tensor,at::Tensor> histogram(const at::Tensor & self, int64_t bins=100, ::std::optional<at::ArrayRef<double>> range=::std::nullopt, const ::std::optional<at::Tensor> & weight={}, bool density=false) {
+    return at::_ops::histogram_bin_ct::call(self, bins, range, weight, density);
+}
+
+}

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

@@ -0,0 +1,24 @@
+#pragma once
+// @generated by torchgen/gen.py from DispatchKeyFunction.h
+
+// NB: The implementing C++ file is RegisterDispatchKey.cpp
+
+// The only #includes we need are for custom classes that have defaults in the C++ API
+#include <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 & huber_loss_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & target, int64_t reduction, double delta);
+TORCH_API at::Tensor & huber_loss_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & target, int64_t reduction, double delta, at::Tensor & grad_input);
+
+} // namespace cpu
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_same_size_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 bool is_same_size(const at::Tensor & self, const at::Tensor & other);
+TORCH_API bool nested_is_same_size(const at::Tensor & self, const at::Tensor & other);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_vulkan_available.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_vulkan_available_ops.h>
+
+namespace at {
+
+
+// aten::is_vulkan_available() -> bool
+inline bool is_vulkan_available() {
+    return at::_ops::is_vulkan_available::call();
+}
+
+}

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/kl_div_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 kl_div(const at::Tensor & self, const at::Tensor & target, int64_t reduction=at::Reduction::Mean, bool log_target=false);
+
+} // namespace compositeimplicitautograd
+} // namespace at

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

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_inv_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 linalg_inv(const at::Tensor & A);
+TORCH_API at::Tensor & linalg_inv_out(const at::Tensor & A, at::Tensor & out);
+} // namespace native
+} // namespace at

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

@@ -0,0 +1,105 @@
+#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_pinv_atol_rtol_tensor {
+  using schema = at::Tensor (const at::Tensor &, const ::std::optional<at::Tensor> &, const ::std::optional<at::Tensor> &, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_pinv")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "atol_rtol_tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_pinv.atol_rtol_tensor(Tensor self, *, Tensor? atol=None, Tensor? rtol=None, bool hermitian=False) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const ::std::optional<at::Tensor> & atol, const ::std::optional<at::Tensor> & rtol, bool hermitian);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const ::std::optional<at::Tensor> & atol, const ::std::optional<at::Tensor> & rtol, bool hermitian);
+};
+
+struct TORCH_API linalg_pinv_atol_rtol_tensor_out {
+  using schema = at::Tensor & (const at::Tensor &, const ::std::optional<at::Tensor> &, const ::std::optional<at::Tensor> &, 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::linalg_pinv")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "atol_rtol_tensor_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_pinv.atol_rtol_tensor_out(Tensor self, *, Tensor? atol=None, Tensor? rtol=None, bool hermitian=False, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, const ::std::optional<at::Tensor> & atol, const ::std::optional<at::Tensor> & rtol, bool hermitian, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const ::std::optional<at::Tensor> & atol, const ::std::optional<at::Tensor> & rtol, bool hermitian, at::Tensor & out);
+};
+
+struct TORCH_API linalg_pinv_atol_rtol_float {
+  using schema = at::Tensor (const at::Tensor &, ::std::optional<double>, ::std::optional<double>, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_pinv")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "atol_rtol_float")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_pinv.atol_rtol_float(Tensor self, *, float? atol=None, float? rtol=None, bool hermitian=False) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, ::std::optional<double> atol, ::std::optional<double> rtol, bool hermitian);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, ::std::optional<double> atol, ::std::optional<double> rtol, bool hermitian);
+};
+
+struct TORCH_API linalg_pinv_atol_rtol_float_out {
+  using schema = at::Tensor & (const at::Tensor &, ::std::optional<double>, ::std::optional<double>, 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::linalg_pinv")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "atol_rtol_float_out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_pinv.atol_rtol_float_out(Tensor self, *, float? atol=None, float? rtol=None, bool hermitian=False, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, ::std::optional<double> atol, ::std::optional<double> rtol, bool hermitian, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, ::std::optional<double> atol, ::std::optional<double> rtol, bool hermitian, at::Tensor & out);
+};
+
+struct TORCH_API linalg_pinv {
+  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::linalg_pinv")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_pinv(Tensor self, float rcond, bool hermitian=False) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, double rcond, bool hermitian);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, double rcond, bool hermitian);
+};
+
+struct TORCH_API linalg_pinv_rcond_tensor {
+  using schema = at::Tensor (const at::Tensor &, const at::Tensor &, bool);
+  using ptr_schema = schema*;
+  // See Note [static constexpr char* members for windows NVCC]
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linalg_pinv")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "rcond_tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_pinv.rcond_tensor(Tensor self, Tensor rcond, bool hermitian=False) -> Tensor")
+  static at::Tensor call(const at::Tensor & self, const at::Tensor & rcond, bool hermitian);
+  static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & rcond, bool hermitian);
+};
+
+struct TORCH_API linalg_pinv_out {
+  using schema = at::Tensor & (const at::Tensor &, double, 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::linalg_pinv")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_pinv.out(Tensor self, float rcond, bool hermitian=False, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, double rcond, bool hermitian, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, double rcond, bool hermitian, at::Tensor & out);
+};
+
+struct TORCH_API linalg_pinv_out_rcond_tensor {
+  using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, 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::linalg_pinv")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out_rcond_tensor")
+  STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linalg_pinv.out_rcond_tensor(Tensor self, Tensor rcond, bool hermitian=False, *, Tensor(a!) out) -> Tensor(a!)")
+  static at::Tensor & call(const at::Tensor & self, const at::Tensor & rcond, bool hermitian, at::Tensor & out);
+  static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & rcond, bool hermitian, at::Tensor & out);
+};
+
+}} // namespace at::_ops

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

@@ -0,0 +1,24 @@
+#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 masked_scatter(const at::Tensor & self, const at::Tensor & mask, const at::Tensor & source);
+TORCH_API at::Tensor & masked_scatter_out(const at::Tensor & self, const at::Tensor & mask, const at::Tensor & source, at::Tensor & out);
+TORCH_API at::Tensor & masked_scatter__cpu(at::Tensor & self, const at::Tensor & mask, const at::Tensor & source);
+TORCH_API at::Tensor & masked_scatter__cuda(at::Tensor & self, const at::Tensor & mask, const at::Tensor & source);
+} // namespace native
+} // namespace at

.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool3d_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 max_pool3d(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, at::IntArrayRef dilation=1, bool ceil_mode=false);
+} // namespace native
+} // namespace at