keineahnung2345
keineahnung2345

PyTorch中的pyi檔案生成機制

PyTorch中的pyi檔案生成機制

  • 前言
  • pyi檔
  • 由py生成pyi.in
  • 由pyi.in生成pyi
    • torch/CMakeLists.txt
    • tools/pyi/gen_pyi.py
    • gen_pyi
      • native_functions
        • rand.names ＆ rand.names_out
        • rand.generator_with_names & rand.generator_with_names_out
        • rand
        • rand.generator
        • rand.out
        • rand.generator_out
        • add.Tensor && add.out
        • add_.Tensor && add.out
        • add.out
      • function_signatures
        • rand.names & rand.names_out
        • rand.generator_with_names & rand.generator_with_names_out
        • rand
        • rand.generator
        • rand.out
        • rand.generator_out
        • add.Tensor && add.out
        • add_.Tensor && add.out
        • add.out
      • sig_groups
        • rand.generator_with_names & rand.generator_with_names_out
        • rand.generator & rand.generator_out
        • rand.names & rand.names_out
        • rand & rand.out
        • add.Tensor & add.out
        • add & add.Tensor & add.out
      • unsorted_function_hints
        • rand
        • add
      • function_hints
        • rand
        • add
      • hinted_function_names
      • all_symbols
      • all_directive
      • env
    • gen_nn_functional
    • datapipe.pyi
    • 生成結果
  • 使用pyi做類型檢查

前言

在PyTorch中如果查找python函數的定義，十有八九會跳轉到torch/_C/_VariableFunctions.pyi這個檔案。但是如果去PyTorch的github repo上尋找這個檔案，只能找到一個跟它名字類似的torch/_C/_VariableFunctions.pyi.in，卻找不到torch/_C/_VariableFunctions.pyi這個檔案本身。

如果打開torch/_C/_VariableFunctions.pyi去看：

# @generated from torch/_C/_VariableFunctions.pyi.in

才發現原來首行就說了：它是在編譯時才由torch/_C/_VariableFunctions.pyi.in動態生成的。

本篇就是要探討PyTorch中pyi檔案的生成機制。pyi檔案的生成過程大致可分為以下兩步：

  1. 由py生成pyi.in

  2. 由pyi.in生成pyi

但在這之前，先來看一下pyi檔在Python中的作用為何。

pyi檔

首先來看一下pyi這個檔案類型的名稱由來，根據What does “i” represent in Python .pyi extension?：

The i in .pyi stands for ‘interface’.

The .pyi extension was first mentioned in this GitHub issue thread where JukkaL says:

I'd probably prefer an extension with just a single dot. It also needs to be something that is not in use (it should not be used by cython, etc.). .pys seems to be used in Windows (or was). Maybe .pyi, where i stands for an interface definition?

可以知道，pyi中的i代表的是interface。

pyi implements "stub" file (definition from Martin Fowler)

Stubs: provide canned answers to calls made during the test, usually not responding at all to anything outside what's programmed in for the test.

而它代表的涵意則是stub(樁/存根)，詳見樁 (計算機)：

樁[1]（Stub / Method Stub ）是指用來替換一部分功能的程序段。樁程序可以用來模擬已有程序的行為（比如一個遠端機器的過程）或是對將要開發的代碼的一種臨時替代。因此，打樁技術在程序移植、分布式計算、通用軟體開發和測試中用處很大。

正如pyi文件是干嘛的？（一文读懂Python的存根文件和类型检查）中所說，而pyi檔的作用只是在IDE中給type hint的，並不是必須的。

在PyTorch中也是一樣，torch/_C/_VariableFunctions.pyi僅用於類型提示。Python函數與C++函數的關聯實際上是由torch/csrc/autograd/generated/python_torch_functions_i.cpp所指定，而該檔案也是在編譯時自動生成的，詳見PyTorch中的python_torch_functions_i.cpp檔案生成機制。

由py生成pyi.in

PyTorch源碼中有以下.pyi.in檔：

torch/_C/__init__.pyi.in
torch/_C/_nn.pyi.in
torch/_C/return_types.pyi.in
torch/_C/_VariableFunctions.pyi.in
torch/nn/functional.pyi.in
torch/utils/data/datapipes/datapipe.pyi.in

根據torch/nn/functional.pyi.in中的注釋：

# These stubs were generated by running stubgen (`stubgen --parse-only functional.py`), followed by manual cleaning.

functional.pyi.in是用mypy的stubgen工具由functional.py生成後手動編輯而成的。

試著自己對torch/nn/functional.py跑跑看stubgen，首先把functional.py這個檔案複製到一個合適的地方，然後下：

stubgen functional.py

如果出現以下跟import相關的錯誤，先手動把對應的行數注釋掉就好：

Critical error during semantic analysis: functional.py:23: error: No parent module -- cannot perform relative import
functional.py:24: error: No parent module -- cannot perform relative import

先只關注以下這一段：

def fractional_max_pool2d_with_indices(
    input: Tensor, kernel_size: BroadcastingList2[int],
    output_size: Optional[BroadcastingList2[int]] = None,
    output_ratio: Optional[BroadcastingList2[float]] = None,
    return_indices: bool = False,
    _random_samples: Optional[Tensor] = None
) -> Tuple[Tensor, Tensor]:
    # ...

fractional_max_pool2d = boolean_dispatch(
    arg_name="return_indices",
    arg_index=4,
    default=False,
    if_true=fractional_max_pool2d_with_indices,
    if_false=_fractional_max_pool2d,
    module_name=__name__,
    func_name="fractional_max_pool2d",
)

生成的functional.pyi裡對應的內容：

# ...
def fractional_max_pool2d_with_indices(input: Tensor, kernel_size: BroadcastingList2[int], output_size: Optional[BroadcastingList2[int]] = ..., output_ratio: Optional[BroadcastingList2[float]] = ..., return_indices: bool = ..., _random_samples: Optional[Tensor] = ...) -> Tuple[Tensor, Tensor]: ...

fractional_max_pool2d: Incomplete
# ...

fractional_max_pool2d_with_indices這個函數的簽名與原本的幾乎一致，而fractional_max_pool2d則因為無法推斷被標注為Incomplete。

照理說.pyi.in檔是由.py檔生成的，但是torch/_C目錄下的.pyi.in檔都沒有對應的.py檔，推測是由多個.py檔合併到同一個.pyi.in檔而來的。

由pyi.in生成pyi

一般來說.pyi檔是由stubgen生成的，但在PyTorch中則是先用stubgen生成並手動編輯後得到pyi.in檔，然後再利用Python腳本由.pyi.in檔生成的。

torch/CMakeLists.txt

torch/CMakeLists.txt

新增一個名為torch_python_stubs的custom target，依賴於如下的pyi檔。(關於add_custom_target和接下來會看到的add_custom_command詳見cmake的add_custom_command及add_custom_target。)

add_custom_target(torch_python_stubs DEPENDS
    "${TORCH_SRC_DIR}/_C/__init__.pyi"
    "${TORCH_SRC_DIR}/_C/_VariableFunctions.pyi"
    "${TORCH_SRC_DIR}/nn/functional.pyi"
    "${TORCH_SRC_DIR}/utils/data/datapipes/datapipe.pyi"
)

查看如下add_custom_command的OUTPUT參數，可以知道這個custom command正是用於生成torch_python_stubs所依賴的前三個pyi檔。至於剩下的datapipe.pyi是如何生成的，詳見datapipe.pyi章節。

file(GLOB_RECURSE torchgen_python "${PROJECT_SOURCE_DIR}/torchgen/*.py")
file(GLOB_RECURSE autograd_python "${TOOLS_PATH}/autograd/*.py")
file(GLOB_RECURSE pyi_python "${TOOLS_PATH}/pyi/*.py")
add_custom_command(
    OUTPUT
    "${TORCH_SRC_DIR}/_C/__init__.pyi"
    "${TORCH_SRC_DIR}/_C/_VariableFunctions.pyi"
    "${TORCH_SRC_DIR}/nn/functional.pyi"
    COMMAND
    "${PYTHON_EXECUTABLE}" -mtools.pyi.gen_pyi
      --native-functions-path "aten/src/ATen/native/native_functions.yaml"
      --tags-path "aten/src/ATen/native/tags.yaml"
      --deprecated-functions-path "tools/autograd/deprecated.yaml"
    DEPENDS
      "${TORCH_SRC_DIR}/_C/__init__.pyi.in"
      "${TORCH_SRC_DIR}/_C/_VariableFunctions.pyi.in"
      "${TORCH_SRC_DIR}/nn/functional.pyi.in"
      "${TORCH_ROOT}/aten/src/ATen/native/native_functions.yaml"
      "${TORCH_ROOT}/aten/src/ATen/native/tags.yaml"
      "${TORCH_ROOT}/tools/autograd/deprecated.yaml"
      ${pyi_python}
      ${autograd_python}
      ${torchgen_python}
    WORKING_DIRECTORY
    "${TORCH_ROOT}"
)

這一段的入口是add_custom_command中的COMMAND，它透過"${PYTHON_EXECUTABLE}" -mtools.pyi.gen_pyi調用tools/pyi/gen_pyi.py，輸入則是DEPENDS區塊中寫的_C/__init__.pyi.in, _C/_VariableFunctions.pyi.in 及nn/functional.pyi.in，程序執行完後會生成OUTPUT區塊中寫的三個pyi檔。

torch/_C/_nn.pyi和torch/_C/return_types.pyi也是由tools/pyi/gen_pyi.py生成的，為什麼沒寫在add_custom_target和add_custom_command的DEPENDS和OUTPUT裡?

新增一個名為torch_python的shared library，運行過後會生成build/lib/libtorch_python.so。

add_library(torch_python SHARED ${TORCH_PYTHON_SRCS})

接著宣告torch_python依賴於torch_python_stubs這個custom target。

add_dependencies(torch_python torch_python_stubs)

在非MacOS的系統上都會建構一個名為nnapi_backend的library，它的依賴中就有torch_python。

# Skip building this library under MacOS, since it is currently failing to build on Mac
# Github issue #61930
if(NOT ${CMAKE_SYSTEM_NAME} MATCHES "Darwin")
  # Add Android Nnapi delegate library
  add_library(nnapi_backend SHARED
          ${TORCH_SRC_DIR}/csrc/jit/backends/nnapi/nnapi_backend_lib.cpp
          ${TORCH_SRC_DIR}/csrc/jit/backends/nnapi/nnapi_backend_preprocess.cpp
          )
  # Pybind11 requires explicit linking of the torch_python library
  target_link_libraries(nnapi_backend PRIVATE torch torch_python pybind::pybind11)
endif()

總結一下，就是有nnapi_backend -> torch_python -> torch_python_stubs -> torch/_C/__init__.pyi, torch/_C/_VariableFunctions.pyi, torch/nn/functional.pyi間的層層依賴，所以要建構nnapi_backend這個library時才會調用tools/pyi/gen_pyi.py去生成.pyi檔。

tools/pyi/gen_pyi.py

CMakeLists.txt中透過"${PYTHON_EXECUTABLE}" -mtools.pyi.gen_pyi調用了tools/pyi/gen_pyi.py， 它的功用是由.pyi.in檔生成.pyi檔。

def main() -> None:
    parser = argparse.ArgumentParser(description="Generate type stubs for PyTorch")
    parser.add_argument(
        "--native-functions-path",
        metavar="NATIVE",
        default="aten/src/ATen/native/native_functions.yaml",
        help="path to native_functions.yaml",
    )
    parser.add_argument(
        "--tags-path",
        metavar="TAGS",
        default="aten/src/ATen/native/tags.yaml",
        help="path to tags.yaml",
    )
    parser.add_argument(
        "--deprecated-functions-path",
        metavar="DEPRECATED",
        default="tools/autograd/deprecated.yaml",
        help="path to deprecated.yaml",
    )
    parser.add_argument(
        "--out", metavar="OUT", default=".", help="path to output directory"
    )
    args = parser.parse_args()
    fm = FileManager(install_dir=args.out, template_dir=".", dry_run=False)
    gen_pyi(
        args.native_functions_path, args.tags_path, args.deprecated_functions_path, fm
    )


if __name__ == "__main__":
    main()

gen_pyi.py中的注釋：

- We start off with a hand-written __init__.pyi.in file.  This
  file contains type definitions for everything we cannot automatically
  generate, including pure Python definitions directly in __init__.py
  (the latter case should be pretty rare).

- We go through automatically bound functions based on the
  type information recorded in native_functions.yaml and
  generate type hints for them (generate_type_hints)

native_functions.yaml中記錄了自動綁定函數（automatically bound functions，猜測是Python與C++函數的綁定）的類型資訊，gen_pyi.py會依據這些類型資訊用generate_type_hints函數（待會會在unsorted_function_hints一節出現）生成類型提示。

gen_pyi

tools/pyi/gen_pyi.py

這個函數的功用是由_C/__init__.pyi.in, _C/_VariableFunctions.pyi.in及torch/_C/return_types.pyi.in生成_C/__init__.pyi, _C/_VariableFunctions.pyi, torch/_VF.pyi和torch/return_types.pyi。

def gen_pyi(
    native_yaml_path: str,
    tags_yaml_path: str,
    deprecated_yaml_path: str,
    fm: FileManager,
) -> None:
    """gen_pyi()

    This function generates a pyi file for torch.
    """
    
    # ...

前三個參數預設為：

  • native_yaml_path：aten/src/ATen/native/native_functions.yaml
  • tags_yaml_path：aten/src/ATen/native/tags.yaml
  • deprecated_yaml_path：tools/autograd/deprecated.yaml

fm建構子中的兩個參數如下：

  • install_dir：args.out，也就是’.’
  • template_dir：‘.’

native_functions

解析native_functions.yaml與tags.yaml，得到native_functions變數：

    native_functions = parse_native_yaml(
        native_yaml_path, tags_yaml_path
    ).native_functions
    native_functions = list(filter(should_generate_py_binding, native_functions))

native_functions是一個NativeFunction的列表，表示aten命名空間裡的函數，其第零個元素如下：

NativeFunction(
    namespace='aten', 
    func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='_cast_Byte', inplace=False, dunder_method=False, functional_overload=False), overload_name=''), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='non_blocking', type=BaseType(name=), default='False', annotation=None),), pre_tensor_options_kwarg_only=(), tensor_options=None, post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), 
    use_const_ref_for_mutable_tensors=False, 
    device_guard=True, 
    device_check=, 
    python_module=None, 
    category_override=None, 
    variants={}, 
    manual_kernel_registration=False, 
    manual_cpp_binding=False, 
    loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=9), 
    autogen=[], 
    ufunc_inner_loop={}, 
    structured=False, 
    structured_delegate=None, 
    structured_inherits=None, 
    precomputed=None, 
    cpp_no_default_args=set(), 
    is_abstract=False, 
    has_composite_implicit_autograd_kernel=True, 
    has_composite_implicit_autograd_nested_tensor_kernel=False, 
    has_composite_explicit_autograd_kernel=False, 
    has_composite_explicit_autograd_non_functional_kernel=False, 
    tags=set()
    )

代表rand函數的元素如下，aten::rand函數有六種overload name，分別為names, generator_with_names, 空字串, generator, out, generator_out。可與native_functions.yaml交互參看：

rand.names ＆ rand.names_out
- func: rand.names(SymInt[] size, *, Dimname[]? names, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
  device_check: NoCheck
  device_guard: False
  dispatch:
    CompositeExplicitAutograd: rand
  autogen: rand.names_out
  tags: nondeterministic_seeded

yaml檔中的autogen欄位中有rand.names_out，對照native_functions中的元素，可以發現NativeFunction的autogen成員也有一個overload_name為names_out的OperatorName。

NativeFunction(
    namespace='aten', 
    func=FunctionSchema(
        name=OperatorName(
            name=BaseOperatorName(
                base='rand', inplace=False, dunder_method=False, functional_overload=False), 
            overload_name='names'), 
        arguments=Arguments(
            pre_self_positional=(), 
            self_arg=None, 
            post_self_positional=(
                Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), 
            pre_tensor_options_kwarg_only=(Argument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, annotation=None),), 
            tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), 
            post_tensor_options_kwarg_only=(), out=()), 
        returns=(Return(name=None, type=BaseType(name=), annotation=None),)), 
    use_const_ref_for_mutable_tensors=False, 
    device_guard=False, 
    device_check=, 
    python_module=None, 
    category_override=None, 
    variants={}, 
    manual_kernel_registration=False, 
    manual_cpp_binding=False, 
    loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4254), 
    autogen=[OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), overload_name='names_out')], 
    ufunc_inner_loop={}, 
    structured=False, 
    structured_delegate=None, 
    structured_inherits=None, 
    precomputed=None, 
    cpp_no_default_args=set(), 
    is_abstract=True, 
    has_composite_implicit_autograd_kernel=False,
    has_composite_implicit_autograd_nested_tensor_kernel=False, 
    has_composite_explicit_autograd_kernel=True, 
    has_composite_explicit_autograd_non_functional_kernel=False, 
    tags={'nondeterministic_seeded'})
rand.generator_with_names & rand.generator_with_names_out
- func: rand.generator_with_names(SymInt[] size, *, Generator? generator, Dimname[]? names, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
  device_check: NoCheck
  device_guard: False
  tags: nondeterministic_seeded
  dispatch:
    CompositeExplicitAutograd: rand
  autogen: rand.generator_with_names_out

yaml檔中的autogen欄位中有rand.generator_with_names_out，對照下面，可以發現NativeFunction的autogen成員也有一個overload_name為generator_with_names_out的OperatorName。

NativeFunction(
    namespace='aten', 
    func=FunctionSchema(
        name=OperatorName(
            name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
            overload_name='generator_with_names'), 
        arguments=Arguments(
            pre_self_positional=(), 
            self_arg=None, 
            post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), 
            pre_tensor_options_kwarg_only=(Argument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, annotation=None), Argument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, annotation=None)), 
            tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), 
            post_tensor_options_kwarg_only=(), 
            out=()), 
        returns=(Return(name=None, type=BaseType(name=), annotation=None),)), 
    use_const_ref_for_mutable_tensors=False, 
    device_guard=False, 
    device_check=, 
    python_module=None, 
    category_override=None, 
    variants={}, 
    manual_kernel_registration=False, 
    manual_cpp_binding=False, 
    loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4262), 
    autogen=[OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), overload_name='generator_with_names_out')], 
    ufunc_inner_loop={}, 
    structured=False, 
    structured_delegate=None, 
    structured_inherits=None, 
    precomputed=None, 
    cpp_no_default_args=set(), 
    is_abstract=True, 
    has_composite_implicit_autograd_kernel=False, 
    has_composite_implicit_autograd_nested_tensor_kernel=False, 
    has_composite_explicit_autograd_kernel=True, 
    has_composite_explicit_autograd_non_functional_kernel=False, 
    tags={'nondeterministic_seeded'})
rand
- func: rand(SymInt[] size, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
  tags: nondeterministic_seeded
  dispatch:
    CompositeExplicitAutograd: rand

NativeFunction(
    namespace='aten', 
    func=FunctionSchema(
        name=OperatorName(
            name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
            overload_name=''), 
            arguments=Arguments(
                pre_self_positional=(), 
                self_arg=None, 
                post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), 
                pre_tensor_options_kwarg_only=(), 
                tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), 
                post_tensor_options_kwarg_only=(), 
                out=()), 
            returns=(Return(name=None, type=BaseType(name=), annotation=None),)), 
            use_const_ref_for_mutable_tensors=False, 
            device_guard=True, 
            device_check=, 
            python_module=None, 
            category_override=None, 
            variants={}, 
            manual_kernel_registration=False, 
            manual_cpp_binding=False, 
            loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4270), 
            autogen=[], 
            ufunc_inner_loop={}, 
            structured=False, 
            structured_delegate=None, 
            structured_inherits=None, 
            precomputed=None, 
            cpp_no_default_args=set(), 
            is_abstract=True, 
            has_composite_implicit_autograd_kernel=False, 
            has_composite_implicit_autograd_nested_tensor_kernel=False, 
            has_composite_explicit_autograd_kernel=True, 
            has_composite_explicit_autograd_non_functional_kernel=False, 
            tags={'nondeterministic_seeded'})
rand.generator
- func: rand.generator(SymInt[] size, *, Generator? generator, ScalarType? dtype=None, Layout? layout=None, Device? device=None, bool? pin_memory=None) -> Tensor
  tags: nondeterministic_seeded
  dispatch:
    CompositeExplicitAutograd: rand

NativeFunction(
    namespace='aten', 
    func=FunctionSchema(
        name=OperatorName(
            name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
            overload_name='generator'), 
            arguments=Arguments(
                pre_self_positional=(), 
                self_arg=None, 
                post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), 
                pre_tensor_options_kwarg_only=(Argument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, annotation=None),), 
                tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), 
                post_tensor_options_kwarg_only=(), 
                out=()), 
            returns=(Return(name=None, type=BaseType(name=), annotation=None),)), 
            use_const_ref_for_mutable_tensors=False, 
            device_guard=True, 
            device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4275), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'})
rand.out
- func: rand.out(SymInt[] size, *, Tensor(a!) out) -> Tensor(a!)
  tags: nondeterministic_seeded
  dispatch:
    CompositeExplicitAutograd: rand_out

NativeFunction(
    namespace='aten', 
    func=FunctionSchema(
        name=OperatorName(
            name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
            overload_name='out'), 
            arguments=Arguments(
                pre_self_positional=(), 
                self_arg=None, 
                post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), 
                pre_tensor_options_kwarg_only=(), 
                tensor_options=None, 
                post_tensor_options_kwarg_only=(), 
                out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), 
            returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), 
            use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4280), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'})
rand.generator_out
- func: rand.generator_out(SymInt[] size, *, Generator? generator, Tensor(a!) out) -> Tensor(a!)
  tags: nondeterministic_seeded

NativeFunction(
    namespace='aten', 
    func=FunctionSchema(
        name=OperatorName(
            name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
            overload_name='generator_out'), 
            arguments=Arguments(
                pre_self_positional=(), 
                self_arg=None, 
                post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), 
                pre_tensor_options_kwarg_only=(Argument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, annotation=None),), 
                tensor_options=None, 
                post_tensor_options_kwarg_only=(), 
                out=(Argument(name='out', 
                type=BaseType(name=), 
                default=None, 
                annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), 
            returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), 
            use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4285), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=False, has_composite_implicit_autograd_kernel=True, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'})

因為rand.names和rand.generator_with_names會生成對應的out版本的函數，所以由native_functions.yaml裡六個rand相關函數最後可以生成C++ aten命名空間裡的八個函數。

add.Tensor && add.out

傳入self及other，回傳結果的add函數。

- func: add.Tensor(Tensor self, Tensor other, *, Scalar alpha=1) -> Tensor
  device_check: NoCheck   # TensorIterator
  structured_delegate: add.out
  variants: function, method
  dispatch:
    SparseCPU, SparseCUDA: add_sparse
    SparseCsrCPU, SparseCsrCUDA: add_sparse_csr
    MkldnnCPU: mkldnn_add
    ZeroTensor: add_zerotensor
    NestedTensorCPU, NestedTensorCUDA: NestedTensor_add_Tensor
  tags: [canonical, pointwise]

NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), 
    overload_name='Tensor'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={, }, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=497), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), 
    overload_name='out'), structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise', 'canonical'})
add_.Tensor && add.out

直接修改self參數的inplace版本。

- func: add_.Tensor(Tensor(a!) self, Tensor other, *, Scalar alpha=1) -> Tensor(a!)
  device_check: NoCheck   # TensorIterator
  variants: method
  structured_delegate: add.out
  dispatch:
    SparseCPU, SparseCUDA: add_sparse_
    SparseCsrCPU, SparseCsrCUDA: add_sparse_csr_
    MkldnnCPU: mkldnn_add_
    NestedTensorCPU, NestedTensorCUDA: NestedTensor_add__Tensor
  tags: pointwise

根據pytorch native README.md：

Tensor(a!) - members of a may be written to thus mutating the underlying data.

Tensor(a!) self這個寫法表示self參數同時是入參也是出參。

NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=True, dunder_method=False, functional_overload=False), 
    overload_name='Tensor'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=()))), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=509), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), 
    overload_name='out'), structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise'})
add.out

有出參out版本的add函數。

- func: add.out(Tensor self, Tensor other, *, Scalar alpha=1, Tensor(a!) out) -> Tensor(a!)
  device_check: NoCheck   # TensorIterator
  structured: True
  structured_inherits: TensorIteratorBase
  ufunc_inner_loop:
    Generic: add (AllAndComplex, BFloat16, Half, ComplexHalf)
    ScalarOnly: add (Bool)
  dispatch:
    SparseCPU: add_out_sparse_cpu
    SparseCUDA: add_out_sparse_cuda
    SparseCsrCPU: add_out_sparse_csr_cpu
    SparseCsrCUDA: add_out_sparse_csr_cuda
    MkldnnCPU: mkldnn_add_out
    MPS: add_out_mps
  tags: pointwise

NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), 
    overload_name='out'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=520), autogen=[], ufunc_inner_loop={: UfuncInnerLoop(name='add', supported_dtypes=, ufunc_key=), : UfuncInnerLoop(name='add', supported_dtypes=, ufunc_key=)}, structured=True, structured_delegate=None, structured_inherits='TensorIteratorBase', precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise'})

function_signatures

    function_signatures = load_signatures(
        native_functions, deprecated_yaml_path, method=False, pyi=True
    )

function_signatures是一個PythonSignatureNativeFunctionPair的列表，其第零個元素如下：

PythonSignatureNativeFunctionPair(
    signature=PythonSignature(
        name='_cast_Byte', 
        input_args=(PythonArgument(name='self', type=BaseType(name=), default=None, default_init=None), PythonArgument(name='non_blocking', type=BaseType(name=), default='False', default_init=None)), 
        input_kwargs=(), 
        output_args=None, 
        returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=None),)), 
        tensor_options_args=(), 
        method=False), 
    function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='_cast_Byte', inplace=False, dunder_method=False, functional_overload=False), overload_name=''), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='non_blocking', type=BaseType(name=), default='False', annotation=None),), pre_tensor_options_kwarg_only=(), tensor_options=None, post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=9), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=False, has_composite_implicit_autograd_kernel=True, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags=set())
    )

代表rand函數的元素如下，共六個，可與剛才的native_functions一一對應：

rand.names & rand.names_out
PythonSignatureNativeFunctionPair(signature=PythonSignature(name='rand', input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), input_kwargs=(PythonArgument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, default_init=None),), output_args=None, returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=None),)), tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), method=False), function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='names'), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, annotation=None),), tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=False, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4254), 
autogen=[OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), overload_name='names_out')], 
ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}))

注意names會生成names_out函數。

rand.generator_with_names & rand.generator_with_names_out
PythonSignatureNativeFunctionPair(signature=PythonSignature(name='rand', input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), input_kwargs=(PythonArgument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, default_init=None), PythonArgument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, default_init=None)), output_args=None, returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=None),)), tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), method=False), function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='generator_with_names'), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, annotation=None), Argument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, annotation=None)), tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=False, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4262), 
autogen=[OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), overload_name='generator_with_names_out')], 
ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}))

注意generator_with_names會生成generator_with_names_out函數。

rand
PythonSignatureNativeFunctionPair(signature=PythonSignature(name='rand', input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), input_kwargs=(), output_args=None, returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=None),)), tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), method=False), function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name=''), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(), tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4270), 
autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}))
rand.generator
PythonSignatureNativeFunctionPair(signature=PythonSignature(name='rand', input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), input_kwargs=(PythonArgument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, default_init=None),), output_args=None, returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=None),)), tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), method=False), function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='generator'), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, annotation=None),), tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4275), 
autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}))
rand.out
PythonSignatureNativeFunctionPair(signature=PythonSignature(name='rand', input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), input_kwargs=(), output_args=PythonOutArgument(name='out', type=BaseType(name=), default='None', default_init=None, outputs=(PythonArgument(name='out', type=BaseType(name=), default=None, default_init=None),)), returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), method=False), function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='out'), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(), tensor_options=None, post_tensor_options_kwarg_only=(), out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4280), 
autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}))
rand.generator_out
PythonSignatureNativeFunctionPair(signature=PythonSignature(name='rand', input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), input_kwargs=(PythonArgument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, default_init=None),), output_args=PythonOutArgument(name='out', type=BaseType(name=), default='None', default_init=None, outputs=(PythonArgument(name='out', type=BaseType(name=), default=None, default_init=None),)), returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), method=False), function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='generator_out'), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4285), 
autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=False, has_composite_implicit_autograd_kernel=True, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}))

代表add函數的元素如下，共三個，也可與剛才的native_functions一一對應。

add.Tensor && add.out
PythonSignatureNativeFunctionPair(signature=PythonSignature(name='add', input_args=(PythonArgument(name='self', type=BaseType(name=), default=None, default_init=None), PythonArgument(name='other', type=BaseType(name=), default=None, default_init=None)), input_kwargs=(PythonArgument(name='alpha', type=BaseType(name=), default='1', default_init=None),), output_args=None, returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=None),)), tensor_options_args=(), method=False), function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='Tensor'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={, }, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=497), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='out'), structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise', 'canonical'}))
add_.Tensor && add.out
PythonSignatureNativeFunctionPair(signature=PythonSignature(name='add_', input_args=(PythonArgument(name='self', type=BaseType(name=), default=None, default_init=None), PythonArgument(name='other', type=BaseType(name=), default=None, default_init=None)), input_kwargs=(PythonArgument(name='alpha', type=BaseType(name=), default='1', default_init=None),), output_args=None, returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), tensor_options_args=(), method=False), function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=True, dunder_method=False, functional_overload=False), overload_name='Tensor'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=()))), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=509), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='out'), structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise'}))
add.out
PythonSignatureNativeFunctionPair(signature=PythonSignature(name='add', input_args=(PythonArgument(name='self', type=BaseType(name=), default=None, default_init=None), PythonArgument(name='other', type=BaseType(name=), default=None, default_init=None)), input_kwargs=(PythonArgument(name='alpha', type=BaseType(name=), default='1', default_init=None),), output_args=PythonOutArgument(name='out', type=BaseType(name=), default='None', default_init=None, outputs=(PythonArgument(name='out', type=BaseType(name=), default=None, default_init=None),)), returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), tensor_options_args=(), method=False), function=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='out'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=520), autogen=[], ufunc_inner_loop={: UfuncInnerLoop(name='add', supported_dtypes=, ufunc_key=), : UfuncInnerLoop(name='add', supported_dtypes=, ufunc_key=)}, structured=True, structured_delegate=None, structured_inherits='TensorIteratorBase', precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise'}))

sig_groups

sig_groups是一個PythonSignatureGroup的列表，PythonSignatureGroup則是由PythonSignature和NativeFunction組成的pair。

PythonSignatureGroup跟PythonSignatureNativeFunctionPair比起來多了一個outplace。

    sig_groups = get_py_torch_functions(function_signatures)

sig_groups的第零個元素如下：

PythonSignatureGroup(
    signature=PythonSignature(name='__and__', input_args=(PythonArgument(name='self', type=BaseType(name=), default=None, default_init=None), PythonArgument(name='other', type=BaseType(name=), default=None, default_init=None)), input_kwargs=(), output_args=None, returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=None),)), tensor_options_args=(), method=False), 
    base=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='and', inplace=False, dunder_method=True, functional_overload=False), overload_name='Tensor'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(), tensor_options=None, post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={, }, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=7635), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=False, has_composite_implicit_autograd_kernel=True, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags=set()), 
    outplace=None)

代表rand函數的四個元素如下。原先的八個函數依據有沒有out被整理成兩兩一對，共四對。

rand.generator_with_names & rand.generator_with_names_out
PythonSignatureGroup(
    signature=PythonSignature(
        name='rand', 
        input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), 
        input_kwargs=(PythonArgument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, default_init=None), PythonArgument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, default_init=None)), 
        output_args=None, 
        returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=None),)), 
        tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), 
        method=False), 
    base=NativeFunction(
        namespace='aten', 
        func=FunctionSchema(
            name=OperatorName(
                name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
                overload_name='generator_with_names'), 
            arguments=Arguments(
                pre_self_positional=(), 
                self_arg=None, 
                post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), 
                pre_tensor_options_kwarg_only=(Argument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, annotation=None), Argument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, annotation=None)), 
                tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), 
                post_tensor_options_kwarg_only=(), 
                out=()), 
            returns=(Return(name=None, type=BaseType(name=), annotation=None),)), 
        use_const_ref_for_mutable_tensors=False, 
        device_guard=False, 
        device_check=, 
        python_module=None, 
        category_override=None, 
        variants={}, 
        manual_kernel_registration=False, 
        manual_cpp_binding=False, 
        loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4262), 
        autogen=[
            OperatorName(
                name=BaseOperatorName(
                    base='rand', 
                    inplace=False, 
                    dunder_method=False, 
                    functional_overload=False), 
                overload_name='generator_with_names_out')], 
        ufunc_inner_loop={}, 
        structured=False, 
        structured_delegate=None, 
        structured_inherits=None, 
        precomputed=None, 
        cpp_no_default_args=set(), 
        is_abstract=True, 
        has_composite_implicit_autograd_kernel=False, 
        has_composite_implicit_autograd_nested_tensor_kernel=False, 
        has_composite_explicit_autograd_kernel=True, 
        has_composite_explicit_autograd_non_functional_kernel=False, 
        tags={'nondeterministic_seeded'}), 
    outplace=None)
rand.generator & rand.generator_out
PythonSignatureGroup(signature=PythonSignature(name='rand', input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), input_kwargs=(PythonArgument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, default_init=None),), output_args=PythonOutArgument(name='out', type=BaseType(name=), default='None', default_init=None, outputs=(PythonArgument(name='out', type=BaseType(name=), default=None, default_init=None),)), returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), method=False), base=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='generator'), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, annotation=None),), tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4275), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}), outplace=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='generator_out'), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='generator', type=OptionalType(elem=BaseType(name=)), default=None, annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4285), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=False, has_composite_implicit_autograd_kernel=True, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}))
rand.names & rand.names_out
PythonSignatureGroup(signature=PythonSignature(name='rand', input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), input_kwargs=(PythonArgument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, default_init=None),), output_args=None, returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=None),)), tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), method=False), base=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='names'), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='names', type=OptionalType(elem=ListType(elem=BaseType(name=), size=None)), default=None, annotation=None),), tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=False, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4254), autogen=[OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='names_out')], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}), outplace=None)
rand & rand.out
PythonSignatureGroup(signature=PythonSignature(name='rand', input_args=(PythonArgument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, default_init=None),), input_kwargs=(), output_args=PythonOutArgument(name='out', type=BaseType(name=), default='None', default_init=None, outputs=(PythonArgument(name='out', type=BaseType(name=), default=None, default_init=None),)), returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), tensor_options_args=(PythonArgument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', default_init=None), PythonArgument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', default_init='torch::tensors::get_default_device()'), PythonArgument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None), PythonArgument(name='requires_grad', type=OptionalType(elem=BaseType(name=)), default='False', default_init=None)), method=False), base=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name=''), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(), tensor_options=TensorOptionsArguments(dtype=Argument(name='dtype', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), layout=Argument(name='layout', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), device=Argument(name='device', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None), pin_memory=Argument(name='pin_memory', type=OptionalType(elem=BaseType(name=)), default='None', annotation=None)), post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4270), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}), outplace=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='rand', inplace=False, dunder_method=False, functional_overload=False), 
overload_name='out'), arguments=Arguments(pre_self_positional=(), self_arg=None, post_self_positional=(Argument(name='size', type=ListType(elem=BaseType(name=), size=None), default=None, annotation=None),), pre_tensor_options_kwarg_only=(), tensor_options=None, post_tensor_options_kwarg_only=(), out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=4280), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=None, structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=True, has_composite_explicit_autograd_non_functional_kernel=False, tags={'nondeterministic_seeded'}))

上面共有四個PythonSignatureGroup元素，先來看第一個元素，其base成員的func的overload_name為generator_with_names，autogen的overload_name則為generator_with_names_out。第二個的則分別為generator和generator_out。第三個的分別為names和names_out。第四個的分別為空字串和out。

到這裡得到了八個rand相關函數。

add.Tensor & add.out
PythonSignatureGroup(signature=PythonSignature(name='add', input_args=(PythonArgument(name='self', type=BaseType(name=), default=None, default_init=None), PythonArgument(name='other', type=BaseType(name=), default=None, default_init=None)), input_kwargs=(PythonArgument(name='alpha', type=BaseType(name=), default='1', default_init=None),), output_args=PythonOutArgument(name='out', type=BaseType(name=), default='None', default_init=None, outputs=(PythonArgument(name='out', type=BaseType(name=), default=None, default_init=None),)), returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), tensor_options_args=(), method=False), base=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='Tensor'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={, }, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=497), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='out'), structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise', 'canonical'}), outplace=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='out'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=520), autogen=[], ufunc_inner_loop={: UfuncInnerLoop(name='add', supported_dtypes=, ufunc_key=), : UfuncInnerLoop(name='add', supported_dtypes=, ufunc_key=)}, structured=True, structured_delegate=None, structured_inherits='TensorIteratorBase', precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise'}))
add & add.Tensor & add.out
PythonSignatureGroup(signature=PythonSignatureDeprecated(name='add', input_args=(PythonArgument(name='self', type=BaseType(name=), default=None, default_init=None), PythonArgument(name='alpha', type=BaseType(name=), default=None, default_init=None), PythonArgument(name='other', type=BaseType(name=), default=None, default_init=None)), input_kwargs=(), output_args=PythonOutArgument(name='out', type=BaseType(name=), default='None', default_init=None, outputs=(PythonArgument(name='out', type=BaseType(name=), default=None, default_init=None),)), returns=PythonReturns(returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), tensor_options_args=(), method=False, deprecated_schema=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name=''), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='alpha', type=BaseType(name=), default=None, annotation=None), Argument(name='other', type=BaseType(name=), default=None, annotation=None)), pre_tensor_options_kwarg_only=(), tensor_options=None, post_tensor_options_kwarg_only=(), out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), deprecated_args_exprs=('out', 'self', 'other', 'alpha')), base=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='Tensor'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=()), returns=(Return(name=None, type=BaseType(name=), annotation=None),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={, }, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=497), autogen=[], ufunc_inner_loop={}, structured=False, structured_delegate=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='out'), structured_inherits=None, precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise', 'canonical'}), outplace=NativeFunction(namespace='aten', func=FunctionSchema(name=OperatorName(name=BaseOperatorName(base='add', inplace=False, dunder_method=False, functional_overload=False), overload_name='out'), arguments=Arguments(pre_self_positional=(), self_arg=SelfArgument(argument=Argument(name='self', type=BaseType(name=), default=None, annotation=None)), post_self_positional=(Argument(name='other', type=BaseType(name=), default=None, annotation=None),), pre_tensor_options_kwarg_only=(Argument(name='alpha', type=BaseType(name=), default='1', annotation=None),), tensor_options=None, post_tensor_options_kwarg_only=(), out=(Argument(name='out', type=BaseType(name=), default=None, annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), returns=(Return(name=None, type=BaseType(name=), annotation=Annotation(alias_set=('a',), is_write=True, alias_set_after=())),)), use_const_ref_for_mutable_tensors=False, device_guard=True, device_check=, python_module=None, category_override=None, variants={}, manual_kernel_registration=False, manual_cpp_binding=False, loc=Location(file='aten/src/ATen/native/native_functions.yaml', line=520), autogen=[], ufunc_inner_loop={: UfuncInnerLoop(name='add', supported_dtypes=, ufunc_key=), : UfuncInnerLoop(name='add', supported_dtypes=, ufunc_key=)}, structured=True, structured_delegate=None, structured_inherits='TensorIteratorBase', precomputed=None, cpp_no_default_args=set(), is_abstract=True, has_composite_implicit_autograd_kernel=False, has_composite_implicit_autograd_nested_tensor_kernel=False, has_composite_explicit_autograd_kernel=False, has_composite_explicit_autograd_non_functional_kernel=False, tags={'pointwise'}))

unsorted_function_hints

     for group in sorted(sig_groups, key=lambda g: g.signature.name):
        name = group.signature.name
        unsorted_function_hints[name] += generate_type_hints(group)

        named_tuple = returns_named_tuple_pyi(group.signature)
        if named_tuple is not None and not group.signature.deprecated:
            # deprecated namedtuples are currently not included for torch functions
            tuple_name, tuple_def = named_tuple
            if tuple_name in namedtuples:
                assert namedtuples[tuple_name] == tuple_def
            else:
                namedtuples[tuple_name] = tuple_def

unsorted_function_hints是一個defaultdict，key為函數名稱，value則為list of string。

rand

代表rand函數的元素如下：

'rand': [
    'def rand(size: Sequence[Union[_int, SymInt]], *, generator: Optional[Generator], names: Optional[Sequence[Union[str, ellipsis, None]]], dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...', 
    'def rand(*size: _int, generator: Optional[Generator], names: Optional[Sequence[Union[str, ellipsis, None]]], dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...', 
    'def rand(size: Sequence[Union[_int, SymInt]], *, generator: Optional[Generator], out: Optional[Tensor]=None, dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...', 
    'def rand(*size: _int, generator: Optional[Generator], out: Optional[Tensor]=None, dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...', 
    'def rand(size: Sequence[Union[_int, SymInt]], *, names: Optional[Sequence[Union[str, ellipsis, None]]], dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...', 
    'def rand(*size: _int, names: Optional[Sequence[Union[str, ellipsis, None]]], dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...', 
    'def rand(size: Sequence[Union[_int, SymInt]], *, out: Optional[Tensor]=None, dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...', 
    'def rand(*size: _int, out: Optional[Tensor]=None, dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...'
    ]

上面是八個overload的rand函數，可以將它們分為四組：有generator及names參數，只有generator參數，只有name參數，沒有generator和names參數。每組又可分為size參數是Sequence的及是int的。到這裡已經可以與torch/_C/_VariableFunctions.pyi一 一對應了。

add

找到名為add的key，其value list共有三個元素，分別對應add.Tensor，add_.Tensor，add.out。

'def add(input: Union[Tensor, Number], other: Union[Tensor, Number], *, alpha: Optional[Number]=1, out: Optional[Tensor]=None) -> Tensor: ...'

'def add(self: Tensor, alpha: Number, other: Tensor) -> Tensor: ...'

'def add(self: Tensor, alpha: Number, other: Tensor, *, out: Tensor) -> Tensor: ...'

function_hints

    function_hints = []
    for name, hints in sorted(unsorted_function_hints.items()):
        if len(hints) > 1:
            hints = ["@overload\n" + h for h in hints]
        function_hints += hints

function_hints是一個list of string：

['@overload\ndef __and_...ensor: ...', '@overload\ndef __and_...ensor: ...', '@overload\ndef __lshi...ensor: ...', '@overload\ndef __lshi...ensor: ...', '@overload\ndef __or__...ensor: ...', '@overload\ndef __or__...ensor: ...', '@overload\ndef __rshi...ensor: ...', '@overload\ndef __rshi...ensor: ...', '@overload\ndef __xor_...ensor: ...', '@overload\ndef __xor_...ensor: ...', 'def _adaptive_avg_po...ensor: ...', 'def _adaptive_avg_po...ensor: ...', 'def _add_batch_dim(i...ensor: ...', '@overload\ndef _add_r...ensor: ...', ...]

其第零個元素如下：

'@overload\ndef __and__(input: Tensor, other: Tensor) -> Tensor: ...'
rand

代表rand函數的八個元素如下。其實跟unsorted_function_hints裡的大同小異，差別只在於前面多加了’@overload\n’。

'@overload\ndef rand(size: Sequence[Union[_int, SymInt]], *, generator: Optional[Generator], names: Optional[Sequence[Union[str, ellipsis, None]]], dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...'

'@overload\ndef rand(*size: _int, generator: Optional[Generator], names: Optional[Sequence[Union[str, ellipsis, None]]], dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...'

'@overload\ndef rand(size: Sequence[Union[_int, SymInt]], *, generator: Optional[Generator], out: Optional[Tensor]=None, dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...'

'@overload\ndef rand(*size: _int, generator: Optional[Generator], out: Optional[Tensor]=None, dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...'

'@overload\ndef rand(size: Sequence[Union[_int, SymInt]], *, names: Optional[Sequence[Union[str, ellipsis, None]]], dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...'

'@overload\ndef rand(*size: _int, names: Optional[Sequence[Union[str, ellipsis, None]]], dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...'

'@overload\ndef rand(size: Sequence[Union[_int, SymInt]], *, out: Optional[Tensor]=None, dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...'

'@overload\ndef rand(*size: _int, out: Optional[Tensor]=None, dtype: Optional[_dtype]=None, layout: Optional[_layout]=None, device: Optional[Union[_device, str, None]]=None, pin_memory: Optional[_bool]=False, requires_grad: Optional[_bool]=False) -> Tensor: ...'
add

代表add函數的三個元素如下：

'@overload\ndef add(input: Union[Tensor, Number], other: Union[Tensor, Number], *, alpha: Optional[Number]=1, out: Optional[Tensor]=None) -> Tensor: ...'

'@overload\ndef add(self: Tensor, alpha: Number, other: Tensor) -> Tensor: ...'

'@overload\ndef add(self: Tensor, alpha: Number, other: Tensor, *, out: Tensor) -> Tensor: ...'

hinted_function_names

    # Generate __all__ directive
    # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    # Include only the functions that contain hints, to prevent undefined
    # symbols to be included in the `__all__` directive.
    hinted_function_names = [
        name for name, hint in unsorted_function_hints.items() if hint
    ]

hinted_function_names是一個list of string，看起來就只是一個有hint的函數名稱的列表：

['sparse_csr_tensor', '_sparse_csr_tensor_unsafe', 'sparse_csc_tensor', '_sparse_csc_tensor_unsafe', 'sparse_bsr_tensor', '_sparse_bsr_tensor_unsafe', 'sparse_bsc_tensor', '_sparse_bsc_tensor_unsafe', 'set_flush_denormal', 'get_default_dtype', 'asarray', 'from_numpy', 'frombuffer', 'numel', ...]

其中也包含了：

'rand', 'rand_like', 'randint_like', 'randn', 'randn_like', 'randperm'

及'add'。

all_symbols

    all_symbols = sorted(list(namedtuples.keys()) + hinted_function_names)

all_symbols

['__and__', '__lshift__', '__or__', '__rshift__', '__xor__', '_adaptive_avg_pool2d', '_adaptive_avg_pool3d', '_add_batch_dim', '_add_relu', '_add_relu_', '_addmm_activation', '_aminmax', '_amp_foreach_non_fin...d_unscale_', '_amp_update_scale_', ...]

其中也包含了：

'rand', 'rand_like', 'randint_like', 'randn', 'randn_like', 'randperm'

及'add'。

all_directive

接下來將all_symbols轉為string，用\n切割成多段個組成一個字串的列表，即為all_directive：

    all_directive = pformat(all_symbols, width=100, compact=True).split("\n")
    all_directive[0] = "__all__ = {}".format(all_directive[0])

第零個元素如下：

"__all__ = ['__and__', '__lshift__', '__or__', '__rshift__', '__xor__', '_adaptive_avg_pool2d',"

其中包含add的元素如下：

" 'adaptive_max_pool1d', 'add', 'addbmm', 'addcdiv', 'addcmul', 'addmm', 'addmv', 'addmv_', 'addr',"

包含rand的元素如下：

" 'rad2deg_', 'rand', 'rand_like', 'randint', 'randint_like', 'randn', 'randn_like', 'randperm',"

最後一個元素如下：

" 'vsplit', 'vstack', 'where', 'xlogy', 'xlogy_', 'zero_', 'zeros', 'zeros_like']"

env

到這裡為止得到了function_hints和all_directive，這兩個變數與其它變數共同組成了env：

    env = {
        "namedtuple_defs": namedtuple_defs,
        "function_hints": function_hints,
        "tensor_method_hints": tensor_method_hints,
        "legacy_class_hints": legacy_class_hints,
        "legacy_storage_base_hints": legacy_storage_base_hints,
        "dtype_class_hints": dtype_class_hints,
        "dispatch_key_hints": dispatch_key_hints,
        "all_directive": all_directive,
    }

運算後的env如下：

{
   "namedtuple_defs":[
      "_fake_quantize_per_t... Tensor)])",
      "_fused_moving_avg_ob... Tensor)])",
      "_linalg_det = NamedT... Tensor)])",
      "_linalg_eigh = Named... Tensor)])",
      "_linalg_slogdet = Na... Tensor)])",
      "_linalg_solve_ex = N... Tensor)])",
      "_linalg_svd = NamedT... Tensor)])",
      "_lu_with_info = Name... Tensor)])",
      "_unpack_dual = Named... Tensor)])",
      "..."
   ],
   "function_hints":[
      "@overload\ndef __and_...ensor: ...",
      "@overload\ndef __and_...ensor: ...",
      "@overload\ndef __lshi...ensor: ...",
      "@overload\ndef __lshi...ensor: ...",
      "@overload\ndef __or__...ensor: ...",
      "@overload\ndef __or__...ensor: ...",
      "@overload\ndef __rshi...ensor: ...",
      "@overload\ndef __rshi...ensor: ...",
      "@overload\ndef __xor_...ensor: ...",
      "..."
   ],
   "tensor_method_hints":[
      "def __abs__(self) ->...ensor: ...",
      "def __add__(self, ot...ensor: ...",
      "@overload\ndef __and_...ensor: ...",
      "@overload\ndef __and_...ensor: ...",
      "@overload\ndef __and_...ensor: ...",
      "def __bool__(self) -....bool: ...",
      "def __complex__(self...mplex: ...",
      "def __div__(self, ot...ensor: ...",
      "def __eq__(self, oth...[override]",
      "..."
   ],
   "legacy_class_hints":[
      "class DoubleTensor(T...nsor): ...",
      "class FloatTensor(Tensor): ...",
      "class LongTensor(Tensor): ...",
      "class IntTensor(Tensor): ...",
      "class ShortTensor(Tensor): ...",
      "class HalfTensor(Tensor): ...",
      "class CharTensor(Tensor): ...",
      "class ByteTensor(Tensor): ...",
      "class BoolTensor(Tensor): ..."
   ],
   "legacy_storage_base_hints":[
      "class StorageBase(object): ..."
   ],
   "dtype_class_hints":[
      "float32: dtype = ...",
      "float: dtype = ...",
      "float64: dtype = ...",
      "double: dtype = ...",
      "float16: dtype = ...",
      "bfloat16: dtype = ...",
      "half: dtype = ...",
      "uint8: dtype = ...",
      "int8: dtype = ...",
      "..."
   ],
   "dispatch_key_hints":[
      "Undefined: DispatchKey = ...",
      "FPGA: DispatchKey = ...",
      "ORT: DispatchKey = ...",
      "Vulkan: DispatchKey = ...",
      "Metal: DispatchKey = ...",
      "MKLDNN: DispatchKey = ...",
      "OpenGL: DispatchKey = ...",
      "OpenCL: DispatchKey = ...",
      "IDEEP: DispatchKey = ...",
      "..."
   ],
   "all_directive":[
      "__all__ = ['__and__...,",
      " ...,",
      " '_aminmax', ...,",
      " ...,",
      " '_cast_Float', ...,",
      " ...,",
      " ...,",
      " ...,",
      " '_convolution_mode...,",
      "..."
   ]
}

接著把env傳入FileManager的成員函數write_with_template：

    # ...
    fm.write_with_template(
        "torch/_C/__init__.pyi",
        "torch/_C/__init__.pyi.in",
        lambda: {
            "generated_comment": "@" + "generated from torch/_C/__init__.pyi.in",
            **env,
        },
    )
    fm.write_with_template(
        "torch/_C/_VariableFunctions.pyi",
        "torch/_C/_VariableFunctions.pyi.in",
        lambda: {
            "generated_comment": "@"
            + "generated from torch/_C/_VariableFunctions.pyi.in",
            **env,
        },
    )
    fm.write_with_template(
        "torch/_VF.pyi",
        "torch/_C/_VariableFunctions.pyi.in",
        lambda: {
            "generated_comment": "@"
            + "generated from torch/_C/_VariableFunctions.pyi.in",
            **env,
        },
    )
    fm.write_with_template(
        "torch/return_types.pyi",
        "torch/_C/return_types.pyi.in",
        lambda: {
            "generated_comment": "@" + "generated from torch/_C/return_types.pyi",
            **env,
        },
    )
    gen_nn_functional(fm)

可以看到這段代碼裡調用了FileManager的write_with_template及gen_nn_functional，以下先看gen_nn_functional。

參考Unpacking Operators in Python的Merging Dictionaries章節，{"a": 1, **my_dict}這種寫法是先把my_dict拆開(unpacking)，然後再與"a": 1共同構成一個新的字典。

lambda: {}這種寫法則表示一個無需參數並回傳一個字典的lambda函數。

注意到在呼叫write_with_template時，最後一個參數後面多了一個,。根據Should I add a trailing comma after the last argument in a function call? [closed]，在呼叫函數時，如果參數是分多行寫的，比較建議的寫法是在最後加上一個,。

回想一開始看到的，共會由六個pyi.in檔生成六個對應的pyi檔，此處只生成了四個pyi檔，剩下兩個（functional.pyi，_nn.pyi）則是在gen_nn_functional中調用FileManager.write_with_template生成。

FileManager.write_with_template函數會以模板為基礎，按照替換函數所指定的方式，生成pyi檔，本函數已獨立成篇，詳見PyTorch檔案生成機制中的FileManager.write_with_template。

gen_nn_functional

gen_nn_functional函數同樣位於tools/pyi/gen_pyi.py，它的作用是由torch/nn/functional.pyi.in及torch/_C/_nn.pyi.in生成torch/nn/functional.pyi和torch/_C/_nn.pyi.in。

def gen_nn_functional(fm: FileManager) -> None:
    # Functions imported into `torch.nn.functional` from `torch`, perhaps being filtered
    # through an `_add_docstr` call
    imports = [
        "conv1d",
        "conv2d",
        "conv3d",
        "conv_transpose1d",
        "conv_transpose2d",
        "conv_transpose3d",
        "conv_tbc",
        "avg_pool1d",
        "relu_",
        "selu_",
        "celu_",
        "rrelu_",
        "pixel_shuffle",
        "pixel_unshuffle",
        "channel_shuffle",
        "native_channel_shuffle",
        "pdist",
        "cosine_similarity",
    ]
    # Functions generated by `torch._jit_internal.boolean_dispatch`
    dispatches = [
        "fractional_max_pool2d",
        "fractional_max_pool3d",
        "max_pool1d",
        "max_pool2d",
        "max_pool3d",
        "adaptive_max_pool1d",
        "adaptive_max_pool2d",
        "adaptive_max_pool3d",
    ]
    # Functions directly imported from `torch._C`
    from_c = [
        "avg_pool2d",
        "avg_pool3d",
        "hardtanh_",
        "elu_",
        "leaky_relu_",
        "logsigmoid",
        "softplus",
        "softshrink",
        "one_hot",
    ]
    import_code = ["from .. import {0} as {0}".format(_) for _ in imports]
    # TODO make these types more precise
    dispatch_code = ["{}: Callable".format(_) for _ in (dispatches + from_c)]
    fm.write_with_template(
        "torch/nn/functional.pyi",
        "torch/nn/functional.pyi.in",
        lambda: {
            "imported_hints": import_code,
            "dispatched_hints": dispatch_code,
        },
    )

    # functional.pyi already contains the definitions for those functions
    # so, we don't export then to it
    from_c.extend(["hardtanh", "leaky_relu", "hardsigmoid"])
    dispatch_code = ["{}: Callable".format(_) for _ in (dispatches + from_c)]
    fm.write_with_template(
        "torch/_C/_nn.pyi",
        "torch/_C/_nn.pyi.in",
        lambda: {
            "imported_hints": import_code,
            "dispatched_hints": dispatch_code,
        },
    )

可以看到這個函數最後也是調用FileManager的write_with_template生成.pyi檔。

write_with_template已獨立成篇，詳見PyTorch檔案生成機制中的FileManager.write_with_template。

datapipe.pyi

回頭看CMakeLists.txt：

file(GLOB_RECURSE datapipe_files "${TORCH_SRC_DIR}/utils/data/datapipes/*.py")
add_custom_command(
    OUTPUT
    "${TORCH_SRC_DIR}/utils/data/datapipes/datapipe.pyi"
    COMMAND
    "${PYTHON_EXECUTABLE}" ${TORCH_SRC_DIR}/utils/data/datapipes/gen_pyi.py
    DEPENDS
    "${TORCH_SRC_DIR}/utils/data/datapipes/datapipe.pyi.in"
    ${datapipe_files}
    WORKING_DIRECTORY
    "${TORCH_ROOT}"
)

datapipe.pyi也是由類似的方式透過utils/data/datapipes/gen_pyi.py由datapipe.pyi.in生成的。

torch/utils/data/datapipes/datapipe.pyi.in中的注釋：

# This base template ("datapipe.pyi.in") is generated from mypy stubgen with minimal editing for code injection
# The output file will be "datapipe.pyi". This is executed as part of torch/CMakeLists.txt
# Note that, for mypy, .pyi file takes precedent over .py file, such that we must define the interface for other
# classes/objects here, even though we are not injecting extra code into them at the moment.

生成結果

以torch/_C/_VariableFunctions.pyi.in為例：

  • generated_comment

    # ${generated_comment}

    被替換成：

    # @generated from torch/_C/_VariableFunctions.pyi.in

  • function_hints

    ${function_hints}

    被替換成：

    @overload
def __and__(input: Tensor, other: Tensor) -> Tensor: ...
# ...
def zeros_like(input: Tensor, *, memory_format: Optional[memory_format] = None, dtype: Optional[_dtype] = None, layout: Optional[_layout] = None, device: Optional[Union[_device, str, None]] = None, pin_memory: Optional[_bool] = False, requires_grad: Optional[_bool] = False) -> Tensor: ...

  • all_directive

    ${all_directive}

    被替換成：

    __all__ = ['__and__', '__lshift__', '__or__', '__rshift__', '__xor__', '_adaptive_avg_pool2d',
# ...
 'view_copy', 'vsplit', 'vstack', 'where', 'xlogy', 'xlogy_', 'zero_', 'zeros', 'zeros_like']

其餘部份皆與torch/_C/_VariableFunctions.pyi.in相同。

使用pyi做類型檢查

在torch/__init__.py中有以下這麼一段：

# Appease the type checker: it can't deal with direct setting of globals().
# Note that we will see "too many" functions when reexporting this way; there
# is not a good way to fix this problem.  Perhaps, try to redesign VariableFunctions
# so that this import is good enough
if TYPE_CHECKING:
    # Some type signatures pulled in from _VariableFunctions here clash with
    # signatures already imported. For now these clashes are ignored; see
    # PR #43339 for details.
    from torch._C._VariableFunctions import *  # type: ignore[misc] # noqa: F403

也就是說，在類型檢查功能有被開啟的情況下，會引入torch._C.VariableFunctions中的所有東西。

其中torch._C._VariableFunctions指的就是我們剛剛看到的torch/_C/_VariableFunctions.pyi。

根據pyi文件是干嘛的？（一文读懂Python的存根文件和类型检查），在py檔和pyi檔名稱相同且置於同一資料夾的情況下不需要import pyi檔類型檢查就會啟動。在此處是因為py檔和pyi檔名稱不同，所以才要手動import pyi？

你可能感兴趣的:(PyTorch,c++,python,pytorch,pyi)

  • 理解Gunicorn：Python WSGI服务器的基石 范范0825 ipythonlinux运维
    理解Gunicorn：PythonWSGI服务器的基石介绍Gunicorn，全称GreenUnicorn，是一个为PythonWSGI（WebServerGatewayInterface）应用设计的高效、轻量级HTTP服务器。作为PythonWeb应用部署的常用工具，Gunicorn以其高性能和易用性著称。本文将介绍Gunicorn的基本概念、安装和配置，帮助初学者快速上手。1.什么是Gunico
  • c++ 的iostream 和 c++的stdio的区别和联系 黄卷青灯77 c++算法开发语言iostreamstdio
    在C++中，iostream和C语言的stdio.h都是用于处理输入输出的库，但它们在设计、用法和功能上有许多不同。以下是两者的区别和联系：区别1.编程风格iostream（C++风格）：C++标准库中的输入输出流类库，支持面向对象的输入输出操作。典型用法是cin（输入）和cout（输出），使用>操作符来处理数据。更加类型安全，支持用户自定义类型的输入输出。#includeintmain(){in
  • Python数据分析与可视化实战指南 William数据分析 pythonpython数据
    在数据驱动的时代，Python因其简洁的语法、强大的库生态系统以及活跃的社区，成为了数据分析与可视化的首选语言。本文将通过一个详细的案例，带领大家学习如何使用Python进行数据分析，并通过可视化来直观呈现分析结果。一、环境准备1.1安装必要库在开始数据分析和可视化之前，我们需要安装一些常用的库。主要包括pandas、numpy、matplotlib和seaborn等。这些库分别用于数据处理、数学
  • python os.environ 江湖偌大 python深度学习
    os.environ['TF_CPP_MIN_LOG_LEVEL']='0'#默认值，输出所有信息os.environ['TF_CPP_MIN_LOG_LEVEL']='1'#屏蔽通知信息（INFO）os.environ['TF_CPP_MIN_LOG_LEVEL']='2'#屏蔽通知信息和警告信息（INFO\WARNING）os.environ['TF_CPP_MIN_LOG_LEVEL']='
  • Python中os.environ基本介绍及使用方法 鹤冲天Pro #Pythonpython服务器开发语言
    文章目录python中os.environos.environ简介os.environ进行环境变量的增删改查python中os.environ的使用详解1.简介2.key字段详解2.1常见key字段3.os.environ.get()用法4.环境变量的增删改查和判断是否存在4.1新增环境变量4.2更新环境变量4.3获取环境变量4.4删除环境变量4.5判断环境变量是否存在python中os.envi
  • Pyecharts数据可视化大屏：打造沉浸式数据分析体验 我的运维人生 信息可视化数据分析数据挖掘运维开发技术共享
    Pyecharts数据可视化大屏：打造沉浸式数据分析体验在当今这个数据驱动的时代，如何将海量数据以直观、生动的方式展现出来，成为了数据分析师和企业决策者关注的焦点。Pyecharts，作为一款基于Python的开源数据可视化库，凭借其丰富的图表类型、灵活的配置选项以及高度的定制化能力，成为了构建数据可视化大屏的理想选择。本文将深入探讨如何利用Pyecharts打造数据可视化大屏，并通过实际代码案例
  • Python教程：一文了解使用Python处理XPath 旦莫 Python进阶python开发语言
    目录1.环境准备1.1安装lxml1.2验证安装2.XPath基础2.1什么是XPath？2.2XPath语法2.3示例XML文档3.使用lxml解析XML3.1解析XML文档3.2查看解析结果4.XPath查询4.1基本路径查询4.2使用属性查询4.3查询多个节点5.XPath的高级用法5.1使用逻辑运算符5.2使用函数6.实战案例6.1从网页抓取数据6.1.1安装Requests库6.1.2代
  • python os.environ_python os.environ 读取和设置环境变量 weixin_39605414 pythonos.environ
    >>>importos>>>os.environ.keys()['LC_NUMERIC','GOPATH','GOROOT','GOBIN','LESSOPEN','SSH_CLIENT','LOGNAME','USER','HOME','LC_PAPER','PATH','DISPLAY','LANG','TERM','SHELL','J2REDIR','LC_MONETARY','QT_QPA
  • 使用Faiss进行高效相似度搜索 llzwxh888 faisspython
    在现代AI应用中，快速和高效的相似度搜索是至关重要的。Faiss（FacebookAISimilaritySearch）是一个专门用于快速相似度搜索和聚类的库，特别适用于高维向量。本文将介绍如何使用Faiss来进行相似度搜索，并结合Python代码演示其基本用法。什么是Faiss？Faiss是一个由FacebookAIResearch团队开发的开源库，主要用于高维向量的相似性搜索和聚类。Faiss
  • python是什么意思中文-在python中%是什么意思 编程大乐趣
    Python中%有两种：1、数值运算：%代表取模，返回除法的余数。如：>>>7%212、%操作符（字符串格式化，stringformatting），说明如下：%[(name)][flags][width].[precision]typecode(name)为命名flags可以有+，-，''或0。+表示右对齐。-表示左对齐。''为一个空格，表示在正数的左侧填充一个空格，从而与负数对齐。0表示使用0填
  • Day1笔记-Python简介&标识符和关键字&输入输出 ~在杰难逃~ Pythonpython开发语言大数据数据分析数据挖掘
    大家好，从今天开始呢，杰哥开展一个新的专栏，当然，数据分析部分也会不定时更新的，这个新的专栏主要是讲解一些Python的基础语法和知识，帮助0基础的小伙伴入门和学习Python，感兴趣的小伙伴可以开始认真学习啦！一、Python简介【了解】1.计算机工作原理编程语言就是用来定义计算机程序的形式语言。我们通过编程语言来编写程序代码，再通过语言处理程序执行向计算机发送指令，让计算机完成对应的工作，编程
  • python八股文面试题分享及解析(1) Shawn________ python
    #1.'''a=1b=2不用中间变量交换a和b'''#1.a=1b=2a,b=b,aprint(a)print(b)结果：21#2.ll=[]foriinrange(3):ll.append({'num':i})print(11)结果:#[{'num':0},{'num':1},{'num':2}]#3.kk=[]a={'num':0}foriinrange(3):#0,12#可变类型，不仅仅改变
  • 【JS】执行时长(100分) |思路参考+代码解析（C++） l939035548 JS算法数据结构c++
    题目为了充分发挥GPU算力，需要尽可能多的将任务交给GPU执行，现在有一个任务数组，数组元素表示在这1秒内新增的任务个数且每秒都有新增任务。假设GPU最多一次执行n个任务，一次执行耗时1秒，在保证GPU不空闲情况下，最少需要多长时间执行完成。题目输入第一个参数为GPU一次最多执行的任务个数，取值范围[1,10000]第二个参数为任务数组长度，取值范围[1,10000]第三个参数为任务数组，数字范围
  • 每日算法&面试题，大厂特训二十八天——第二十天（树） 肥学 ⚡算法题⚡面试题每日精进java算法数据结构
    目录标题导读算法特训二十八天面试题点击直接资料领取导读肥友们为了更好的去帮助新同学适应算法和面试题，最近我们开始进行专项突击一步一步来。上一期我们完成了动态规划二十一天现在我们进行下一项对各类算法进行二十八天的一个小总结。还在等什么快来一起肥学进行二十八天挑战吧！！特别介绍小白练手专栏，适合刚入手的新人欢迎订阅编程小白进阶python有趣练手项目里面包括了像《机器人尬聊》《恶搞程序》这样的有趣文章
  • Python快速入门 —— 第三节：类与对象 孤华暗香 Python快速入门python开发语言
    第三节：类与对象目标：了解面向对象编程的基础概念，并学会如何定义类和创建对象。内容：类与对象：定义类：class关键字。类的构造函数：__init__()。类的属性和方法。对象的创建与使用。示例：classStudent:def__init__(self,name,age,major):self.name&#
  • pyecharts——绘制柱形图折线图 2224070247 信息可视化pythonjava数据可视化
    一、pyecharts概述自2013年6月百度EFE(ExcellentFrontEnd）数据可视化团队研发的ECharts1.0发布到GitHub网站以来，ECharts一直备受业界权威的关注并获得广泛好评，成为目前成熟且流行的数据可视化图表工具，被应用到诸多数据可视化的开发领域。Python作为数据分析领域最受欢迎的语言，也加入ECharts的使用行列，并研发出方便Python开发者使用的数据
  • Python 实现图片裁剪（附代码） | Python工具 剑客阿良_ALiang
    前言本文提供将图片按照自定义尺寸进行裁剪的工具方法，一如既往的实用主义。环境依赖ffmpeg环境安装，可以参考我的另一篇文章：windowsffmpeg安装部署_阿良的博客-CSDN博客本文主要使用到的不是ffmpeg，而是ffprobe也在上面这篇文章中的zip包中。ffmpy安装：pipinstallffmpy-ihttps://pypi.douban.com/simple代码不废话了，上代码
  • 【华为OD技术面试真题 - 技术面】- python八股文真题题库（4) 算法大师 华为od面试python
    华为OD面试真题精选专栏：华为OD面试真题精选目录:2024华为OD面试手撕代码真题目录以及八股文真题目录文章目录华为OD面试真题精选**1.Python中的`with`**用途和功能自动资源管理示例：文件操作上下文管理协议示例代码工作流程解析优点2.\_\_new\_\_和**\_\_init\_\_**区别__new____init__区别总结3.**切片（Slicing）操作**基本切片语法
  • python os 环境变量 CV矿工 python开发语言numpy
    环境变量：环境变量是程序和操作系统之间的通信方式。有些字符不宜明文写进代码里，比如数据库密码，个人账户密码，如果写进自己本机的环境变量里，程序用的时候通过os.environ.get（）取出来就行了。os.environ是一个环境变量的字典。环境变量的相关操作importos"""设置/修改环境变量：os.environ[‘环境变量名称’]=‘环境变量值’#其中key和value均为string类
  • Python爬虫解析工具之xpath使用详解 eqa11 python爬虫开发语言
    文章目录Python爬虫解析工具之xpath使用详解一、引言二、环境准备1、插件安装2、依赖库安装三、xpath语法详解1、路径表达式2、通配符3、谓语4、常用函数四、xpath在Python代码中的使用1、文档树的创建2、使用xpath表达式3、获取元素内容和属性五、总结Python爬虫解析工具之xpath使用详解一、引言在Python爬虫开发中，数据提取是一个至关重要的环节。xpath作为一门
  • 基于CODESYS的多轴运动控制程序框架：逻辑与运动控制分离，快速开发灵活操作 GPJnCrbBdl python开发语言
    基于codesys开发的多轴运动控制程序框架，将逻辑与运动控制分离，将单轴控制封装成功能块，对该功能块的操作包含了所有的单轴控制（归零、点动、相对定位、绝对定位、设置当前位置、伺服模式切换等等）。程序框架由主程序按照状态调用分归零模式、手动模式、自动模式、故障模式，程序状态的跳转都已完成，只需要根据不同的工艺要求完成所需的动作即可。变量的声明、地址的规划都严格按照C++的标准定义，能帮助开发者快速
  • C++ | Leetcode C++题解之第409题最长回文串 Ddddddd_158 经验分享C++Leetcode题解
    题目：题解：classSolution{public:intlongestPalindrome(strings){unordered_mapcount;intans=0;for(charc:s)++count[c];for(autop:count){intv=p.second;ans+=v/2*2;if(v%2==1andans%2==0)++ans;}returnans;}};
  • C++菜鸟教程 - 从入门到精通 第二节 DreamByte c++
    一.上节课的补充(数据类型)1.前言继上节课,我们主要讲解了输入,输出和运算符,我们现在来补充一下数据类型的知识上节课遗漏了这个知识点,非常的抱歉顺便说一下,博主要上高中了,更新会慢,2-4周更新一次对了,正好赶上中秋节,小编跟大家说一句:中秋节快乐!2.int类型上节课,我们其实只用了int类型int类型,是整数类型,它们存贮的是整数,不能存小数(浮点数)定义变量的方式很简单inta;//定义一
  • 【华为OD技术面试真题 - 技术面】- python八股文真题题库（1） 算法大师 华为od面试python
    华为OD面试真题精选专栏：华为OD面试真题精选目录:2024华为OD面试手撕代码真题目录以及八股文真题目录文章目录华为OD面试真题精选1.数据预处理流程数据预处理的主要步骤工具和库2.介绍线性回归、逻辑回归模型线性回归（LinearRegression）模型形式：关键点：逻辑回归（LogisticRegression）模型形式：关键点：参数估计与评估：3.python浅拷贝及深拷贝浅拷贝（Shal
  • nosql数据库技术与应用知识点 皆过客，揽星河 NoSQLnosql数据库大数据数据分析数据结构非关系型数据库
    Nosql知识回顾大数据处理流程数据采集(flume、爬虫、传感器)数据存储(本门课程NoSQL所处的阶段)Hdfs、MongoDB、HBase等数据清洗(入仓)Hive等数据处理、分析(Spark、Flink等)数据可视化数据挖掘、机器学习应用(Python、SparkMLlib等)大数据时代存储的挑战(三高)高并发(同一时间很多人访问)高扩展(要求随时根据需求扩展存储)高效率(要求读写速度快)
  • 《Python数据分析实战终极指南》 xjt921122 python数据分析开发语言
    对于分析师来说，大家在学习Python数据分析的路上，多多少少都遇到过很多大坑**，有关于技能和思维的**：Excel已经没办法处理现有的数据量了，应该学Python吗？找了一大堆Python和Pandas的资料来学习，为什么自己动手就懵了？跟着比赛类公开数据分析案例练了很久，为什么当自己面对数据需求还是只会数据处理而没有分析思路？学了对比、细分、聚类分析，也会用PEST、波特五力这类分析法，为啥
  • Python中深拷贝与浅拷贝的区别 yuxiaoyu.
    转自：http://blog.csdn.net/u014745194/article/details/70271868定义：在Python中对象的赋值其实就是对象的引用。当创建一个对象，把它赋值给另一个变量的时候，python并没有拷贝这个对象，只是拷贝了这个对象的引用而已。浅拷贝：拷贝了最外围的对象本身，内部的元素都只是拷贝了一个引用而已。也就是，把对象复制一遍，但是该对象中引用的其他对象我不复
  • Python开发常用的三方模块如下： 换个网名有点难 python开发语言
    Python是一门功能强大的编程语言，拥有丰富的第三方库，这些库为开发者提供了极大的便利。以下是100个常用的Python库，涵盖了多个领域：1、NumPy，用于科学计算的基础库。2、Pandas，提供数据结构和数据分析工具。3、Matplotlib，一个绘图库。4、Scikit-learn，机器学习库。5、SciPy，用于数学、科学和工程的库。6、TensorFlow，由Google开发的开源机
  • Python编译器 鹿鹿~ Python编译器Pythonpython开发语言后端
    嘿嘿嘿我又来了啊有些小盆友可能不知道Python其实是有编译器的，也就是PyCharm。你们可能会问到这个是干嘛的又不可以吃也不可以穿好像没有什么用，其实你还说对了这个还真的不可以吃也不可以穿，但是它用来干嘛的呢。用来编译你所打出的代码进行运行（可能这里说的有点不对但是只是个人认为）现在我们来说说PyCharm是用来干嘛的。PyCharm是一种PythonIDE，带有一整套可以帮助用户在使用Pyt
  • 一文掌握python面向对象魔术方法（二） 程序员neil pythonpython开发语言
    接上篇：一文掌握python面向对象魔术方法（一）-CSDN博客目录六、迭代和序列化：1、__iter__(self):定义迭代器，使得类可以被for循环迭代。2、__getitem__(self,key):定义索引操作，如obj[key]。3、__setitem__(self,key,value):定义赋值操作，如obj[key]=value。4、__delitem__(self,key):定义
  • java责任链模式 3213213333332132 java责任链模式村民告县长
    责任链模式，通常就是一个请求从最低级开始往上层层的请求，当在某一层满足条件时，请求将被处理，当请求到最高层仍未满足时，则请求不会被处理。 就是一个请求在这个链条的责任范围内，会被相应的处理，如果超出链条的责任范围外，请求不会被相应的处理。 下面代码模拟这样的效果： 创建一个政府抽象类,方便所有的具体政府部门继承它。 package 责任链模式; /** *
  • linux、mysql、nginx、tomcat 性能参数优化 ronin47
    一、linux 系统内核参数 /etc/sysctl.conf文件常用参数 net.core.netdev_max_backlog = 32768 #允许送到队列的数据包的最大数目 net.core.rmem_max = 8388608 #SOCKET读缓存区大小 net.core.wmem_max = 8388608 #SOCKET写缓存区大
  • php命令行界面 dcj3sjt126com PHPcli
    常用选项 php -v php -i PHP安装的有关信息 php -h 访问帮助文件 php -m 列出编译到当前PHP安装的所有模块 执行一段代码 php -r 'echo "hello, world!";' php -r 'echo "Hello, World!\n";' php -r '$ts = filemtime("
  • Filter&Session 171815164 session
    Filter HttpServletRequest requ = (HttpServletRequest) req; HttpSession session = requ.getSession(); if (session.getAttribute("admin") == null) { PrintWriter out = res.ge
  • 连接池与Spring,Hibernate结合 g21121 Hibernate
            前几篇关于Java连接池的介绍都是基于Java应用的，而我们常用的场景是与Spring和ORM框架结合，下面就利用实例学习一下这方面的配置。         1.下载相关内容：     &nb
  • [简单]mybatis判断数字类型 53873039oycg mybatis
           昨天同事反馈mybatis保存不了int类型的属性,一直报错，错误信息如下:       Caused by: java.lang.NumberFormatException: For input string: "null" at sun.mis
  • 项目启动时或者启动后ava.lang.OutOfMemoryError: PermGen space 程序员是怎么炼成的 eclipsejvmtomcatcatalina.sheclipse.ini
       在启动比较大的项目时，因为存在大量的jsp页面，所以在编译的时候会生成很多的.class文件，.class文件是都会被加载到jvm的方法区中，如果要加载的class文件很多，就会出现方法区溢出异常 java.lang.OutOfMemoryError: PermGen space.     解决办法是点击eclipse里的tomcat，在
  • 我的crm小结 aijuans crm
    各种原因吧，crm今天才完了。主要是接触了几个新技术： Struts2、poi、ibatis这几个都是以前的项目中用过的。 Jsf、tapestry是这次新接触的，都是界面层的框架，用起来也不难。思路和struts不太一样，传说比较简单方便。不过个人感觉还是struts用着顺手啊，当然springmvc也很顺手，不知道是因为习惯还是什么。jsf和tapestry应用的时候需要知道他们的标签、主
  • spring里配置使用hibernate的二级缓存几步 antonyup_2006 javaspringHibernatexmlcache
    ．在spring的配置文件中 applicationContent.xml，hibernate部分加入 xml 代码 <prop key="hibernate.cache.provider_class">org.hibernate.cache.EhCacheProvider</prop> <prop key="hi
  • JAVA基础面试题 百合不是茶 抽象实现接口String类接口继承抽象类继承实体类自定义异常
    /*   * 栈（stack）：主要保存基本类型（或者叫内置类型）（char、byte、short、   *int、long、 float、double、boolean）和对象的引用，数据可以共享，速度仅次于   * 寄存器（register），快于堆。堆（heap）：用于存储对象。   */  &
  • 让sqlmap文件 "继承" 起来 bijian1013 javaibatissqlmap
            多个项目中使用ibatis , 和数据库表对应的 sqlmap文件（增删改查等基本语句)，dao, pojo 都是由工具自动生成的, 现在将这些自动生成的文件放在一个单独的工程中，其它项目工程中通过jar包来引用 ，并通过"继承"为基础的sqlmap文件，dao,pojo 添加新的方法来满足项
  • 精通Oracle10编程SQL(13)开发触发器 bijian1013 oracle数据库plsql
    /* *开发触发器 */ --得到日期是周几 select to_char(sysdate+4,'DY','nls_date_language=AMERICAN') from dual; select to_char(sysdate,'DY','nls_date_language=AMERICAN') from dual; --建立BEFORE语句触发器 CREATE O
  • 【EhCache三】EhCache查询 bit1129 ehcache
    本文介绍EhCache查询缓存中数据，EhCache提供了类似Hibernate的查询API，可以按照给定的条件进行查询。   要对EhCache进行查询，需要在ehcache.xml中设定要查询的属性   数据准备 @Before public void setUp() { //加载EhCache配置文件 Inpu
  • CXF框架入门实例 白糖_ springWeb框架webserviceservlet
    CXF是apache旗下的开源框架，由Celtix + XFire这两门经典的框架合成，是一套非常流行的web service框架。 它提供了JAX-WS的全面支持，并且可以根据实际项目的需要，采用代码优先（Code First）或者 WSDL 优先（WSDL First）来轻松地实现 Web Services 的发布和使用，同时它能与spring进行完美结合。 在apache cxf官网提供
  • angular.equals boyitech AngularJSAngularJS APIAnguarJS 中文APIangular.equals
    angular.equals   描述: 比较两个值或者两个对象是不是 相等。还支持值的类型，正则表达式和数组的比较。   两个值或对象被认为是 相等的前提条件是以下的情况至少能满足一项： 两个值或者对象能通过=== （恒等） 的比较 两个值或者对象是同样类型，并且他们的属性都能通过angular
  • java-腾讯暑期实习生-输入一个数组A[1,2,...n]，求输入B，使得数组B中的第i个数字B[i]=A[0]*A[1]*...*A[i-1]*A[i+1] bylijinnan java
    这道题的具体思路请参看 何海涛的微博：http://weibo.com/zhedahht import java.math.BigInteger; import java.util.Arrays; public class CreateBFromATencent { /** * 题目：输入一个数组A[1,2,...n]，求输入B，使得数组B中的第i个数字B[i]=A
  • FastDFS 的安装和配置 修订版 Chen.H linuxfastDFS分布式文件系统
    FastDFS Home:http://code.google.com/p/fastdfs/ 1. 安装 http://code.google.com/p/fastdfs/wiki/Setup http://hi.baidu.com/leolance/blog/item/3c273327978ae55f93580703.html 安装libevent (对libevent的版本要求为1.4.
  • [强人工智能]拓扑扫描与自适应构造器 comsci 人工智能
          当我们面对一个有限拓扑网络的时候,在对已知的拓扑结构进行分析之后,发现在连通点之后,还存在若干个子网络,且这些网络的结构是未知的,数据库中并未存在这些网络的拓扑结构数据....这个时候,我们该怎么办呢?       那么,现在我们必须设计新的模块和代码包来处理上面的问题
  • oracle merge into的用法 daizj oraclesqlmerget into
    Oracle中merge into的使用 http://blog.csdn.net/yuzhic/article/details/1896878 http://blog.csdn.net/macle2010/article/details/5980965 该命令使用一条语句从一个或者多个数据源中完成对表的更新和插入数据. ORACLE 9i 中，使用此命令必须同时指定UPDATE 和INSE
  • 不适合使用Hadoop的场景 datamachine hadoop
    转自：http://dev.yesky.com/296/35381296.shtml。 　　Hadoop通常被认定是能够帮助你解决所有问题的唯一方案。 当人们提到“大数据”或是“数据分析”等相关问题的时候，会听到脱口而出的回答：Hadoop!  实际上Hadoop被设计和建造出来，是用来解决一系列特定问题的。对某些问题来说，Hadoop至多算是一个不好的选择，对另一些问题来说，选择Ha
  • YII findAll的用法 dcj3sjt126com yii
    看文档比较糊涂，其实挺简单的： $predictions=Prediction::model()->findAll("uid=:uid",array(":uid"=>10));   第一个参数是选择条件：”uid=10″。其中:uid是一个占位符，在后面的array(“:uid”=>10)对齐进行了赋值； 更完善的查询需要
  • vim 常用 NERDTree 快捷键 dcj3sjt126com vim
    下面给大家整理了一些vim NERDTree的常用快捷键了，这里几乎包括了所有的快捷键了，希望文章对各位会带来帮助。 切换工作台和目录 ctrl + w + h 光标 focus 左侧树形目录ctrl + w + l 光标 focus 右侧文件显示窗口ctrl + w + w 光标自动在左右侧窗口切换ctrl + w + r 移动当前窗口的布局位置 o 在已有窗口中打开文件、目录或书签，并跳
  • Java把目录下的文件打印出来 蕃薯耀 列出目录下的文件文件夹下面的文件目录下的文件
    Java把目录下的文件打印出来 >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 蕃薯耀 2015年7月11日 11:02:
  • linux远程桌面----VNCServer与rdesktop hanqunfeng Desktop
    windows远程桌面到linux，需要在linux上安装vncserver，并开启vnc服务，同时需要在windows下使用vnc-viewer访问Linux。vncserver同时支持linux远程桌面到linux。   linux远程桌面到windows，需要在linux上安装rdesktop，同时开启windows的远程桌面访问。   下面分别介绍，以windo
  • guava中的join和split功能 jackyrong java
    guava库中，包含了很好的join和split的功能，例子如下： 1） 将LIST转换为使用字符串连接的字符串    List<String> names = Lists.newArrayList("John", "Jane", "Adam", "Tom");
  • Web开发技术十年发展历程 lampcy androidWeb浏览器html5
    回顾web开发技术这十年发展历程： Ajax 03年的时候我上六年级，那时候网吧刚在小县城的角落萌生。传奇，大话西游第一代网游一时风靡。我抱着试一试的心态给了网吧老板两块钱想申请个号玩玩，然后接下来的一个小时我一直在，注，册，账，号。 彼时网吧用的512k的带宽，注册的时候，填了一堆信息，提交，页面跳转，嘣，”您填写的信息有误，请重填”。然后跳转回注册页面，以此循环。我现在时常想，如果当时a
  • 架构师之mima-----------------mina的非NIO控制IOBuffer(说得比较好) nannan408 buffer
    1.前言。   如题。 2.代码。   IoService IoService是一个接口，有两种实现：IoAcceptor和IoConnector；其中IoAcceptor是针对Server端的实现，IoConnector是针对Client端的实现；IoService的职责包括： 1、监听器管理 2、IoHandler 3、IoSession
  • ORA-00054:resource busy and acquire with NOWAIT specified Everyday都不同 oraclesessionLock
    [Oracle] 今天对一个数据量很大的表进行操作时，出现如题所示的异常。此时表明数据库的事务处于“忙”的状态，而且被lock了，所以必须先关闭占用的session。   step1，查看被lock的session：   select t2.username, t2.sid, t2.serial#, t2.logon_time from v$locked_obj
  • javascript学习笔记 tntxia JavaScript
      javascript里面有6种基本类型的值:number、string、boolean、object、function和undefined。number：就是数字值，包括整数、小数、NaN、正负无穷。string:字符串类型、单双引号引起来的内容。boolean:true、false object:表示所有的javascript对象，不用多说function:我们熟悉的方法，也就是
  • Java enum的用法详解 xieke90 enum枚举
    Java中枚举实现的分析：     示例：  public static enum SEVERITY{ INFO,WARN,ERROR }     enum很像特殊的class，实际上enum声明定义的类型就是一个类。 而这些类都是类库中Enum类的子类      (java.l
按字母分类： ABCDEFGHIJKLMNOPQRSTUVWXYZ其他