numba中的numpy数组高级索引

我正在将一个函数转换为numba（nopython模式）。在这个函数中，我需要访问2D数组的一些元素，这些元素的行索引包含在列表中，列是固定的。 在通常的python中，我会：

array[list,column_index]

但在numba中编译时，我收到以下错误：

TypingError: Invalid use of Function(<built-in function getitem>) with argument(s) of type(s): (array(float64, 2d, C), (reflected list(int64), int64))
 * parameterized
In definition 0:
    All templates rejected with literals.
In definition 1:
    All templates rejected without literals.
In definition 2:
    All templates rejected with literals.
In definition 3:
    All templates rejected without literals.
In definition 4:
    All templates rejected with literals.
In definition 5:
    All templates rejected without literals.
In definition 6:
    TypeError: unsupported array index type reflected list(int64) in (reflected list(int64), int64)
    raised from C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\typing\arraydecl.py:71
In definition 7:
    TypeError: unsupported array index type reflected list(int64) in (reflected list(int64), int64)
    raised from C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\typing\arraydecl.py:71
This error is usually caused by passing an argument of a type that is unsupported by the named function.

Traceback (most recent call last):

  File "<ipython-input-123-c5ca441bb6b7>", line 1, in <module>
    tmp = L.numba_montecarlo_barab(SEED,w,R,T,mpx,0.5)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\dispatcher.py", line 367, in _compile_for_args
    raise e

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\dispatcher.py", line 324, in _compile_for_args
    return self.compile(tuple(argtypes))

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler_lock.py", line 32, in _acquire_compile_lock
    return func(*args, **kwargs)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\dispatcher.py", line 655, in compile
    cres = self._compiler.compile(args, return_type)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\dispatcher.py", line 82, in compile
    pipeline_class=self.pipeline_class)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 926, in compile_extra
    return pipeline.compile_extra(func)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 374, in compile_extra
    return self._compile_bytecode()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 857, in _compile_bytecode
    return self._compile_core()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 844, in _compile_core
    res = pm.run(self.status)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler_lock.py", line 32, in _acquire_compile_lock
    return func(*args, **kwargs)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 255, in run
    raise patched_exception

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 246, in run
    stage()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 717, in stage_nopython_backend
    self._backend(lowerfn, objectmode=False)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 666, in _backend
    lowered = lowerfn()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 653, in backend_nopython_mode
    self.metadata)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 1050, in native_lowering_stage
    lower.create_cpython_wrapper(flags.release_gil)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\lowering.py", line 272, in create_cpython_wrapper
    release_gil=release_gil)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\targets\cpu.py", line 149, in create_cpython_wrapper
    builder.build()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\callwrapper.py", line 122, in build
    self.build_wrapper(api, builder, closure, args, kws)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\callwrapper.py", line 155, in build_wrapper
    val = cleanup_manager.add_arg(builder.load(obj), ty)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\callwrapper.py", line 58, in add_arg
    cleanup_arg()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\callwrapper.py", line 41, in cleanup_arg
    self.api.reflect_native_value(ty, native.value, self.env_manager)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\pythonapi.py", line 1417, in reflect_native_value
    impl(typ, val, c)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\targets\boxing.py", line 740, in reflect_list
    raise TypeError(msg)

TypeError: Failed in nopython mode pipeline (step: nopython mode backend)
cannot reflect element of reflected container: reflected list(reflected list(reflected list(int64)))

产生以下错误：

TypingError: Invalid use of Function(<built-in function getitem>) with argument(s) of type(s): (array(float64, 2d, C), (reflected list(int64), int64))
 * parameterized
In definition 0:
    All templates rejected with literals.
In definition 1:
    All templates rejected without literals.
In definition 2:
    All templates rejected with literals.
In definition 3:
    All templates rejected without literals.
In definition 4:
    All templates rejected with literals.
In definition 5:
    All templates rejected without literals.
In definition 6:
    TypeError: unsupported array index type reflected list(int64) in (reflected list(int64), int64)
    raised from C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\typing\arraydecl.py:71
In definition 7:
    TypeError: unsupported array index type reflected list(int64) in (reflected list(int64), int64)
    raised from C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\typing\arraydecl.py:71
This error is usually caused by passing an argument of a type that is unsupported by the named function.

Traceback (most recent call last):

  File "<ipython-input-123-c5ca441bb6b7>", line 1, in <module>
    tmp = L.numba_montecarlo_barab(SEED,w,R,T,mpx,0.5)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\dispatcher.py", line 367, in _compile_for_args
    raise e

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\dispatcher.py", line 324, in _compile_for_args
    return self.compile(tuple(argtypes))

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler_lock.py", line 32, in _acquire_compile_lock
    return func(*args, **kwargs)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\dispatcher.py", line 655, in compile
    cres = self._compiler.compile(args, return_type)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\dispatcher.py", line 82, in compile
    pipeline_class=self.pipeline_class)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 926, in compile_extra
    return pipeline.compile_extra(func)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 374, in compile_extra
    return self._compile_bytecode()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 857, in _compile_bytecode
    return self._compile_core()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 844, in _compile_core
    res = pm.run(self.status)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler_lock.py", line 32, in _acquire_compile_lock
    return func(*args, **kwargs)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 255, in run
    raise patched_exception

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 246, in run
    stage()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 717, in stage_nopython_backend
    self._backend(lowerfn, objectmode=False)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 666, in _backend
    lowered = lowerfn()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 653, in backend_nopython_mode
    self.metadata)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\compiler.py", line 1050, in native_lowering_stage
    lower.create_cpython_wrapper(flags.release_gil)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\lowering.py", line 272, in create_cpython_wrapper
    release_gil=release_gil)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\targets\cpu.py", line 149, in create_cpython_wrapper
    builder.build()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\callwrapper.py", line 122, in build
    self.build_wrapper(api, builder, closure, args, kws)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\callwrapper.py", line 155, in build_wrapper
    val = cleanup_manager.add_arg(builder.load(obj), ty)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\callwrapper.py", line 58, in add_arg
    cleanup_arg()

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\callwrapper.py", line 41, in cleanup_arg
    self.api.reflect_native_value(ty, native.value, self.env_manager)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\pythonapi.py", line 1417, in reflect_native_value
    impl(typ, val, c)

  File "C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site-packages\numba\targets\boxing.py", line 740, in reflect_list
    raise TypeError(msg)

TypeError: Failed in nopython mode pipeline (step: nopython mode backend)
cannot reflect element of reflected container: reflected list(reflected list(reflected list(int64)))

回溯（最近一次呼叫最后一次）：
文件“”，第1行，在
tmp=L.numba_montecarlo_barab（种子，w，R，T，mpx，0.5）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\dispatcher.py”，第367行，位于\u compile\u for \u args中
提高e
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\dispatcher.py”，第324行，位于\u-compile\u中，用于\u-args
返回self.compile（元组（argtypes））
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler\u lock.py”，第32行，在\u acquire\u compile\u lock中
返回函数（*args，**kwargs）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\dispatcher.py”，第655行，编译
cres=self.\u compiler.compile（参数，返回类型）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\dispatcher.py”，第82行，编译
管道类=自管道类）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第926行，在compile\u extra中
返回管道.compile_extra（func）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第374行，在compile\u extra中
返回self.\u编译\u字节码（）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第857行，位于编译字节码中
返回self.\u编译\u核心（）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第844行，在编译内核中
res=pm.run（self.status）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler\u lock.py”，第32行，在\u acquire\u compile\u lock中
返回函数（*args，**kwargs）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第255行，正在运行
引发修补的\u异常
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第246行，正在运行
阶段（
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第717行，位于stage\u nopython\u后端
self.\u后端（lowerfn，objectmode=False）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第666行，在_backend中
降低=降低fn（）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第653行，处于后端模式
self.com（元数据）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\compiler.py”，第1050行，在本机阶段
降低。创建\u cpython\u包装（flags.release\u gil）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\looving.py”，第272行，位于create\U cpython\U包装器中
释放（gil=释放（gil）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\targets\cpu.py”，第149行，位于create\U cpython\U包装器中
builder.build（）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\callwrapper.py”，第122行，内部版本
self.build_包装器（api、生成器、闭包、参数、kws）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\callwrapper.py”，第155行，在build\u wrapper中
val=cleanup\u manager.add\u arg（builder.load（obj），ty）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\callwrapper.py”，第58行，在add_arg中
清理_arg（）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\callwrapper.py”，第41行，在cleanup\u arg中
self.api.reflect\u native\u值（ty、native.value、self.env\u管理器）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\pythonapi.py”，第1417行，在reflect\u native\u值中
impl（典型值、val、c）
文件“C:\ProgramData\Anaconda2\envs\Python 3.5\lib\site packages\numba\targets\boxing.py”，第740行，在反射列表中
raise TypeError（msg）
TypeError:在nopython模式管道中失败（步骤：nopython模式后端）
无法反映反射容器的元素：反射列表（反射列表（反射列表（int64）））

转换为元组也不起作用，但我知道元组不受支持

---

在numba nopython模式下使用高级索引的正确方法是什么？

我也有同样的问题。请尝试以下操作：

array[：，column_index].take（list）

。它返回与

array[np.array（list），column_index]

相同的错误，我刚刚运行了

a[：，1]。在
numba.jit下take（[1,2,3]）
，一切正常。当我做
a[[1,2,3]，1]
时，我得到了相同的错误。版本：numba:0.41.0；numpy 1.15.4；Python 3.7.1。所以，简单的索引和take对我来说很有用
a
是一个10x10形状的随机数组。是的，它确实应该可以工作。最后，我决定放弃高级索引，将所有内容放入for循环中，但错误依然存在。这是因为传递了一个列表作为参数，而numba目前不支持这个参数。非常感谢<代码>数组[np.array（list），column\u index]

可能也会起作用。