Как я могу решить эту проблему с Numba - неправильное использование функции

Question

Я столкнулся с проблемой, которую не понимаю при использовании пакета numba python. Ошибка говорит:

TypingError: Invalid use of Function(<built-in function setitem>) with argument(s) of type(s): (array(float64, 3d, C), array(bool, 3d, C), Literal[int](0))
 * 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:
    TypeError: unsupported array index type array(bool, 3d, C) in [array(bool, 3d, C)]
    raised from C:\Users\sw14928\AppData\Local\Continuum\anaconda3\lib\site-packages\numba\typing\arraydecl.py:71
In definition 5:
    TypeError: unsupported array index type array(bool, 3d, C) in [array(bool, 3d, C)]
    raised from C:\Users\sw14928\AppData\Local\Continuum\anaconda3\lib\site-packages\numba\typing\arraydecl.py:71
In definition 6:
    All templates rejected with literals.
In definition 7:
    All templates rejected without literals.
This error is usually caused by passing an argument of a type that is unsupported by the named function.

Мой код выглядит так:

@njit
def kaczmarz(finalArray,responseArray):
    #KaczResult =  np.zeros_like(finalArray)#convolve(sourceOntable, response, method = 'direct', mode='same')
    solution = np.zeros_like(finalArray)
    #solution = finalArray
    #remnants =  np.zeros_like(finalArray)
    widthFinalArray = finalArray.shape[0]
    depthFinalArray = finalArray.shape[1]
    widthRes = responseArray.shape[0]
    depthRes = responseArray.shape[1]
    xRad = int(widthRes/ 2)
    yRad = int(depthRes/ 2)
    #usefulResponseIndicies = response>0
    rawValues = widthFinalArray * depthFinalArray * finalArray.shape[2]
    damping = 0.01
    for i in range(5000):
        print(i)
        update = np.zeros_like(solution)
        count = 0
        for x in range(finalArray.shape[0]):
            for y in range(finalArray.shape[1]):
                #randomPoint = np.random.randint(rawValues)
                #coords =  np.unravel_index(randomPoint, finalArray.shape)
                #x = coords[0]
                #y = coords[1]

                x = np.random.randint(finalArray.shape[0])
                y = np.random.randint(finalArray.shape[1])
                #print(x,y,z)

                x1 = x-xRad
                if x1 < 0: x1 = 0
                if x1 > widthFinalArray: x1 = widthFinalArray

                x2 = xRad+x
                if x2 < 0: x2 = 0
                if x2 > widthFinalArray: x2 = widthFinalArray

                x3 = xRad-x
                if x3 < 0: x3 = 0
                if x3 > widthRes: x3 = widthRes

                x4 = xRad-x+widthFinalArray
                if x4 < 0: x4 = 0
                if x4 > widthRes: x4 = widthRes

                y1 = y-yRad
                if y1 < 0: y1 = 0
                if y1 > depthFinalArray: y1 = depthFinalArray

                y2 = yRad+y
                if y2 < 0: y2 = 0
                if y2 > depthFinalArray: y2 = depthFinalArray

                y3 = yRad-y
                if y3 < 0: y3 = 0
                if y3 > depthRes: y3 = depthRes

                y4 = yRad-y+depthFinalArray
                if y4 < 0: y4 = 0
                if y4 > depthRes: y4 = depthRes

                subSection = solution[x1:x2+1, y1:y2,0]
                #shape = np.shape(subSection)
                #test = np.sum(subSection)

                subResponse = responseArray[x3:x4,y3:y4]
                KaczResult = np.sum(subResponse * subSection)
                remnant = finalArray[x,y,0]-KaczResult

                subUpdate = subResponse*remnant

                #print(np.max(subResponse))
                #print(np.max(subResponse))
                update[x1:x2+1, y1:y2,0] += subUpdate
                count += 1

                solution[solution < 0] = 0
        solution += damping*update/count

    return solution