使用 np.ndarray 调用函数说“类型错误：缺少 1 个必需的位置参数：”

Question

I'm writing the code to rotate matrix(1x3) on Python 3.5.2.

To rotate matrix, I made a function.(Not a Class)

def rotate_rpy(posvec: Vector, rotvec: Vector) -> Vector :
  return np.dot(np.dot(np.dot (posvec, rotate_rpy(rotvec[0]) ), rotate_pitch(rotvec[1]) ), rotate_yaw(rotvec[2]))

newpose = rotate_rpy(mypose, rotateang)#enbug

it is getting an Error like this

Traceback (most recent call last): 
  File "rotation_matrix.py", line 50, in <module>
    newpose = rotate_rpy(mypose, rotateang)#enbug
  File "rotation_matrix.py", line 35, in rotate_rpy
    return np.dot(np.dot(np.dot (posvec, rotate_rpy(rotvec[0]) ), rotate_pitch(rotvec[1]) ), rotate_yaw(rotvec[2]))
TypeError: rotate_rpy() missing 1 required positional argument: 'rotvec'

maybe a silly question, but I dont Understand what is the problem.

Most question on the Internet says something about instantiate, but it is just a function. not class. Anyways I just tried changing argument:

newpose = rotate_rpy(mypose, mypose, rotateang)#enbug

then

TypeError: rotate_rpy() takes 2 positional arguments but 3 were given

and

newpose = rotate_rpy(mypose)#enbug

then

TypeError: rotate_rpy() missing 1 required positional argument: 'rotvec'

I trust it silly but

newpose = rotate_rpy()#enbug

then

TypeError: rotate_rpy() missing 2 required positional arguments: 'posvec' and 'rotvec'

I'm confuced

3 arg - too much arg

2 arg - 1 arg misssing

1 arg - 1 arg missing

0 arg - 2 arg missing

It isn't corresponding. I have no idea. Please help me...

Full Code Below:

#coding:utf-8

import numpy as np

def rotate_roll (th):
  _rot_vec_roll = {
    { 1.,          0. ,         0.},
    { 0.,   np.cos(th), np.sin(th)},
    { 0., - np.sin(th), np.cos(th)}
  }
  return _rot_vec_roll

def rotate_pitch (th):
  _rot_vec_pitch = {
    {  np.cos(th),0. , np.sin(th)},
    {  0.,        1.,          0.},
    {- np.sin(th),1.,  np.cos(th)}
  }
  return _rot_vec_pitch

def rotate_yaw (th):
  _rot_vec_yaw = {
    {   np.cos(th), np.sin(th), 0.},
    { - np.sin(th), np.cos(th), 0.},
    {           0.,         0., 1.}
  }
  return _rot_vec_yaw

# R2 = A * R1 
# A = roll_vec * pitch_vec * yaw_vec

Vector = np.ndarray(shape=(1,3), dtype=np.float)

def rotate_rpy(posvec: Vector, rotvec: Vector) -> Vector :
  return np.dot(np.dot(np.dot (posvec, rotate_rpy(rotvec[0]) ), rotate_pitch(rotvec[1]) ), rotate_yaw(rotvec[2]))

mypose = np.ndarray(shape=(1,3), dtype=np.float)
mypose = np.array([3.0,1.0,1.0], dtype=float)

print(mypose)

base = np.pi / 6.0

rotateang = np.ndarray(shape=(1,3), dtype=np.float)
rotateang = np.array([base, base/2.0, base], dtype=float)

print(rotateang)

newpose = np.ndarray(shape=(1,3), dtype=np.float)
newpose = rotate_rpy(mypose, rotateang)#enbug
print(newpose);

and full error below:

[ 3.  1.  1.]
[ 0.52359878  0.26179939  0.52359878]
Traceback (most recent call last):
  File "rotation_matrix.py", line 50, in <module>
    newpose = rotate_rpy(mypose, rotateang)#enbug
  File "rotation_matrix.py", line 35, in rotate_rpy
    return np.dot(np.dot(np.dot (posvec, rotate_rpy(rotvec[0]) ), rotate_pitch(rotvec[1]) ), rotate_yaw(rotvec[2]))
TypeError: rotate_rpy() missing 1 required positional argument: 'rotvec'

Answer 1

The function rotate_rpy is defined to call itself infinitely recursively (which is a problem on its own):

def rotate_rpy(posvec, rotvec):
    return np.dot(np.dot(np.dot (posvec, rotate_rpy(rotvec[0]), ....)

It also calls itself inconsistently, with only one parameter. Python interpreter cannot catch the first error by fortunately reports the second error.