NumPy

• Indexing

  • SciPy.org, Indexing

• where

  • SciPy.org, numpy.where

  • PinkWink, numpy의 where 함수 사용법

• 2d list ￫ numpy array

  • numpy.array()

    • 코드

    a_list = [[1, 2], [3, 4]]
    an_array = np.array(a_list)
    
    print(an_array)
    

    • 출력

    [[1 2]
    [3 4]]
    

    • 참조

      • kite, How to convert a 2D list into a NumPy array in Python

• numpy array ￫ list

  • tolist()

    • 코드

      • 1D array

      >>> a = np.array([1, 2])
      >>> list(a)
      [1, 2]
      >>> a.tolist()
      [1, 2]
      

      • 2D array

      >>> a = np.array([[1, 2], [3, 4]])
      >>> list(a)
      [array([1, 2]), array([3, 4])]
      >>> a.tolist()
      [[1, 2], [3, 4]]
      

    • 참조

      • SciPy.org, numpy.ndarray.tolist

• Copy numpy array into higher dimensions

  • np.expand_dims and np.concatenate

    • 코드

    arr = np.expand_dims(arr, axis=2)
    arr = np.concatenate((arr,arr,arr), axis=2)
    

    • 참조

      • StackOverflow, how to copy numpy array value into higher dimensions

  • np.repeat

    • 코드

    arr3D = np.repeat(arr[...,None], 3, axis=2)
    

    • 참조

      • StackOverflow, how to copy numpy array value into higher dimensions