• Implemented using arraylist, so append is
    • Thus, use python lists until you know for sure when the array won’t grow anymore, and convert to numpy array.
  • {python}arr = np.zeros() or {python}arr = np.empty() to initialize.
    • ! np.arrange(3) initializes an int, while np.arrange(3.) initilizes a float.
  • {python}arr = np.zeros_like(ref_arr) will initiallize arr with the same dimensions and datatype as ref_arr:
  • Sequence based initialization:
    • Linspace vs arrange differences. Conclusion: linspace is better:
  • Indexing arrays (is a pointer, except the “fancy indexing”)