`np.multiply` vs. `*`

In the course of our work over on dipy, we stumbled into the following behavior:

The following line: https://github.com/nipy/dipy/blob/master/dipy/denoise/shift_twist_convolution.pyx#L68

raises this error (on np 1.11, but not earlier):

Traceback (most recent call last):
  File "contextual_enhancement.py", line 149, in <module>
    sh_order=8, test_mode=True)
  File "dipy/denoise/shift_twist_convolution.pyx", line 68, in dipy.denoise.shift_twist_convolution.convolve (dipy/denoise/shift_twist_convolution.c:2147)
TypeError: 'numpy.float64' object cannot be interpreted as an index

When changed to this:

https://github.com/stephanmeesters/dipy/blob/enhance_5d_update_pr/dipy/denoise/shift_twist_convolution.pyx#L69

This error no longer occurs. What does the error mean (it’s a bit opaque in this context), and how do we avoid it in the future?

Thanks!

[seberg]

The reason should be that output is actually a sequence and not an ndarray but you actually

[seberg]

Close it, as the last comment said:

[1, 2, 3] * np.float64(4.1)

raises and mismatches with 0-D arrays. But that is tricky and most of the discussion is about things that are not relevant anymore. We should just open a new issue when we run into it again.