Newest Version of Numpy does not work [have to degrade to 1.16.0]

[claytonstrawn]

Bug report

Bug summary

If you upgrade to the most recent numpy version, yt.load does not function correctly

Code for reproduction

pip install --upgrade numpy
python

import yt
yt.load('/Users/claytonstrawn/VELA_v2_27/10MpcBox_csf512_a0.500.d')

Actual outcome

yt : [INFO     ] 2020-06-18 16:56:12,637 discovered particle_data:/Users/claytonstrawn/VELA_v2_27/PMcrs0a0.500.DAT
yt : [INFO     ] 2020-06-18 16:56:12,638 discovered particle_stars:/Users/claytonstrawn/VELA_v2_27/stars_a0.500.dat
yt : [INFO     ] 2020-06-18 16:56:12,665 Using root level of 13
yt : [INFO     ] 2020-06-18 16:56:12,667 Discovered 6 species of particles
yt : [INFO     ] 2020-06-18 16:56:12,667 Particle populations:   5287936    634880    106752     17168   2090807   1939360 
Traceback (most recent call last):
  File "/Users/claytonstrawn/anaconda2/envs/myenv_testforinterns/lib/python3.6/site-packages/numpy/core/function_base.py", line 117, in linspace
    num = operator.index(num)
TypeError: 'float' object cannot be interpreted as an integer

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/convenience.py", line 70, in load
    return candidates[0](*args, **kwargs)
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/frontends/art/data_structures.py", line 175, in __init__
    unit_system=unit_system)
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/data_objects/static_output.py", line 307, in __init__
    self._parse_parameter_file()
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/frontends/art/data_structures.py", line 356, in _parse_parameter_file
    self.current_time = self.quan(b2t(self.parameters['t']), 'Gyr')
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/frontends/art/io.py", line 612, in b2t
    return a2t(b2a(tb))
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/frontends/art/io.py", line 591, in b2a
    return find_root(f_b2a, 1e-4, 1.1)
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/frontends/art/io.py", line 558, in find_root
    assert(sign(f(a)) != sign(f(b)))
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/frontends/art/io.py", line 590, in <lambda>
    f_b2a = lambda at: a2b(at, **kwargs)-bt
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/frontends/art/io.py", line 582, in a2b
    val = quad(f_a2b, 1, at)
  File "/Users/claytonstrawn/yt_trident_temp/yt/yt/frontends/art/io.py", line 570, in quad
    spacings = np.logspace(np.log10(xmin), np.log10(xmax), n)
  File "<__array_function__ internals>", line 6, in logspace
  File "/Users/claytonstrawn/anaconda2/envs/myenv_testforinterns/lib/python3.6/site-packages/numpy/core/function_base.py", line 272, in logspace
    y = linspace(start, stop, num=num, endpoint=endpoint, axis=axis)
  File "<__array_function__ internals>", line 6, in linspace
  File "/Users/claytonstrawn/anaconda2/envs/myenv_testforinterns/lib/python3.6/site-packages/numpy/core/function_base.py", line 121, in linspace
    .format(type(num)))
TypeError: object of type <class 'float'> cannot be safely interpreted as an integer.

This seems to be the same on all branches of yt. I was able to fix it by downgrading to numpy 1.16.0 so I know it's a numpy dependency.

[munkm]

Hi Clayton,

I've verified this issue on my machine in the art frontend.

I'm working on a fix. It looks like we have a similar issue as one of the fixes applied in #2448. I'll tag you when I submit the PR.

[neutrinoceros]

note, this also looks related to #2419