Replace np.sctypes for numpy 2.0 compat #1250

mscheltienne · 2023-08-28T13:38:09Z

With nightly numpy release, the following attribute error is raised:

AttributeError: `np.sctypes` was removed in the NumPy 2.0 release. . Did you mean: 'dtypes'?

corresponding to this bullet point in the release notes:

np.issctype, np.maximum_sctype, np.obj2sctype, np.sctype2char, np.sctypeDict, np.sctypes, np.issubsctype were all removed from the main namespace without replacement, as they where niche members.

This PR replaces:

np.sctypes calls with np.core.sctypes
arr.newbyteorder with arr.view(arr.dtype.newbyteorder())
some implicit casts when multiplying by python int to explicit casts then multiply
np.can_cast usage

codecov · 2023-08-28T14:01:27Z

Codecov Report

Attention: 3 lines in your changes are missing coverage. Please review.

Comparison is base (4588058) 92.22% compared to head (a71eebf) 92.25%.
Report is 2 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1250      +/-   ##
==========================================
+ Coverage   92.22%   92.25%   +0.02%     
==========================================
  Files          99       99              
  Lines       12454    12466      +12     
  Branches     2559     2564       +5     
==========================================
+ Hits        11486    11500      +14     
+ Misses        644      643       -1     
+ Partials      324      323       -1

Files	Coverage Δ
nibabel/conftest.py	`100.00% <100.00%> (ø)`
nibabel/ecat.py	`88.21% <100.00%> (ø)`
nibabel/freesurfer/io.py	`94.24% <100.00%> (ø)`
nibabel/quaternions.py	`99.02% <100.00%> (+<0.01%)`	⬆️
nibabel/spatialimages.py	`94.14% <100.00%> (+0.02%)`	⬆️
nibabel/streamlines/trk.py	`93.98% <100.00%> (ø)`
nibabel/nifti1.py	`92.79% <88.88%> (+0.03%)`	⬆️
nibabel/casting.py	`87.23% <77.77%> (+0.85%)`	⬆️

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

effigies · 2023-08-28T14:52:26Z

Thank you! I'm a little bit dreading cleaning up the tests for numpy 2 (#1242), but happy to take preemptive fixes that don't break existing numpy.

Would you care to add yourself to the Zenodo and author list?

effigies · 2023-08-28T14:55:01Z

Actually, would you mind running pipx run blue nibabel to sort out the styling?

larsoner · 2023-08-28T15:12:31Z

I suspect this is just a workaround since the release notes say:

... np.sctypes, np.issubsctype were all removed from the main namespace without replacement, as they where niche members.

So really it seems like the code should be updated not to use sctypes at all :(

mscheltienne · 2023-08-28T15:28:55Z

I don't read this as a deprecation of sctypes entirely, rather as a removal of the "shortcut" np.sctypes, since they explicitly mention namespace.

effigies · 2023-08-28T15:31:09Z

I asked on: numpy/numpy#24376

Not sure how likely they are to see a question on a closed PR, so might ask again in a couple days if we don't get a response.

mscheltienne · 2023-08-28T15:34:05Z

Perfect, at least that will be a clear answer :)

larsoner · 2023-08-28T15:34:22Z

I don't read this as a deprecation of sctypes entirely, rather as a removal of the "shortcut" np.sctypes, since they explicitly mention namespace.

Looks like np.core should not be used

numpy/numpy#24376 (comment)

mscheltienne · 2023-08-28T15:36:14Z

Well that was quick. I don't have time today, but I can have a look tomorrow to replace sctypes with something else.

mscheltienne · 2023-08-29T11:27:03Z

nibabel/casting.py

+# np.sctypes is deprecated in numpy 2.0 and np.core.sctypes should not be used instead.
+sctypes = {
+    "int": [np.int8, np.int16, np.int32, np.int64],
+    "uint": [np.uint8, np.uint16, np.uint32, np.uint64],
+    "float": [np.float16, np.float32, np.float64, np.float128],
+    "complex": [np.complex64, np.complex128, np.complex256],
+    "others": [bool, object, bytes, str, np.void],
+}
+


I'm not entirely convince this is enough to keep a working status on all platforms. It looks to me like np.sctypes is defined per platform, and if this is correct, we might need to add some logic to define the mapping between strings and scalar types per platform as well.

This reverts commit 6c30a84.

mscheltienne · 2023-08-29T15:38:20Z

nibabel/casting.py

+sctypes = {
+    'int': [np.int8, np.int16, np.int32, np.int64],
+    'uint': [np.uint8, np.uint16, np.uint32, np.uint64],
+    'float': [np.float16, np.float32, np.float64, np.longdouble],
+    'complex': [np.complex64, np.complex128, np.complex256],
+    'others': [bool, object, bytes, str, np.void],
+}


I'm out of idea to correctly build this mapping dict on all platforms. I never worked with longlong, longdouble, clongdouble and I am not familiar with the issues that are attached to them. A couple of ideas:

Don't use the size aliases and let numpy figure that out (6c30a84). It should work, but compared to that commit the list(set(...)) should be sorted by precision. One point eluded me however: np.byte correctly points to np.int8, np.ubyte to np.uint8, and so on.. except for np.longlong and np.ulonglong which points to themselves. np.ulonglong == np.uint64 is False and I guess that would cause issues.

create the mapping based on attributes, e.g. hasattr(np, "float128").

Any good idea?

…2 to the list

effigies · 2023-09-11T08:14:56Z

Hi, thanks for this, and I think this will work. I need to find a little time to think clearly about this problem. Just to let you know that I haven't forgotten about this, just a little slow to get back to it.

larsoner · 2023-09-12T18:10:14Z

Looks like this branch is getting out of date, trying to use it in MNE I get:

/opt/hostedtoolcache/Python/3.11.5/x64/lib/python3.11/site-packages/nibabel/quaternions.py:32: in <module>
    MAX_FLOAT = np.maximum_sctype(float)
/opt/hostedtoolcache/Python/3.11.5/x64/lib/python3.11/site-packages/numpy/__init__.py:383: in __getattr__
    raise AttributeError(
E   AttributeError: `np.maximum_sctype` was removed in the NumPy 2.0 release.

nibabel/nifti1.py

larsoner · 2023-10-26T19:57:48Z

Green! (Other than some coverage misses that I think probably exist in master as well?)

@effigies feel free to review/merge if you're happy, only ended up at +170 −97

nibabel/casting.py

nibabel/conftest.py

nibabel/quaternions.py

effigies · 2023-10-31T15:30:50Z

nibabel/tests/test_floating.py

+    # TODO: Should this actually still overflow? Does it matter?
+    if FP_OVERFLOW_WARN:
+        ctx = pytest.raises(OverflowError)
+    else:
+        ctx = nullcontext()
+    with ctx:
        as_int(val)
-    with pytest.raises(OverflowError):
+    with ctx:
        as_int(-val)


What happens with numpy 2.0? This seems like it should still fail.

It does not overflow. Agreed it's weird that it does not

nibabel/tests/test_volumeutils.py

effigies · 2023-10-31T16:15:55Z

nibabel/tests/test_volumeutils.py

+    # This is the normal rule - no upcast from Python scalar
+    # (on NumPy 2.0 it *will* upcast from a np.float64 scalar!)
+    assert (f32_arr * 1.).dtype == np.float32
+    assert (f32_arr + 1.).dtype == np.float32


The goal here seems to be contrasting to numpy casting rules, not numpy + Python scalar casting rules. I would just branch on numpy version about the expected dtype for using f64(...).

Co-authored-by: Chris Markiewicz <[email protected]>

effigies · 2023-11-02T14:21:35Z

Okay, I'm starting to understand what's going on here. Numpy previously had broken behavior around longdoubles and ints. If you install Python 3.6 and numpy 1.11, you get things like:

In [1]: import numpy as np

In [2]: val = np.longdouble(2**1024) * 2
---------------------------------------------------------------------------
OverflowError                             Traceback (most recent call last)
<ipython-input-2-4bc60941a286> in <module>
----> 1 val = np.longdouble(2**1024) * 2

OverflowError: int too large to convert to float

In [6]: val = np.longdouble(2**1023)

In [7]: val
Out[7]: 8.9884656743115795386e+307

In [8]: val * 2
Out[8]: 1.7976931348623159077e+308

In [9]: val * 4
Out[9]: 3.5953862697246318155e+308

In [10]: int(val * 4)
---------------------------------------------------------------------------
OverflowError                             Traceback (most recent call last)
<ipython-input-10-af59c627c981> in <module>
----> 1 int(val * 4)

OverflowError: cannot convert float infinity to integer

However, if you use Python 3.8 and numpy 1.20 (nibabel current minimum requirements):

In [1]: import numpy as np

In [2]: nexp64 = 1024

In [3]: val = np.longdouble(2**1024) * 2

In [4]: val
Out[4]: 3.5953862697246318155e+308

In [5]: int(val)
Out[5]: 359538626972463181545861038157804946723595395788461314546860162315465351611001926265416954644815072042240227759742786715317579537628833244985694861278948248755535786849730970552604439202492188238906165904170011537676301364684925762947826221081654474326701021369172596479894491876959432609670712659248448274432

I think at this point, probably the correct thing is just to get rid of int_to_float and as_int. But that's not needed for this PR.

further info: numpy/numpy#17325 nipy/nibabel#1250

mscheltienne added 2 commits August 28, 2023 15:35

replace np.sctypes with np.core.sctypes

2623523

rm unused imports

d550152

mscheltienne changed the title ~~Replace np.sctypes with np.core.sctypes~~ Replace np.sctypes with np.core.sctypes for numpy 2.0 compat Aug 28, 2023

run blue instead of black

6e873c6

add author entries

0f5ad6e

mscheltienne added 3 commits August 29, 2023 13:24

manually define the mapping between str and scalar types in casting.py

1dac328

rm unused imports

7966160

rm unused variable definition

07cea85

mscheltienne commented Aug 29, 2023

View reviewed changes

fix blue

d3d23db

mscheltienne changed the title ~~Replace np.sctypes with np.core.sctypes for numpy 2.0 compat~~ Replace np.sctypes for numpy 2.0 compat Aug 29, 2023

mscheltienne added 4 commits August 29, 2023 15:55

fix missing import

ac9e16f

try without using the sized aliases

6c30a84

Revert "try without using the sized aliases"

3639711

This reverts commit 6c30a84.

try with sized aliases again and np.longdouble instead of np.float128

4cdba01

mscheltienne commented Aug 29, 2023

View reviewed changes

use combination of getattr and hasattr, include float96 and complex19…

2398cc3

…2 to the list

mscheltienne added 2 commits September 12, 2023 20:30

rm use of np.maximum_sctype

fcd8dd0

rm unused import

edd95db

mscheltienne added 2 commits October 26, 2023 13:36

more fix

97e3aa9

fix more stuff

a765af0

larsoner reviewed Oct 26, 2023

View reviewed changes

nibabel/nifti1.py Outdated Show resolved Hide resolved

larsoner added 13 commits October 26, 2023 10:34

FIX: check

df96ae3

FIX: check

83111ea

FIX: Python types

6ffea1b

FIX: Preserve

86b0597

FIX: Simplify

fbbd801

FIX: Maybe

4630e0d

FIX: Better

49b1d41

Merge branch 'master' into np.sctypes

e61e7a0

FIX: Revert

5eb5e54

FIX: ComplexWarning

a1ddae8

FIX: Context

cd362aa

FIX: One more

c32b0d2

FIX: Explicit

d4596b7

effigies reviewed Oct 31, 2023

View reviewed changes

larsoner and others added 5 commits November 1, 2023 14:49

Apply suggestions from code review

0f746c0

Co-authored-by: Chris Markiewicz <[email protected]>

FIX: Style

e3c72e1

STY: Flake

e3a7495

FIX: Test val equiv

1bc593c

FIX: Version

a71eebf

effigies merged commit dabfc46 into nipy:master Nov 2, 2023
51 of 52 checks passed

mscheltienne deleted the np.sctypes branch November 2, 2023 14:26

MarDiehl added a commit to MarDiehl/DAMASK that referenced this pull request Jun 19, 2024

compatible with NumPy 2.0

0a8fd23

further info: numpy/numpy#17325 nipy/nibabel#1250

MarDiehl added a commit to MarDiehl/DAMASK that referenced this pull request Jun 19, 2024

compatible with NumPy 2.0

1bf4ed6

further info: numpy/numpy#17325 nipy/nibabel#1250

MarDiehl added a commit to MarDiehl/DAMASK that referenced this pull request Jun 19, 2024

compatible with NumPy 2.0

ca5a1a3

further info: numpy/numpy#17325 nipy/nibabel#1250

MarDiehl added a commit to MarDiehl/DAMASK that referenced this pull request Jun 19, 2024

compatible with NumPy 2.0

6bb914e

further info: numpy/numpy#17325 nipy/nibabel#1250

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Replace np.sctypes for numpy 2.0 compat #1250

Replace np.sctypes for numpy 2.0 compat #1250

mscheltienne commented Aug 28, 2023 •

edited by larsoner

Loading

codecov bot commented Aug 28, 2023 •

edited

Loading

effigies commented Aug 28, 2023

effigies commented Aug 28, 2023

larsoner commented Aug 28, 2023

mscheltienne commented Aug 28, 2023

effigies commented Aug 28, 2023

mscheltienne commented Aug 28, 2023

larsoner commented Aug 28, 2023

mscheltienne commented Aug 28, 2023

mscheltienne Aug 29, 2023

mscheltienne Aug 29, 2023

effigies commented Sep 11, 2023

larsoner commented Sep 12, 2023

larsoner commented Oct 26, 2023

effigies Oct 31, 2023

larsoner Nov 1, 2023

effigies Oct 31, 2023

effigies commented Nov 2, 2023

Replace np.sctypes for numpy 2.0 compat #1250

Replace np.sctypes for numpy 2.0 compat #1250

Conversation

mscheltienne commented Aug 28, 2023 • edited by larsoner Loading

codecov bot commented Aug 28, 2023 • edited Loading

Codecov Report

effigies commented Aug 28, 2023

effigies commented Aug 28, 2023

larsoner commented Aug 28, 2023

mscheltienne commented Aug 28, 2023

effigies commented Aug 28, 2023

mscheltienne commented Aug 28, 2023

larsoner commented Aug 28, 2023

mscheltienne commented Aug 28, 2023

mscheltienne Aug 29, 2023

Choose a reason for hiding this comment

mscheltienne Aug 29, 2023

Choose a reason for hiding this comment

effigies commented Sep 11, 2023

larsoner commented Sep 12, 2023

larsoner commented Oct 26, 2023

effigies Oct 31, 2023

Choose a reason for hiding this comment

larsoner Nov 1, 2023

Choose a reason for hiding this comment

effigies Oct 31, 2023

Choose a reason for hiding this comment

effigies commented Nov 2, 2023

mscheltienne commented Aug 28, 2023 •

edited by larsoner

Loading

codecov bot commented Aug 28, 2023 •

edited

Loading