ulpwise 0.3.1

• ulpwise survey takes f16 and bf16 in --dtypes (the default stays f32,f64): torch and jax get half
  precision rows, measured in float16 or bfloat16 ulps against the same mpmath reference; numpy has no
  bfloat16 and scipy.special computes a float16 input in float32, so those two have no bf16 rows and scipy
  has no f16 rows. A backend without a kernel for a function in a dtype (torch's Bessel and Airy functions,
  erfcx, ndtri, log_ndtr and zeta in half precision) is logged and skipped before the reference is
  computed. The torch default tolerances of the two dtypes are (1.6e-2, 1e-5) and (1e-3, 1e-5), as in
  torch.testing. A log spaced domain wider than the dtype is now clipped to the dtype's finite range before
  the points are spaced: the float16 grid of sqrt and log over (1e-300, 1e300) kept 32 of 600 points and
  acosh 24, they have about 570 and 490 now; the float32 grids of the same functions go from about 100 points
  to about 600, the float64 grids are unchanged. The (0, 1) and (-1, 1) grids of ndtri, logit, asin,
  acos, atanh and erfinv stop at the float below 1 in half precision too, where 1 - 1e-4 and 1 - 1e-3
  rounded to 1. Two corrections to the ulp metric for every dtype: the threshold from which an exact result
  counts as a correct overflow to infinity was fmax * (1 + eps / 2), about one ulp above the largest finite
  value, and is now fmax + ulp / 2, the rounding threshold; and the spacing at the largest finite value was
  numpy's inf, which turned every finite result there into a 0 ulp error, it is ulpwise.spacing now.
  An unknown dtype raises ValueError instead of a KeyError from the grid.
• next_up, next_down, spacing, neighbours, binade_edges and all_floats accept f16 and bf16 (and the
  float16, half, bfloat16, torch.float16, torch.bfloat16 aliases) like ulp_distance and special already
  did; they raised unsupported dtype before. The 16 bit versions are pure Python on the bit patterns, with the
  conventions of the Rust ones: the largest float steps to infinity, both zeros to the smallest subnormal, spacing
  is the distance from |x| to the next float and stays finite at the largest float, neighbours returns finite
  values only. A test walks every finite float16 and bfloat16 and checks the six functions against each other, against
  numpy's float16 nextafter and spacing and against torch's bfloat16 nextafter.
• ulpwise scan read the first positional argument of a method call as the value the method is applied
  to, so log(exp(x).sum(-1)) was not a logsumexp-by-hand finding while log(exp(x).sum()) and
  log(exp(x).sum(dim=-1)) were. A method call whose first argument is a dimension (an integer, None
  or a tuple of them) is now read as applying to its receiver, like the argument-less form.
• ulpwise scan printed a SyntaxWarning (a DeprecationWarning before Python 3.12) for every invalid
  escape sequence in the files it scanned, four lines on kornia. The parse now ignores those warnings;
  a file that does not parse is skipped as before.
• ulpwise scan reported Windows paths with backslashes (kornia\geometry\conversions.py), so the same
  report read differently from the one made on Linux or macOS. Paths are now written with forward
  slashes on every platform; --run still imports the module from either form. The CI pytest job on
  ubuntu, macOS and windows runs tests/test_scan.py now, and that file also covers the reasons
  --run gives for a function it does not run (a class, a constant, an instance method, a builtin
  without a signature, a nested name, a module that fails to import) and the ulp comparison of the
  results across infinities, NaNs and an all zero reference.
• ulpwise survey without scipy installed died at ndtri with a ModuleNotFoundError and lost every
  row computed before it: the reference started its Newton iteration from scipy.special.ndtri. It now
  starts from a bisection on math.erfc, returns an exact zero at p = 0.5 (a Newton step there leaves a
  rounding residual, which the ulp metric would read as 1e16 ulps against the exact 0.0 of torch and scipy)
  and gives the same rows as before on the 600 point float64 and float32 grids. The survey also evaluates
  the reference only when some requested backend implements the function; a numpy only run no longer
  spends its time on references for the torch and scipy only entries. The pytest job on ubuntu, macOS and
  windows (Python 3.12 and 3.9) runs tests/test_survey.py now, with numpy as the only backend.
• ulpwise survey printed a numpy RuntimeWarning for every backend call that divided by zero or
  overflowed at the edge values of the grid (reciprocal and log at 0, reciprocal at the overflow
  edge), hundreds of lines on a full run. Those points are counted in the nonfinite column already, so
  the survey now evaluates the backends under np.errstate(all="ignore"), and a test runs reciprocal and
  log with warnings turned into errors.
• The README said that two kornia fixes in the corpus were merged but not released; the corpus has 31
  kornia cases now, and the pytorch/rl, peft and pytorch (#198006) fixes are in the same state (kornia
  0.8.3 and 0.9.0rc1, torchrl 0.14.0, peft 0.21.2 and torch 2.14.1 predate them or cherry-pick other
  changes), while the timm and ultralytics fixes are released. The sentence says so now, and
  tests/test_corpus.py checks that the README table has one row per case with the merge date from
  cases.json, so a case added without its row, or a row with a stale date, fails the suite.
• tests/test_survey.py checks every reference of the survey registry against the libraries it measures:
  each entry is surveyed on a 24 point float64 grid against torch, numpy and scipy and its median error must
  stay under 32 ulp, which a reference that is another function fails by fifteen orders of magnitude (torch's
  polygamma_1 is the worst true median at 9 ulp), and the piecewise references (selu, elu, entr,
  gelu_tanh, log_ndtr, lgamma, digamma, erfinv, ndtri, zeta, sinc, spherical_bessel_j0)
  are pinned once per branch, since a slip in one branch moves a third of the grid and keeps the median.
  survey.py goes from 77 to 94 percent covered.
• With torch installed but expecttest missing, torch.testing._internal does not import and
  ulpwise survey silently used the dtype default as the OpInfo tolerance, so the op tol columns
  looked like an override that was never read. The survey now logs one line with the import error and
  the survey extra installs expecttest. The regression corpus CI job, which has torch but had no
  expecttest, failed on the two op_db tests for the same reason; it installs expecttest now and the
  tests skip with the reason where op_db is not importable.
• ulpwise survey looked up torch's OpInfo tolerance by op name and took the first op_db entry
  with that name. polygamma has one entry per order, and the first, polygamma_n_0, has no
  override, so the op fail column of polygamma_1 and polygamma_2 in float32 was computed with
  the dtype default (rtol 1.3e-6, atol 1e-5) instead of torch's polygamma_n_1 and _n_2
  override (rtol 0.01, atol 1e-4). Entry gains opinfo_variant, torch_opinfo_tolerance
  takes a variant and otherwise prefers the base variant, and the two registry entries name
  theirs. In studies/accuracy-survey-2026-09 the polygamma_1 f32 row goes from 86 to 7 inputs
  failing the op's tolerance; the measured errors are unchanged (erratum in its README).
• tests/test_survey.py covers the survey itself now: survey() on sqrt against numpy and torch
  (numpy within half an ulp, torch under one), an inexact backend, a backend that raises or returns
  the wrong shape, the backend resolution, the op_db tolerance lookup including the variant case,
  the csv and markdown writers, versions(), main() and the survey subcommand. CI installs
  mpmath for the Python job so these tests run there instead of being skipped.
• Six more kornia cases, each present on kornia 0.8.3 and fixed on main: the Hessian of So3.exp
  at the identity (#4972, nine nan for -I / 4 and zeros), sampson_epipolar_distance of a point on
  its epiline with squared=False (#5116, sqrt(eps) = 1e-4 for 0) and of the same F scaled by
  1e-4 (#5116, a third lower), RandomHue on a float64 image (#5131, 8.7e-8 rad past a half turn,
  the float32 rounding error of pi), MS_SSIMLoss(sigmas=(0.5, 1.3), reduction="none") (#5143, an
  even 6-pixel window and a (1, 15, 19) map for a (1, 3, 16, 20) input) and MS_SSIMLoss on a
  uint8 pair (#5353, expected scalar type Byte but found Float; with data_range=255 it now scores
  the pair divided by 255 at the default to 1e-6). Six more: So3.right_jacobian of a float16
  45 rad rotation (#4967, the identity for a matrix of 0.2 to 0.4), average_quaternions with a member
  stored as 3 q (#4980, 41.8 degrees for the 22.5 degree bisector), Hyperplane.through a float16
  triangle with legs of 300 (#5104, the normal (0, 0, 1) for (0, 0, -1)), filter2d with a
  per-sample kernel on a channels-last image (#5301, torch's view error), lovasz_hinge_loss of a
  float16 prediction (#5303, a float32 loss) and get_box_kernel1d (#5357, a stride-0 view where
  one write zeroed all three taps). The kornia block is 31 cases, all present on 0.8.3 and all
  fixed on main. One more ultralytics case, scale_masks with the dataloader's ratio_pad and an
  odd letterbox padding (#26379, a padded row survived the crop and the bottom rows of every mask
  faded to 0) and verify_image_label on two triangles that tile one square (#26377, the second
  polygon was dropped as a duplicate of the first because only the class and the box were compared);
  both carry fixed_in_release 8.4.165, the first tag with the fixes), and verify_image_label on a
  pose row for a detect task (#26357, the row was read as a polygon, its box was out of bounds and
  the image was dropped as corrupt; fixed_in_release 8.4.164). The corpus is 50 cases.
• The manual workflow_dispatch run of CI is now strict about the corpus like the Monday run (the
  changelog said so already, the workflow set the variable for schedule only).
• The release workflow's manual dry run (workflow_dispatch without publish) now ends in a
  collect job that downloads the artifacts the way the publish job does and checks that there
  are five wheels and one sdist, so a dry run covers the whole pipeline short of the upload. The
  workflow actions moved to actions/checkout@v7, setup-python@v7, upload-artifact@v7 and
  download-artifact@v8 (the first Dependabot pull requests), and Cargo.lock to pyo3 0.29.3.
• Dependabot (.github/dependabot.yml) opens weekly pull requests for Cargo.lock and the
  workflow actions, so pyo3 patch releases land through CI-tested pull requests instead of a
  hand-run cargo update; Python dependencies stay unpinned on purpose.
• Eight more kornia cases, each present on kornia 0.8.3 and fixed on main: Se2.exp/Se2.log at
  theta = 1e-8 (#4960, the translation came back as (1, 2) from exp and (1e-8, -5e-9) from log),
  point_line_distance of a homogeneous point with w = 2 (#4975, 4.0 for 1.5), Quaternion.__pow__
  of -1 (#5003, the zero quaternion), So2 from a (B, 1) angle times (B, 2) points (#5005,
  (B, B, 2)), solve_cubic of 1e-30 x^3 + 2x - 6 (#5024, roots [0, 0, 0] for 3),
  RgbToGrayscale on uint8 (#5111, all zeros), conv_soft_argmax2d with a far-away peak (#5134,
  the weak window's coordinates moved from 0.8834 to 1.0) and get_gaussian_discrete_kernel1d(1, sigma) (#5376, 3 taps). Six more in the same shape: Se3.exp's d t / d omega at the identity
  (#4963, nan on 0.8.3, zero on main before the fix), Quaternion.polar_angle's gradient at the
  identity (#4981, nan), decompose_essential_matrix of a (3, 3) input (#4998, (1, 3, 3)),
  Hyperplane.through's gradients for orthogonal equal-length edges (#5058, nan),
  Vector3.normalized of a float16 zero vector (#5084, NaN) and otsu_threshold's mask below
  zero (#5182, all False). The whole kornia block of the corpus, 19 cases, reads present on 0.8.3;
  the mean_average_precision recall thresholds (#5101) were checked and left out because 0.8.3
  already scores the exact-tenth case right.
• CI runs the regression corpus every Monday (and on workflow_dispatch) with
  ULPWISE_CORPUS_STRICT=1, which turns an unexpected pass into a failure: a release fixed a case
  and fixed_in_release is stale. Push and pull request runs stay non strict. The corpus job also
  installs ultralytics, so the ultralytics case runs there instead of being skipped.
• Corpus metadata: kornia #4838 is fixed by kornia #5124 (merged 2026-09-30) and kornia #4897 by
  kornia #4941 (merged 2026-09-26); both cases now carry the pr and merged_at, and stay expected
  failures because no kornia release after 0.9.0rc1 (2026-07-19) exists yet. ultralytics #26330 sets
  fixed_in_release to 8.4.164: the tag v8.4.164 (2026-09-27, the first release after the merge)
  carries the use_obb check in verify_labels and v8.4.163 does not, so the case is a hard
  failure, not an expected one, from 8.4.164 on. peft #3777 keeps fixed_in_release: null: the
  v0.21.1 and v0.21.2 tags still carry the weight.size()[2:4] shortcut in lora/layer.py.