ponder-lab / Hybridize-Functions-Refactoring / #2513
82%
main: 83%

LAST BUILD BRANCH: dependabot/github_actions/actions/setup-python-7
DEFAULT BRANCH: main
Ran
Jobs 1
Files 33
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 81.766%. Remained the same
#2513

push

github

web-flow 
Bump Ariadne to 0.52.31 (#779)

* Bump Ariadne to 0.52.30

0.52.30 makes the wala/ML#365 worklist engine the only type-resolution
path (seeding-order independent, closes the wala/ML#718 DenseLayer3dProj
residual) and lands the origin refinements for #774: iteration products
classify through their producer (wala/ML#729), unresolved-iterable
enumerate no longer defaults to a TensorFlow origin (wala/ML#730/#728),
and the parameter origin seed/barrier is user-code-only (wala/ML#731).
Shape-element scalars broadcast correctly (wala/ML#723).

The suite is green with no pin advances.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01He1ULx8nRoHzGicMryry1A

* Advance the Ariadne bump to 0.52.31

0.52.31 lands wala/ML#732 (non-tensor slice-constructor and iterator
objects no longer count as tensor defs), the last piece of the #774
origin arc. With the #777 detector unchanged, all four `deep_recommenders`
`Cora` data-prep methods now decline the barren precondition, closing
#774's original four-method report end to end. Suite green, no pin
advances.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01He1ULx8nRoHzGicMryry1A

* Vendor the Cora data-prep subject as a #774 regression fixture

The #774 resolution (the four `deep_recommenders` `Cora` numpy/scipy
data-prep methods declining) was guarded only by the whole-project corpus
and synthetic distillations, not by a unit pin on the real subject.
`testCoraDataPrepBarren` vendors `datasets/cora.py`'s `Cora` class and a
numpy-feeding driver, asserting the end-to-end contract on 0.52.31:
`build_graph`, `encode_labels`, and the nested `_get_labels`/`_sample_mask`
report no tensor computation (barren), while the numpy-on-parameters
siblings `sample_train_nodes` and the outer `split_labels` still compute and
are declined by that precondition (code 16) instead.

Co-Authored-By: Claude ... (continued)

2139 of 2616 relevant lines covered (81.77%)

0.82 hits per line

Jobs
ID Job ID Ran Files Coverage
1 #2513.1 33
81.77
Source Files on build #2513