ponder-lab / Hybridize-Functions-Refactoring / #2626
83%
main: 84%

LAST BUILD BRANCH: gh-readonly-queue/main/pr-842-77b7f38fa055d4987502c13054242de54cacb165
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: 83.129%. Remained the same
#2626

push

github

web-flow 
Make the dynamic-shape numpy fixture run to completion (#820)

* Make the dynamic-shape numpy fixture run to completion

The module-level driver called `head_over_dynamic` on the symbolic Keras input and crashed: `get_shape(inp)` is `[None, 5]`, so `np.prod` feeds `None` into the reshape target, which is exactly the hazard the fixture models. The driver now witnesses the crash (`try`/`except` plus an assertion that it happened) instead of exiting nonzero on it, keeping the call site and its argument typing intact for the analysis. The issue's originally reported dtype mismatch belongs to an earlier form of the driver that the wala/ML#722 rework already replaced; the run-to-completion violation is what remained.

Full suite green, pinning that the classification evidence survives the guarded call.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GcqABsog1rtb4m2CDZdcRD

* Narrow the crash witness to TypeError from review

A bare `except Exception` could mask unrelated breakage (an import error, a framework API change) while still passing the assertion. The dynamic-dimension reshape raises `TypeError` (runtime-verified), so only that is witnessed; anything else propagates and fails the run.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GcqABsog1rtb4m2CDZdcRD

---------

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

2444 of 2940 relevant lines covered (83.13%)

0.83 hits per line

Jobs
ID Job ID Ran Files Coverage
1 #2626.1 33
83.13
Source Files on build #2626