AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
翻譯準備中
尚未偵測到更新。GitHub 發布版本後會顯示在這裡。
個人 GitHub 帳號下公開儲存庫的擁有者可以登入並驗證所有權,自行管理介紹,無需管理員核准。
登入並產生驗證檔案,在 30 分鐘內將其提交至此儲存庫的預設分支。其他分支或分叉儲存庫中的檔案無效。然後在下方檢查檔案,即可無需管理員核准直接管理專案。請保留檔案以維持管理權限。
如需登入,將前往登入頁面。登入後請回到此處,再次點擊所選的驗證按鈕。
在專案資料夾中執行 /nomorevibe verify,即可驗證所有權並自行管理介紹。
一分鐘完成驗證正在載入完整 README。 以下為已儲存的摘錄。
A Python frontend to Ken Brakke's Surface Evolver
(version 2.70a). Evolver runs in-process as a compiled extension module built with
nanobind; the mesh comes back as NumPy arrays.
Documentation:, with an API
reference, a tutorial (a double bubble, from arrays to a FEM-ready mesh), worked
examples and performance numbers; built from docs/ with Sphinx (see Development).檢視完整 README ↗Soap films, bubbles, drops on walls, liquid bridges and solder joints from Python. pySurfaceEvolver runs Ken Brakke's Surface Evolver, the standard program for surfaces of least energy, inside Python: surfaces go in and come back as NumPy arrays.
relax() takes a surface to equilibrium and says how it went: converged,
stable, mesh health. It looks after the mesh and switches to Newton steps on its
own.git clone https://github.com/gautelinga/pySurfaceEvolver
cd pySurfaceEvolver
pip install ".[all]" # with PyVista for plots and meshio for mesh files
This builds the engine from source with a C and C++ compiler (and, for the fast Newton solver, gfortran with LAPACK/BLAS); builds are tested on Linux. See Install for the details.
import pysurfaceevolver as pyse
from pysurfaceevolver import constraints as C, shapes
drop = shapes.cap(radius=1, height=0.6) # a rough start: relaxing meets the volume
ev = pyse.Evolver()
ev.load_arrays(drop.vertices, drop.faces,
constraints={1: C.plane((0, 0, 1), 0.0, contact_angle=60)},
vertex_constraints={1: drop.rims["rim"]},
bodies=[pyse.Body(volume=1.0)])
result = ev.relax(levels=2) # refine twice, relax to equilibrium
print(result)
print(f"pressure {ev.body(1).pressure:.4f}, contact angles {ev.contact_angles(1).mean():.1f} degrees")
ev.plot(walls=True) # the drop, with the table drawn
IterationResult(converged after 291 gradient steps and 3 Newton, residual 7.5e-13, stable, health ok)
pressure 1.7366, contact angles 59.2 degrees
https://gautelinga.github.io/pySurfaceEvolver/
relax() does and where its defaults stop working.Bug reports and questions are welcome as issues. Building, testing and the layout of the code are described in Development.
pySurfaceEvolver's own code is MIT-licensed (LICENSE). Surface Evolver itself (most
of src/) is by Kenneth A. Brakke and freely available under his terms; the software
bundled in the wheels keeps its own licenses (THIRD_PARTY_LICENSES.txt).