Active learning for materials test campaigns in the Uncertainty Engine.
Open the how-to
— docs/active-learning.html rendered straight from
this repository, no checkout needed.
It is a short guide to the Engine's own canvas: where to start (two tests at opposite ends of the range), how to import the workflows, how to run the loop, and how to tell when the campaign is finished. Two further tabs cover getting around the Engine and the words used throughout.
The workflows it refers to are in workflows/, exported from the
canvas, plus example.csv to start from. They are the same graphs the notebooks
build, so the Studio and the SDK run the same thing.
Each test type has a folder under material_tests/, holding
its notebook, any helpers, and a README covering the model and the method.
| Material test | Notebook | The question | Inputs | |
|---|---|---|---|---|
| DBTT | DBTT_active_learning.ipynb |
Where is the ductile–brittle transition temperature? | temperature | |
GP_learnable_uncertainty.ipynb (helper) |
How much of this model's uncertainty could more testing remove? | a dataset already in the project | ||
DBTT_design_study.ipynb (helper) |
Which acquisition, and which kernel? | temperature | ||
| Creep | Creep_active_learning.ipynb |
What stress gives 100,000 h life at the design temperature? | stress and temperature | |
| Fatigue | Fatigue_active_learning.ipynb |
What amplitude survives 10⁷ cycles? | stress amplitude |
Every notebook is self-contained: it imports nothing from this repository, so the Colab badges work without a checkout. Everything meant to be changed is gathered in a single Settings cell at the top.
Google Colab. Click a badge. The repository is public, so no GitHub
authorisation is needed. Put your credentials in the Secrets panel (the 🔑 key
icon in the left sidebar) as UE_USERNAME, UE_PASSWORD and
UE_PROJECT_NAME, with Notebook access switched on for each. The first cell
installs whatever the runtime is missing.
VS Code, or any local Jupyter.
python -m venv .venv && .venv/bin/pip install -r requirements.txt
cp .env.example .env # then fill it inPick the .venv kernel in VS Code's top-right kernel picker. .env is
gitignored and must never be committed; copy the example rather than editing it,
so the template keeps its keys blank.
UE_PROJECT_NAME must name a project that already exists — projects are
created in the Engine GUI, not from the SDK. load_credentials() tries Colab
secrets first, then .env, then the environment, and reports only where the
credentials came from, never what they are, so the notebooks' outputs are safe
to commit.
All three notebooks write their resources and workflows into the same project
under distinct names (dbtt_*, creep_*, fatigue_*), so they do not collide
and can be run in any order.