Machine Learning in Engineering Science | NCKU + LAiMM | Fall 2026
Purpose: a fully worked, self-paced in-class practice with an optional extension. Weeks 11 and 18 are optional examination-readiness materials. These notebooks do not create weekly graded assignments.
Before running: choose a Python 3 kernel in Jupyter or a CPU runtime in Colab. Run every cell from top to bottom with Shift+Enter. Setup installs missing free packages. No API key or external dataset download is required.
Learning target: complete the calculation, inspect intermediate values, and explain why the protocol supports a limited claim.
Time: about 25-45 minutes including reading, prediction, execution, and explain-back.
Keep the seed fixed for the first run. Record actual outputs and package versions rather than copying a claimed performance number.
import sys, subprocess, importlib.util, platform, json, random, hashlib
from pathlib import Path
required = {"numpy": "numpy", "matplotlib": "matplotlib", "sklearn": "scikit-learn"}
missing = [package for module, package in required.items() if importlib.util.find_spec(module) is None]
if missing:
subprocess.check_call([sys.executable, "-m", "pip", "install", *missing])
import numpy as np
import matplotlib.pyplot as plt
SEED = 2026
random.seed(SEED)
np.random.seed(SEED)
rng = np.random.default_rng(SEED)
ARTIFACTS = Path("artifacts")
ARTIFACTS.mkdir(exist_ok=True)
print("Python:", platform.python_version(), "NumPy:", np.__version__, "Seed:", SEED)
Checkpoint. The environment cell must finish before any analysis cell. Restart and run all if a variable is missing.
Create a small original artifact and an explicit claim. A result card should distinguish measured behavior from broader ambitions.
signal=rng.normal(size=100)
card={"task":"mean of a generated normal sample","seed":SEED,"n":100,
"observed_mean":float(signal.mean()),"units":"dimensionless synthetic",
"limitation":"one random sample; no real engineering measurement"}
Checkpoint. The sample mean will not generally equal zero exactly; zero is the population expectation.
Save the array with a hash based on its bytes and reload it. This checks artifact integrity, not scientific validity.
path=ARTIFACTS/"final-synthetic.npy"; np.save(path,signal)
digest=hashlib.sha256(path.read_bytes()).hexdigest()
restored=np.load(path)
assert np.array_equal(signal,restored)
card["file_sha256"]=digest
print(card)
Checkpoint. A classmate needs both the artifact and its interpretation. An unlabeled array is insufficient.
Write claim/evidence/limitation answers and identify what would change your conclusion. This is optional rehearsal.
questions=["What exactly was measured?","Can another person rerun it?","What would falsify the claim?"]
summary={"result_card":card,"oral_questions":questions,"AI_use":"Disclose any AI assistance and the checks you performed."}
for question in questions: print(question)
Checkpoint. Replace the toy card with your project evidence before the final examination.
Export a JSON record with a seed, package version, and measured result. Download the artifacts directory before leaving Colab; runtime storage is temporary.
record={"week":17,"seed":SEED,"numpy":np.__version__,"python":platform.python_version(),"result":summary}
(ARTIFACTS/"week-17-evidence.json").write_text(json.dumps(record,indent=2))
print(json.dumps(record,indent=2))
Checkpoint. The evidence file must match the result you actually obtained. A saved screenshot alone does not permit rerunning the calculation.
1. Predict the effect of one change before editing code. Change one parameter, input, or boundary condition only.
2. Rerun all cells in a fresh kernel. Keep the original evidence record and compare the observed difference.
3. Explain the result using the mechanism, not just the final number.
4. Name one assumption that limits the claim, and a new test that could challenge it.
Troubleshooting: if installation fails, restore network access; if a variable is undefined, run earlier cells; if your output differs, compare package versions, seed, and data construction. Do not repeatedly edit code until a preferred score appears.
Completion criterion: one executed notebook, one evidence JSON, and a brief claim/evidence/limitation statement.