ML Term Project 2023: A Grad Course Retrospective

ML_Termproject2023

ML_Termproject2023 is a final project for a graduate-level machine learning course. It was completed over 7 commits between November and December 2023.

Project Structure

flowchart TD
    A[문제 정의] --> B[데이터 준비]
    B --> C[Baseline 모델]
    C --> D[모델 개선]
    D --> E[하이퍼파라미터 튜닝]
    E --> F[결과 분석]
    F --> G[보고서 작성]

Model Comparison Experiments

from sklearn.model_selection import cross_val_score
from sklearn.metrics import accuracy_score

models = {
    'Logistic Regression': LogisticRegression(),
    'Random Forest': RandomForestClassifier(n_estimators=100),
    'SVM': SVC(kernel='rbf'),
    'XGBoost': XGBClassifier(),
    'Neural Network': MLPClassifier(hidden_layer_sizes=(100, 50)),
}

results = {}
for name, model in models.items():
    scores = cross_val_score(model, X_train, y_train, cv=5)
    results[name] = {
        'mean': scores.mean(),
        'std': scores.std()
    }

Hyperparameter Tuning

from sklearn.model_selection import GridSearchCV

param_grid = {
    'n_estimators': [100, 200, 500],
    'max_depth': [5, 10, 15, None],
    'min_samples_split': [2, 5, 10],
}

grid_search = GridSearchCV(
    RandomForestClassifier(),
    param_grid,
    cv=5,
    scoring='accuracy',
    n_jobs=-1,
    verbose=1
)
grid_search.fit(X_train, y_train)

Lessons Learned

1. Baselines matter: Good preprocessing beats a more complex model
2. Cross-validation is non-negotiable: A single train/test split risks overfitting
3. Feature importance analysis: Model interpretability matters in research

This was an undergraduate course project, but it laid a solid foundation in the fundamentals of ML experimentation.