Module 1: AI Foundations

# The practical model selection flowchart in code

def select_model(problem_type, data_type, dataset_size):
    """Simple model selection heuristic."""

    # Rule 1: Tabular/structured data
    if data_type == "tabular":
        if dataset_size < 1000:
            return "LogisticRegression / RandomForest"
        else:
            return "XGBoost / LightGBM"

    # Rule 2: Text data
    if data_type == "text":
        if dataset_size < 100:
            return "Few-shot with LLM (GPT-4, Claude)"
        elif dataset_size < 10000:
            return "Fine-tuned BERT / Sentence Transformers"
        else:
            return "Fine-tuned LLM or custom Transformer"

    # Rule 3: Image data
    if data_type == "image":
        if dataset_size < 500:
            return "Pre-trained ViT / ResNet with transfer learning"
        else:
            return "Fine-tuned vision model"

    # Rule 4: Time series
    if data_type == "time_series":
        if problem_type == "forecasting":
            return "Prophet / XGBoost with lag features"
        else:
            return "Temporal Fusion Transformer"

    return "Start with a baseline, iterate"

# Example usage
print(select_model("classification", "tabular", 50000))
# Output: "XGBoost / LightGBM"

print(select_model("generation", "text", 50))
# Output: "Few-shot with LLM (GPT-4, Claude)"

import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import classification_report

# Simulate a dataset
np.random.seed(42)
X = np.random.randn(1000, 10)  # 1000 samples, 10 features
y = (X[:, 0] + X[:, 1] * 2 > 0).astype(int)  # Binary target

# Step 1: Split your data (ALWAYS do this first)
X_train, X_temp, y_train, y_temp = train_test_split(
    X, y, test_size=0.3, random_state=42
)
X_val, X_test, y_val, y_test = train_test_split(
    X_temp, y_temp, test_size=0.5, random_state=42
)

print(f"Train: {len(X_train)}, Val: {len(X_val)}, Test: {len(X_test)}")
# Train: 700, Val: 150, Test: 150

# Step 2: Train and evaluate
model = RandomForestClassifier(n_estimators=100, random_state=42)
model.fit(X_train, y_train)

# Step 3: Check for overfitting
train_acc = model.score(X_train, y_train)
val_acc = model.score(X_val, y_val)
print(f"Train accuracy: {train_acc:.3f}")
print(f"Val accuracy:   {val_acc:.3f}")
# If train >> val, you're overfitting!

# Step 4: Final evaluation (only do this once!)
print("\nFinal Test Results:")
print(classification_report(y_test, model.predict(X_test)))