refactor: restrukturisasi folder eksperimen ke tasks modular

tasks/mlp/03_model_selection.py

@@ -0,0 +1,67 @@
+"""
+Tugas 3: Contoh Implementasi Berbagai Arsitektur Model Sesuai Karakteristik Data
+"""
+
+import tensorflow as tf
+from tensorflow import keras
+from tensorflow.keras.layers import Conv2D, Dense, Embedding, Flatten, LSTM, MaxPooling2D
+from tensorflow.keras.models import Sequential
+from xgboost import XGBClassifier
+
+
+def build_models():
+    print("=" * 60)
+    print("TUGAS 3: CONTOH ARSITEKTUR MODEL SELECTION")
+    print("=" * 60)
+
+    # (a) Prediksi Stok Barang dari Deret Waktu (LSTM)
+    # Input shape: (samples, time_steps=12, features=1)
+    model_a = Sequential([
+        LSTM(32, input_shape=(12, 1)),
+        Dense(1)
+    ], name="LSTM_Stock_Forecasting")
+    model_a.compile(optimizer='adam', loss='mse')
+    print("\n[Case A] Model LSTM untuk Prediksi Stok Barang (Time Series):")
+    model_a.summary()
+
+    # (b) Klasifikasi Foto Produk Rusak vs Normal (CNN)
+    # Input shape contoh citra RGB 128x128
+    model_b = Sequential([
+        Conv2D(32, (3, 3), activation='relu', input_shape=(128, 128, 3)),
+        MaxPooling2D((2, 2)),
+        Flatten(),
+        Dense(64, activation='relu'),
+        Dense(1, activation='sigmoid')  # binary: rusak vs normal
+    ], name="CNN_Defect_Classification")
+    model_b.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
+    print("\n[Case B] Model CNN untuk Klasifikasi Foto Produk (Computer Vision):")
+    model_b.summary()
+
+    # (c) Analisis Sentimen Ulasan Play Store (LSTM + Embedding)
+    vocab_size = 5000
+    max_length = 100
+    model_c = Sequential([
+        Embedding(input_dim=vocab_size, output_dim=64, input_length=max_length),
+        LSTM(64),
+        Dense(32, activation='relu'),
+        Dense(1, activation='sigmoid')  # binary: positif vs negatif
+    ], name="LSTM_Sentiment_Analysis")
+    model_c.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
+    print("\n[Case C] Model LSTM + Embedding untuk Analisis Sentimen (NLP):")
+    model_c.summary()
+
+    # (d) Prediksi Churn Pelanggan Tabular (ML Klasik - XGBoost)
+    # X_train memiliki dimensi (n_samples, 15)
+    model_d = XGBClassifier(
+        n_estimators=100,
+        max_depth=4,
+        learning_rate=0.1,
+        random_state=42
+    )
+    print("\n[Case D] Model XGBoost untuk Prediksi Churn (Tabular 15 fitur):")
+    print(model_d)
+
+
+if __name__ == "__main__":
+    build_models()
+