Create file train_model.py

train_model.py

@@ -0,0 +1,55 @@
+import tensorflow as tf
+import numpy as np
+from datasets import load_dataset
+from sklearn.preprocessing import LabelBinarizer
+from tensorflow.keras.applications import MobileNetV2
+from tensorflow.keras.layers import Dense, GlobalAveragePooling2D
+from tensorflow.keras.models import Model
+from tensorflow.keras.optimizers import Adam
+from tqdm import tqdm
+
+# Load dataset dari Hugging Face
+dataset = load_dataset("tanganke/stanford_cars")
+NUM_CLASSES = 196
+IMG_SIZE = 224
+BATCH_SIZE = 32
+EPOCHS = 5
+
+# Preprocessing fungsi
+def preprocess(example):
+    image = example["image"].resize((IMG_SIZE, IMG_SIZE))
+    image = np.array(image) / 255.0
+    return image, example["label"]
+
+# Apply preprocessing
+X_train, y_train = zip(*[preprocess(x) for x in tqdm(dataset["train"])])
+X_test, y_test = zip(*[preprocess(x) for x in tqdm(dataset["test"])])
+
+X_train = np.array(X_train)
+X_test = np.array(X_test)
+
+# One-hot encoding label
+lb = LabelBinarizer()
+y_train = lb.fit_transform(y_train)
+y_test = lb.transform(y_test)
+
+# Load base model MobileNetV2
+base_model = MobileNetV2(include_top=False, input_shape=(IMG_SIZE, IMG_SIZE, 3), weights='imagenet')
+x = GlobalAveragePooling2D()(base_model.output)
+x = Dense(256, activation='relu')(x)
+preds = Dense(NUM_CLASSES, activation='softmax')(x)
+
+model = Model(inputs=base_model.input, outputs=preds)
+
+# Freeze base layers
+for layer in base_model.layers:
+    layer.trainable = False
+
+model.compile(optimizer=Adam(1e-4), loss='categorical_crossentropy', metrics=['accuracy'])
+
+# Training
+model.fit(X_train, y_train, validation_data=(X_test, y_test), epochs=EPOCHS, batch_size=BATCH_SIZE)
+
+# Simpan model
+model.save("model/car_model.h5")
+print("✅ Model saved to model/car_model.h5")