train.py

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
from tensorflow.keras.applications import MobileNetV2
import pathlib, os

# Dataset yolu - ZIP'i çıkarttığın klasörü buraya yaz
DATA_DIR = pathlib.Path(os.path.expanduser("~/Downloads/archive/Garbage classification/Garbage classification"))

IMG_SIZE = (224, 224)
BATCH_SIZE = 32
EPOCHS = 10

# Veriyi yükle
train_ds = keras.utils.image_dataset_from_directory(
    DATA_DIR,
    validation_split=0.2,
    subset="training",
    seed=123,
    image_size=IMG_SIZE,
    batch_size=BATCH_SIZE
)

val_ds = keras.utils.image_dataset_from_directory(
    DATA_DIR,
    validation_split=0.2,
    subset="validation",
    seed=123,
    image_size=IMG_SIZE,
    batch_size=BATCH_SIZE
)

class_names = train_ds.class_names
print("Kategoriler:", class_names)

# Performans için cache
AUTOTUNE = tf.data.AUTOTUNE
train_ds = train_ds.cache().shuffle(1000).prefetch(buffer_size=AUTOTUNE)
val_ds = val_ds.cache().prefetch(buffer_size=AUTOTUNE)

# MobileNetV2 - Transfer Learning
base_model = MobileNetV2(input_shape=(224, 224, 3), include_top=False, weights='imagenet')
base_model.trainable = False

model = keras.Sequential([
    layers.Rescaling(1./127.5, offset=-1),
    base_model,
    layers.GlobalAveragePooling2D(),
    layers.Dropout(0.3),
    layers.Dense(6, activation='softmax')
])

model.compile(
    optimizer='adam',
    loss='sparse_categorical_crossentropy',
    metrics=['accuracy']
)

model.summary()

# Eğitim
history = model.fit(train_ds, validation_data=val_ds, epochs=EPOCHS)

# Kaydet
model.save("garbage_model.h5")
print("\nModel kaydedildi: garbage_model.h5")
print("Son doğruluk:", round(history.history['val_accuracy'][-1]*100, 2), "%")