from fastapi import FastAPI, UploadFile, File, HTTPException from fastapi.middleware.cors import CORSMiddleware import uvicorn import numpy as np from PIL import Image import io import os try: import tensorflow as tf from tensorflow.keras.models import load_model from tensorflow.keras.preprocessing.image import img_to_array TF_AVAILABLE = True print("TensorFlow berhasil diimport") print("TensorFlow Version:", tf.__version__) except Exception as e: TF_AVAILABLE = False print("TensorFlow Import Error:", e) app = FastAPI( title="Rock Detection API", version="1.0" ) app.add_middleware( CORSMiddleware, allow_origins=["*"], # sementara untuk testing allow_credentials=True, allow_methods=["*"], allow_headers=["*"], ) MODEL_PATH = "model_cnn.h5" # 7 CLASS SESUAI MODEL CNN CLASS_NAMES = [ "Quartzite", "Marble", "Coal", "Sandstone", "Limestone", "Basalt", "Granite" ] # KATEGORI BATU ROCK_INFO = { "Quartzite": "Metamorphic", "Marble": "Metamorphic", "Coal": "Sedimentary", "Sandstone": "Sedimentary", "Limestone": "Sedimentary", "Basalt": "Igneous", "Granite": "Igneous" } TARGET_SIZE = (224, 224) model = None @app.get("/") def root(): return { "message": "Rock Detection API Aktif" } @app.on_event("startup") async def load_ml_model(): global model if not TF_AVAILABLE: print("WARNING: TensorFlow tidak tersedia") return if not os.path.exists(MODEL_PATH): print(f"ERROR: File {MODEL_PATH} tidak ditemukan") return try: model = load_model(MODEL_PATH) print(f"\nModel {MODEL_PATH} berhasil dimuat!") print("Input Shape :", model.input_shape) print("Output Shape:", model.output_shape) model.summary() except Exception as e: print("ERROR LOAD MODEL:", e) def preprocess_image(image_bytes: bytes): try: img = Image.open(io.BytesIO(image_bytes)).convert("RGB") img = img.resize(TARGET_SIZE) img_array = img_to_array(img) img_array = img_array / 255.0 img_array = np.expand_dims(img_array, axis=0) return img_array except Exception as e: raise ValueError(f"Error preprocessing image: {e}") @app.post("/predict") async def predict(file: UploadFile = File(...)): if not file.content_type.startswith("image/"): raise HTTPException( status_code=400, detail="File harus berupa gambar" ) if model is None: raise HTTPException( status_code=500, detail="Model CNN belum dimuat" ) try: contents = await file.read() processed_image = preprocess_image(contents) print("\nProcessed Image Shape:") print(processed_image.shape) predictions = model.predict(processed_image) print("\nPredictions:") print(predictions) print("\nPrediction Shape:") print(predictions.shape) class_idx = np.argmax(predictions[0]) confidence = float(predictions[0][class_idx]) predicted_class = CLASS_NAMES[class_idx] rock_category = ROCK_INFO[predicted_class] return { "rock_type": predicted_class, "rock_category": rock_category, "confidence": round(confidence, 4), "message": "Prediksi berhasil" } except Exception as e: print("PREDICT ERROR:", e) raise HTTPException( status_code=500, detail=str(e) ) if __name__ == "__main__": uvicorn.run( "main:app", host="0.0.0.0", port=7860, reload=True )