3 API BUAH FRESHLY

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+freshly_saved_model_banana/variables/variables.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+freshly_saved_model_mango/variables/variables.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
+freshly_saved_model_orange/variables/variables.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -0,0 +1,22 @@
+# Menggunakan image Python 3.9
+FROM python:3.9
+
+# Mengatur direktori kerja
+WORKDIR /code
+
+# Menyalin requirements dan menginstalnya
+COPY ./requirements.txt /code/requirements.txt
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+# Hugging Face Spaces menyarankan untuk membuat user non-root
+RUN useradd -m -u 1000 user
+USER user
+ENV HOME=/home/user \
+    PATH=/home/user/.local/bin:$PATH
+
+# Pindah ke direktori home user
+WORKDIR $HOME/app
+COPY --chown=user . $HOME/app
+
+# Hugging Face Spaces MEWAJIBKAN aplikasi berjalan di port 7860
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

freshly_saved_model_banana/keras_metadata.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be4d385a196ee895a8f75e57d26b7c753d785efd1f562390af025ead2425bd1e
+size 48687

freshly_saved_model_banana/saved_model.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a98ab46180a5ff2bec82b47914468f1efcd8a1c760b4037c5c98bf60e52dc93
+size 477488

freshly_saved_model_banana/variables/variables.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7a878cee0fe1e50768d947816566eebec30c02b254ac28312612c77972e8232
+size 6147460

freshly_saved_model_mango/keras_metadata.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a8b482b679df62df5a07acbc3c0760a61a9e03a93dd2ac3e5f023db2235a217
+size 48597

freshly_saved_model_mango/saved_model.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e2cf97675a1f9ad633f3976de18b2fc355abdb19379a9d9bb43ff1aca39bca9
+size 478335

freshly_saved_model_mango/variables/variables.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bbccdfe41a861c5f7ef7cde942bfc9c14901cba82a09ff1e8d89908d655444d
+size 6147420

freshly_saved_model_orange/keras_metadata.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b57fe9ad6f00eb7f3afd29b0dce8a9fa50416a656c7aedd8dcc72f3fd3e2d08
+size 48600

freshly_saved_model_orange/saved_model.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:021a3bb64a6559007abbc0ddbaf466b6b634a6bfca787822532b8d9775bd2cd7
+size 479457

freshly_saved_model_orange/variables/variables.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e785bb1c07cf7b2b637cc3de1d7315b46f6c2934286900b3564eb72267e53661
+size 6147420

main.py

@@ -0,0 +1,92 @@
+import io
+import numpy as np
+import tensorflow as tf
+from fastapi import FastAPI, File, UploadFile, HTTPException
+from PIL import Image
+
+app = FastAPI(
+    title="Freshly Universal API",
+    description="API untuk mendeteksi kematangan berbagai jenis buah dan sayur",
+    version="2.0.0"
+)
+
+# 1. Konfigurasi Model & Kelasnya
+# Pastikan nama kelas sesuai dengan urutan saat Anda melatih masing-masing model
+CONFIG = {
+    "banana":  ['banana_ripe', 'banana_rotten', 'banana_unripe'],
+    "mango":   ['mango_ripe', 'mango_rotten', 'mango_unripe'],
+    "orange":  ['orange_ripe', 'orange_rotten', 'orange_unripe'],
+    #"chili":   ['chili_ripe', 'chili_rotten', 'chili_unripe'],
+    #"paprika": ['paprika_ripe', 'paprika_rotten', 'paprika_unripe'],
+    #"tomato":  ['tomato_ripe', 'tomato_rotten', 'tomato_unripe']
+}
+
+IMG_SIZE = (224, 224)
+MEAN = np.array([0.485, 0.456, 0.406])
+STD = np.array([0.229, 0.224, 0.225])
+
+# 2. Dictionary untuk menyimpan semua model di RAM
+models = {}
+
+# Memuat semua model secara otomatis saat server menyala
+print("Memulai proses pemuatan semua model...")
+for fruit_type in CONFIG.keys():
+    model_folder = f"freshly_saved_model_{fruit_type}"
+    try:
+        print(f" -> Memuat model {fruit_type.upper()} dari {model_folder}...")
+        models[fruit_type] = tf.keras.models.load_model(model_folder)
+    except Exception as e:
+        print(f" [X] Gagal memuat model {fruit_type}: Pastikan folder {model_folder} ada!")
+
+print("\nSemua model yang tersedia siap digunakan!")
+
+# 3. Fungsi Preprocessing (Bisa dipakai untuk semua model karena format inputnya sama)
+def preprocess_image(image_bytes):
+    try:
+        img = Image.open(io.BytesIO(image_bytes)).convert('RGB')
+        img = img.resize(IMG_SIZE)
+        img_array = tf.keras.utils.img_to_array(img)
+        img_array = img_array / 255.0
+        img_array = (img_array - MEAN) / STD
+        img_array = np.expand_dims(img_array, axis=0)
+        return img_array
+    except Exception as e:
+        raise ValueError(f"Gagal memproses gambar: {str(e)}")
+
+# 4. Endpoint Dinamis (Menerima parameter jenis buah di URL)
+@app.post("/predict/{fruit_type}")
+async def predict_fruit(fruit_type: str, file: UploadFile = File(...)):
+    # Cek apakah jenis buah yang diminta (misal: "apple") ada di daftar kita
+    fruit_type = fruit_type.lower()
+    if fruit_type not in models:
+        raise HTTPException(
+            status_code=404, 
+            detail=f"Model untuk '{fruit_type}' tidak ditemukan. Tersedia: {list(models.keys())}"
+        )
+    
+    if not file.content_type.startswith('image/'):
+        raise HTTPException(status_code=400, detail="File harus berupa gambar.")
+    
+    try:
+        contents = await file.read()
+        img_tensor = preprocess_image(contents)
+        
+        # Prediksi menggunakan model yang Sesuai dengan URL
+        active_model = models[fruit_type]
+        class_names = CONFIG[fruit_type]
+        
+        predictions = active_model.predict(img_tensor)
+        pred_index = np.argmax(predictions[0])
+        confidence = float(predictions[0][pred_index])
+        
+        return {
+            "fruit_type": fruit_type,
+            "filename": file.filename,
+            "predicted_class": class_names[pred_index],
+            "confidence": round(confidence * 100, 2),
+            "all_probabilities": {
+                class_names[i]: round(float(predictions[0][i]) * 100, 2) for i in range(len(class_names))
+            }
+        }
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))

requirements.txt

@@ -0,0 +1,6 @@
+fastapi==0.104.1
+uvicorn==0.24.0
+tensorflow-cpu==2.10.0
+Pillow==10.1.0
+python-multipart==0.0.6
+numpy==1.26.2