Spaces:
Running
Running
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from fastapi import FastAPI, UploadFile, File
|
2 |
+
from fastapi.middleware.cors import CORSMiddleware
|
3 |
+
import numpy as np
|
4 |
+
from tensorflow.keras.models import load_model
|
5 |
+
from tensorflow.keras.preprocessing.image import load_img, img_to_array
|
6 |
+
from io import BytesIO
|
7 |
+
|
8 |
+
app = FastAPI()
|
9 |
+
|
10 |
+
# Enable CORS to allow requests from frontend (React)
|
11 |
+
app.add_middleware(
|
12 |
+
CORSMiddleware,
|
13 |
+
allow_origins=["*"], # Change ["http://localhost:5173"] for better security
|
14 |
+
allow_credentials=True,
|
15 |
+
allow_methods=["*"],
|
16 |
+
allow_headers=["*"],
|
17 |
+
)
|
18 |
+
|
19 |
+
# Load your model
|
20 |
+
model = load_model("densenet201_food_classification.h5")
|
21 |
+
|
22 |
+
# Define class indices
|
23 |
+
class_indices = {
|
24 |
+
0: "burger",
|
25 |
+
1: "butter_naan",
|
26 |
+
2: "chai",
|
27 |
+
3: "chapati",
|
28 |
+
4: "chole_bhature",
|
29 |
+
5: "dal_makhani",
|
30 |
+
6: "dhokla",
|
31 |
+
7: "fried_rice",
|
32 |
+
8: "idli",
|
33 |
+
9: "jalebi",
|
34 |
+
10: "kaathi_rolls",
|
35 |
+
11: "kadai_paneer",
|
36 |
+
12: "kulfi",
|
37 |
+
13: "masala_dosa",
|
38 |
+
14: "momos",
|
39 |
+
15: "paani_puri",
|
40 |
+
16: "pakode",
|
41 |
+
17: "pav_bhaji",
|
42 |
+
18: "pizza",
|
43 |
+
19: "samosa"
|
44 |
+
}
|
45 |
+
|
46 |
+
def predict_image(image, model):
|
47 |
+
try:
|
48 |
+
img = load_img(image, target_size=(224, 224))
|
49 |
+
image_array = img_to_array(img) / 255.0
|
50 |
+
image_array = np.expand_dims(image_array, axis=0)
|
51 |
+
|
52 |
+
predictions = model.predict(image_array)
|
53 |
+
class_idx = np.argmax(predictions)
|
54 |
+
class_label = class_indices.get(class_idx, "Unknown")
|
55 |
+
confidence = float(predictions[0][class_idx])
|
56 |
+
|
57 |
+
return class_label, confidence
|
58 |
+
except Exception as e:
|
59 |
+
return None, None
|
60 |
+
|
61 |
+
@app.post("/predict/")
|
62 |
+
async def predict(file: UploadFile = File(...)):
|
63 |
+
try:
|
64 |
+
image_data = await file.read()
|
65 |
+
image = BytesIO(image_data)
|
66 |
+
|
67 |
+
class_label, confidence = predict_image(image, model)
|
68 |
+
|
69 |
+
if class_label is None:
|
70 |
+
return {"error": "Prediction failed"}
|
71 |
+
|
72 |
+
return {"predicted_class": class_label, "confidence": f"{confidence:.2f}"}
|
73 |
+
except Exception as e:
|
74 |
+
return {"error": f"Internal Server Error: {str(e)}"}
|