app.py

from keras import (
    saving,
    preprocessing,
    applications
)
import fastapi
from fastapi import UploadFile, File, HTTPException
from PIL import Image
import io
import time
import numpy as np


app = fastapi.FastAPI()

model = saving.load_model("hf://Yumeng-Liu/trash-classifier")

CLASSES = ['biological', 'cardboard', 'glass', 'metal', 'paper', 'plastic', 'trash']
THRESHOLD = 5.0e-1


def get_prediction(img: Image) -> str:
    img = img.resize((224, 224))
    img_array = preprocessing.image.img_to_array(img)
    img_array = np.expand_dims(img_array, axis=0)  # Add an extra dimension to match the model's input shape
    img_array = applications.mobilenet_v2.preprocess_input(img_array)

    prediction_array = model.predict(img_array)
    predicted_class_idx = np.argmax(prediction_array[0])

    if prediction_array[0][predicted_class_idx] > THRESHOLD:
        prediction = CLASSES[predicted_class_idx]
    else:
        prediction = "none"

    return prediction


@app.get("/")
def read_root():
    return {"Hello": "World"}


@app.post("/predict-image")
async def predict(received_image: UploadFile = File(...)):
    try:
        contents = received_image.file.read()
        # Open the binary data as an image
        image = Image.open(io.BytesIO(contents))

        print("Image received")
        # You can now work with the `image` object
        print(image.format, image.size, image.mode)  # Example: JPEG (1920, 1080) RGB
        print("")

        prediction_result = get_prediction(image)
        print(prediction_result)

        # Perform further processing, e.g., save it, analyze it, etc.
        return {
            "result": prediction_result
        }

    except Exception as e:
        print(e)
        raise HTTPException(status_code=500, detail='Something went wrong')

    finally:
        received_image.file.close()


if __name__ == "__main__":
    print("Starting app")
    while True:
        time.sleep(10)