app.py

from fastapi import FastAPI, File, UploadFile
from tensorflow.keras.applications.mobilenet_v2 import MobileNetV2, preprocess_input, decode_predictions
from tensorflow.keras.preprocessing import image
import numpy as np
from io import BytesIO
from PIL import Image
from fastapi import FastAPI
from fastapi.responses import HTMLResponse
import os


app = FastAPI()


# Desativa logs menos importantes do TensorFlow
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
# Desativa as otimizações do oneDNN
os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0'

@app.get("/logs=container")
async def container_logs():
    return {"status": "No logs available"}




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


app = FastAPI()

# Carregar o modelo MobileNetV2
model = MobileNetV2(weights="imagenet")

def prepare_image(img):
    img = img.resize((224, 224))
    img_array = image.img_to_array(img)
    img_array = np.expand_dims(img_array, axis=0)
    return preprocess_input(img_array)

@app.post("/predict")
async def predict(file: UploadFile = File(...)):
    contents = await file.read()
    img = Image.open(BytesIO(contents)).convert("RGB")
    processed_image = prepare_image(img)
    
    predictions = model.predict(processed_image)
    results = decode_predictions(predictions, top=3)[0]
    return [{"label": label, "probability": float(prob)} for (_, label, prob) in results]