FastAPI uploaded image as np.float32 to a Deep Learning model (OpenCV)-CodePudding

I have a TensorFlow Keras Deep Learning model in the form of an h5 file. How can i pass a Fast API uploaded image as np.float32 to my model.

import numpy as np
import cv2
from fastapi import FastAPI, File, UploadFile
import numpy as np
from tensorflow.keras.models import load_model
import tensorflow as tf

model=load_model("complete_model.h5")
app = FastAPI()

def prepare(image):
    IMG_SIZE = 224
    new_array = cv2.resize(image, (IMG_SIZE, IMG_SIZE)) 
    return new_array.reshape(-1, IMG_SIZE,IMG_SIZE,3)

@app.post("/")
async def root(file: UploadFile = File(...)):
    global model
    content = await file.read()
    nparr = np.fromstring(content, np.uint8)
    img = cv2.imdecode(nparr, cv2.IMREAD_COLOR).astype(np.float32)
    prediction = model.predict(prepare(img))
    return prediction

When uploading image using Swagger UI i get an error

line 137, in jsonable_encoder
data = dict(obj)
TypeError: 'numpy.float32' object is not iterable

Normal code

import numpy as np
import numpy as np
from tensorflow.keras.models import load_model
import tensorflow as tf
import cv2

model=load_model("complete_model.h5")

def prepare(image):
    IMG_SIZE = 224
    new_array = cv2.resize(image, (IMG_SIZE, IMG_SIZE)) 
    return new_array.reshape(-1, IMG_SIZE,IMG_SIZE,3)

img = cv2.imread("./test.jpeg").astype(np.float32)
prediction = model.predict(prepare(img))
print(prediction)

Result in terminal:

[[0.25442022 0.74557984]]

How can I get the same result while using Fast API

CodePudding user response：

The error is thrown when returning the response (i.e., prediction in your case) from your endpoint. It looks like FastAPI is trying to convert the numpy array to a dict. Have a look a the discussion here as well. Thus, what you could do is to convert the numpy array to JSON:

return json.dumps(prediction.tolist())

On OpenAPI (Swagger UI), you will still be able to see the expected result. However, if you need to convert it back to a numpy array, you can parse the JSON string as shown below.

arr = np.array(json.loads(resp.json()))  # resp.json() if using Python requests