Python-Flask server for analyse image with one hand written digit
To analyse images, I use tensorflow
Server -> Python 3.8 with Flask 2.2.3
Model -> Tensorflow-macos 2.9
I use the MNIST dataset
The model is train until it reached 0.99 accuracy
I use the adam algorithm for the optimizer and sparse_categorical_crossentropy for the loss
You can install dependencies with the following command :
pip install -r requirements{.linux | .macos}.txt
and run the server :
python server/app.py
Or you can use docker :
docker build -t image-name
docker run -p PORT:PORT image-name
/analyseImage
: take one image in the body and return the result : {"analyse": "result"}