updated dependencies and comments

README.md

@@ -25,7 +25,7 @@ The code has been tested using:
 - [Flask] (3.0): a microframework for [Python] based on Werkzeug, Jinja 2 and good intentions.
 - [Gunicorn] (22.0): a [Python] [WSGI] HTTP Server for UNIX.
 - [NGINX] (1.25): a free, open-source, high-performance HTTP server, reverse proxy, and IMAP/POP3 proxy server.
-- [Docker] (26.0): an open platform for developers and sysadmins to build, ship, and run distributed applications, whether on laptops, data center VMs, or the cloud.
+- [Docker] (26.1): an open platform for developers and sysadmins to build, ship, and run distributed applications, whether on laptops, data center VMs, or the cloud.
 - [Docker Compose] (2.26): a tool for defining and running multi-container [Docker] applications.
 - [Keras] ([TensorFlow] built-in): a high-level neural networks [API], written in [Python] and capable of running on top of [TensorFlow].
 - [TensorFlow] (2.16): an open source software [Deep Learning] library for high performance numerical computation using data flow graphs.
@@ -158,7 +158,7 @@ It is possible to execute tests of [Flask] microservice created with [pytest] fr
 ~/app# make test
 ...
 ============================= test session starts ==============================
-platform linux -- Python 3.10.14, pytest-8.1.1, pluggy-1.5.0
+platform linux -- Python 3.10.14, pytest-8.1.2, pluggy-1.5.0
 rootdir: /app/tests
 collected 2 items
 

app/app/api.py

@@ -18,10 +18,15 @@
 
 @api.route("/predictlabel", methods=["POST"])
 def predict():
-    # result dictionary that will be returned from the view
+    """
+    Predict the label of an uploaded image with the Deep Learning model.
+
+    Returns:
+        dict: The JSON response with the prediction results dictionary.
+    """
+
     result = {"success": False}
 
-    # ensure an image was properly uploaded to our endpoint
     if request.method == "POST" and request.files.get("file"):
         # read image as grayscale
         image_req = request.files["file"].read()
@@ -33,19 +38,13 @@ def predict():
 
         # classify the input image generating a list of predictions
         model = current_app.config["model"]
-        preds = model.predict(preprocessed_image)
+        predictions = model.predict(preprocessed_image)
 
         # add generated predictions to result
-        result["predictions"] = []
-
-        for i in range(0, 10):
-            pred = {"label": str(i), "probability": str(preds[0][i])}
-            result["predictions"].append(pred)
-
-        result["most_probable_label"] = str(np.argmax(preds[0]))
-
-        # indicate that the request was a success
+        result["predictions"] = [
+            {"label": str(i), "probability": str(pred)} for i, pred in enumerate(predictions[0])
+        ]
+        result["most_probable_label"] = str(np.argmax(predictions[0]))
         result["success"] = True
 
-    # return result dictionary as JSON response to client
     return jsonify(result)

app/app/model.py

@@ -13,9 +13,11 @@
 
 
 def init_model():
-    """Function that loads Deep Learning model.
+    """
+    Load the pre-trained Deep Learning model.
+
     Returns:
-        model: Loaded Deep Learning model.
+        model (tensorflow.keras.Model): The loaded Deep Learning model.
     """
 
     model = load_model(current_app.config["MODEL_PATH"])
@@ -24,16 +26,22 @@ def init_model():
 
 
 def preprocess_image(image):
-    """Function that preprocess image.
+    """
+    Preprocess an image for the Deep Learning model.
+
+    Args:
+        image (numpy.ndarray): The input image.
+
     Returns:
-        image: Preprocessed image.
+        preprocessed_image (numpy.ndarray): The preprocessed image.
     """
 
     # invert grayscale image
-    image = util.invert(image)
-    # resize and reshape image for model
-    image = transform.resize(image, (28, 28), anti_aliasing=True, mode="constant")
-    image = np.array(image)
-    image = image.reshape((1, 28 * 28))
+    inverted_image = util.invert(image)
+
+    # resize and reshape image
+    resized_image = transform.resize(inverted_image, (28, 28), anti_aliasing=True, mode="constant")
+    resized_image = np.array(resized_image)
+    preprocessed_image = resized_image.reshape((1, 28 * 28))
 
-    return image
+    return preprocessed_image

app/config.py

@@ -10,31 +10,54 @@
 
 
 class DefaultConfig:
-    if os.environ.get("SECRET_KEY"):
-        SECRET_KEY = os.environ.get("SECRET_KEY")
-    else:
+    """
+    Default configuration class.
+    """
+
+    SECRET_KEY = os.environ.get("SECRET_KEY")
+    if not SECRET_KEY:
         raise ValueError("SECRET KEY NOT FOUND!")
 
     MODEL_PATH = os.environ.get("MODEL_PATH") or "/app/mnist_model.keras"
 
     @staticmethod
     def init_app(app):
+        """
+        Initialize the application with the default configuration.
+        """
+
         print("PRODUCTION CONFIG")
 
 
 class DevConfig(DefaultConfig):
+    """
+    Development configuration class.
+    """
+
     DEBUG = True
 
     @classmethod
     def init_app(cls, app):
+        """
+        Initialize the application with the development configuration.
+        """
+
         print("DEVELOPMENT CONFIG")
 
 
 class TestConfig(DefaultConfig):
+    """
+    Testing configuration class.
+    """
+
     TESTING = True
 
     @classmethod
     def init_app(cls, app):
+        """
+        Initialize the application with the testing configuration.
+        """
+
         print("TESTING CONFIG")
 
 

app/tests/conftest.py

@@ -13,10 +13,27 @@
 
 @pytest.fixture
 def app():
+    """
+    Create a Flask app instance for testing.
+
+    Returns:
+        flask.Flask: The Flask app instance.
+    """
+
     app = create_app("testing")
     return app
 
 
 @pytest.fixture
 def client(app):
+    """
+    Create a Flask test client for the app.
+
+    Args:
+        app (flask.Flask): The Flask app instance.
+
+    Returns:
+        flask.testing.FlaskClient: The Flask test client.
+    """
+
     return app.test_client()

app/tests/test_app.py

@@ -10,13 +10,30 @@
 
 
 def test_index(client):
+    """
+    Test the index route.
+
+    Args:
+        client (flask.testing.FlaskClient): The Flask test client.
+    """
+
     response = client.get("/")
-    # check response
+
+    # assert response status
     assert response.status_code == 200
+
+    # assert response data
     assert response.data == b"Deep Learning on Flask"
 
 
 def test_api(client):
+    """
+    Test the API endpoint to predict the label of an uploaded image with the Deep Learning model.
+
+    Args:
+        client (flask.testing.FlaskClient): The Flask test client.
+    """
+
     # server REST API endpoint url and example image path
     SERVER_URL = "http://127.0.0.1:5000/api/predictlabel"
     IMAGE_PATH = "../app/static/4.jpg"
@@ -26,7 +43,7 @@ def test_api(client):
         payload = {"file": image}
         response = client.post(SERVER_URL, data=payload)
 
-        # check response
+        # assert response status
         assert response.status_code == 200
 
         # JSON format
@@ -40,11 +57,13 @@ def test_api(client):
         if json_response["success"]:
             # most probable label
             print(json_response["most_probable_label"])
+
             # predictions
             for dic in json_response["predictions"]:
                 print(f"label {dic['label']} probability: {dic['probability']}")
 
+            # assert the most probable label is 4
             assert json_response["most_probable_label"] == "4"
         # failed
         else:
-            raise AssertionError()
+            raise AssertionError("API endpoint /predictlabel failed")

docker-compose.yml

@@ -1,5 +1,3 @@
-version: "3.8"
-
 services:
   web:
     build: .

requirements.txt

@@ -2,8 +2,8 @@ Flask==3.0.3
 gunicorn==22.0.0
 numpy==1.26.4
 Pillow==10.3.0
-pytest==8.1.1
+pytest==8.1.2
 requests==2.31.0
-ruff==0.4.1
+ruff==0.4.2
 scikit-image==0.23.2
 tensorflow==2.16.1

requirements_dev.txt

@@ -1,3 +1,3 @@
 -r requirements.txt
-jupyterlab==4.1.6
+jupyterlab==4.1.8
 matplotlib==3.8.4