Upload folder using huggingface_hub

Dockerfile

@@ -1,17 +1,16 @@
-# Use a minimal base image with Python 3.9 installed
 FROM python:3.9-slim
 
-# Set the working directory inside the container to /app
+# Set the working directory inside the container
 WORKDIR /app
 
-# Copy all files from the current directory on the host to the container's /app directory
+# Copy all files from the current directory to the container's working directory
 COPY . .
 
-# Install Python dependencies listed in requirements.txt
-RUN pip3 install -r requirements.txt
+# Install dependencies from the requirements file without using cache to reduce image size
+RUN pip install --no-cache-dir --upgrade -r requirements.txt
 
-# Define the command to run the Streamlit app on port 8501 and make it accessible externally
-#CMD ["streamlit", "run", "app.py", "--server.port=8501", "--server.address=0.0.0.0", "--server.enableXsrfProtection=false"]
-CMD ["streamlit", "run", "app.py", "--server.port=7860", "--server.address=0.0.0.0", "--server.enableXsrfProtection=false"]
-
-# NOTE: Disable XSRF protection for easier external access in order to make batch predictions
+# Define the command to start the application using Gunicorn with 4 worker processes
+# - `-w 4`: Uses 4 worker processes for handling requests
+# - `-b 0.0.0.0:7860`: Binds the server to port 7860 on all network interfaces
+# - `app:app`: Runs the Flask app (assuming `app.py` contains the Flask instance named `app`)
+CMD ["gunicorn", "-w", "4", "-b", "0.0.0.0:7860", "app:cust_predictor_api"]

app.py

@@ -1,61 +1,96 @@
+# Import necessary libraries
+import numpy as np
+import joblib  # For loading the serialized model
+import pandas as pd  # For data manipulation
+from flask import Flask, request, jsonify  # For creating the Flask API
 
-import streamlit as st
-import pandas as pd
-import joblib
-
-import warnings
-warnings.filterwarnings("ignore", message=".*ScriptRunContext.*")
-
-# Load the trained model
-def load_model():
-    return joblib.load("customer_prediction_model_v1_0.joblib")
-
-model = load_model()
-
-# Set the title of the Streamlit app
-st.title("ExtraaLearn Customer Predictor")
-st.subheader("Online Prediction")
-
-# Collect user input based on dataset columns
-# Collect user input for property features
-age                        = st.number_input("age", min_value=5, max_value=90, step=1, value=30)
-website_visits             = st.number_input("website_visits", min_value=0, step=1, value=1)
-time_spent_on_website      = st.number_input("time_spent_on_website", min_value=0, step=1, value=1)
-page_views_per_visit       = st.number_input("page_views_per_visit", min_value=0, step=1, value=1)
-current_occupation         = st.selectbox("current_occupation", ["Professional", "Student", "Unemployed"])
-first_interaction          = st.selectbox("first_interaction", ["Mobile App", "Website"])
-profile_completed          = st.selectbox("profile_completed", ["Medium", "High", "Low"])
-last_activity              = st.selectbox("last_activity", ["Website Activity", "Email Activity", "Phone Activity"])
-print_media_type1          = st.selectbox("print_media_type1", ["Yes", "No"])
-print_media_type2          = st.selectbox("print_media_type2", ["Yes", "No"])
-digital_media              = st.selectbox("digital_media", ["Yes", "No"])
-educational_channels       = st.selectbox("educational_channels", ["Yes", "No"])
-referral                   = st.selectbox("referral", ["Yes", "No"])               
-
-# Convert user input into a DataFrame
-input_data = pd.DataFrame([{
-	'age'                       :  'age',
-	'website_visits'            :  'website_visits',
-	'time_spent_on_website'     :  'time_spent_on_website',
-	'page_views_per_visit'      :  'page_views_per_visit',
-	'current_occupation'        :  'current_occupation',
-	'first_interaction'         :  'first_interaction',
-	'profile_completed'         :  'profile_completed',
-	'last_activity'             :  'last_activity',
-	'print_media_type1'         :  'print_media_type1',
-	'print_media_type2'         :  'print_media_type2',
-	'digital_media'             :  'digital_media',
-	'educational_channels'      :  'educational_channels',
-	'referral'                  :  'referral'
-}])
-
-# Set classification threshold
-classification_threshold = 0.5
-
-# Predict button
-if st.button("Predict"):
-    prediction_proba = model.predict_proba(input_data)[0, 1]
-    prediction = (prediction_proba >= classification_threshold).astype(int)
-    result = "Join" if prediction == 1 else "not join"
-    st.write(f"Prediction: The customer is likely to **{result}**.")
-    st.write(f"Churn Probability: {prediction_proba:.2f}")
+# Initialize the Flask application
+cust_predictor_api = Flask("ExtraaLearn Customer Predictor")
+
+# Load the trained machine learning model
+model = joblib.load("customer_prediction_model_v1_0.joblib")
+
+# Define a route for the home page (GET request)
+@cust_predictor_api.get('/')
+def home():
+    """
+    This function handles GET requests to the root URL ('/') of the API.
+    It returns a simple welcome message.
+    """
+    return "Welcome to the ExtraaLearn Customer Prediction API!"
+
+classification_threshold = 0.45
+
+# Define an endpoint for customer prediction (POST request)
+@cust_predictor_api.post('/v1/cust_lead')
+def predict_cust_lead():
+    """
+    This function handles POST requests to the '/v1/cust_lead' endpoint.
+    It expects a JSON payload containing customer details and returns
+    the predicted customer probability as a JSON response.
+    """
+    # Get the JSON data from the request body
+    cust_data = request.get_json()
+
+    # Extract relevant features from the JSON data
+    sample = {
+		'age' 	: cust_data['age'],
+		'current_occupation' 	: cust_data['current_occupation'],
+		'first_interaction' 	: cust_data['first_interaction'],
+		'profile_completed' 	: cust_data['profile_completed'],
+		'website_visits' 	: cust_data['website_visits'],
+		'time_spent_on_website' 	: cust_data['time_spent_on_website'],
+		'page_views_per_visit' 	: cust_data['page_views_per_visit'],
+		'last_activity' 	: cust_data['last_activity'],
+		'print_media_type1' 	: cust_data['print_media_type1'],
+		'print_media_type2' 	: cust_data['print_media_type2'],
+		'digital_media' 	: cust_data['digital_media'],
+		'educational_channels' 	: cust_data['educational_channels'],
+		'referral' 	: cust_data['referral']
+    }
+
+    # Convert the extracted data into a Pandas DataFrame
+    input_data = pd.DataFrame([sample])
+
+    # Make prediction
+    predicted_cust = model.predict_proba(input_data)[0][1]
+
+    # convert continuous prob as 0/1
+    predicted_cust = (predicted_cust >= classification_threshold).astype(int)
+
+    # Return the actual prediction status
+    return jsonify({'Predicted customer status': predicted_cust})
+
+
+# Define an endpoint for batch prediction (POST request)
+@cust_predictor_api.post('/v1/cust_lead_batch')
+def predict_cust_lead_batch():
+    """
+    This function handles POST requests to the '/v1/cust_lead_batch' endpoint.
+    It expects a CSV file containing property details for multiple properties
+    and returns the predicted status as a dictionary in the JSON response.
+    """
+    # Get the uploaded CSV file from the request
+    file = request.files['file']
+
+    # Read the CSV file into a Pandas DataFrame
+    input_data = pd.read_csv(file)
+
+    # Make predictions for all properties in the DataFrame (get log_prices)
+    predicted_cust_list = model.predict_proba(input_data)[0][1]
+    predicted_cust_list = predicted_cust_list.tolist()
+
+    # Calculate actual prices
+    predicted_cust_list = [round(float(np.exp(log_price)), 2) for log_price in predicted_log_prices]
+    predicted_cust_list = [(predicted_cust >= classification_threshold).astype(int)  for predicted_cust in predicted_cust_list]
+
+    # Create a dictionary of predictions with customer IDs as keys
+    ids = input_data['ID'].tolist()
+    output_dict = dict(zip(ids, predicted_cust_list))
+
+    # Return the predictions dictionary as a JSON response
+    return output_dict
+
+# Run the Flask application in debug mode if this script is executed directly
+if __name__ == '__main__':
+    cust_predictor_api.run(debug=True)

requirements.txt

@@ -1,7 +1,11 @@
-
 pandas==2.2.2
 numpy==2.0.2
 scikit-learn==1.6.1
 xgboost==2.1.4
 joblib==1.4.2
+Werkzeug==2.2.2
+flask==2.2.2
+gunicorn==20.1.0
+requests==2.28.1
+uvicorn[standard]
 streamlit==1.43.2