Workflows.txt

START:
1. TRAINING PHASE (Done separately)
   - Collect or load dataset containing symptoms and corresponding diseases
   - Split data into features (symptoms) and labels (diseases)
   - Preprocess data (encoding)
   - Train the machine learning model on training data
   - Test the model on the test data to evaluate performance
   - Once satisfied, save the trained model to a file (using `joblib` or `pickle`)
     - Save model to "{name of model}.pkl"(name of the best model you choose)

2. FLASK APPLICATION PHASE
   - Import Flask, render_template, request
   - Import the necessary library to load the ML model (`pickle`)
   - Import any libraries needed for input preprocessing (e.g., `numpy`, `pandas`)

   2.1 START FLASK APP
       - Initialize Flask app

   2.2 LOAD SAVED MODEL
       - Load the previously saved ML model from the file (e.g., "model.pkl")

3. DEFINE ROUTES
   3.1 HOMEPAGE ROUTE (GET)
       - Create a `GET` route to display the homepage
       - Render an HTML page containing a form for user input (symptoms)

   3.2 PREDICTION ROUTE (POST)
       - When user submits symptoms via the form:
         - Extract user input from the request
         - Preprocess the input data (e.g., convert symptoms to numerical format)
         - Use the ML model to predict the disease based on symptoms
         - Capture the model’s prediction
         - Retrieve additional information based on the prediction:
           - Disease Description
           - Precautions
           - Medications
           - Recommended Workouts
           - Suggested Diets

4. RETURN DETAILED INFORMATION TO USER
   - Pass the disease description, precautions, medications, workouts, and diets back to the frontend
   - Display all information on the webpage

5. RUN FLASK SERVER
   - Start the Flask app, allowing users to interact with the web interface