svelte-numerology-classifier-graphql Svelte Themes

Svelte Numerology Classifier Graphql

Text classification machine learning

Download

Numerology Prediction Application

A full-stack web application that predicts numerology numbers (1-33) based on keywords or text input using machine learning. Built with Svelte frontend, ASP.NET Core backend with GraphQL API, and scikit-learn for ML predictions.

Features

  • ML-Powered Predictions: Uses Support Vector Machine (SVM) to classify text into numerology numbers 1-33
  • Modern Frontend: Clean, responsive Svelte interface with real-time predictions
  • GraphQL API: Flexible and efficient API using HotChocolate
  • Persistent Storage: SQLite database stores all predictions with timestamps
  • History Tracking: View all past predictions with refresh capability
  • 330+ Training Samples: Comprehensive training dataset for accurate predictions

Technology Stack

Frontend

  • Svelte 4 - Reactive UI framework
  • Vite - Fast build tool and dev server
  • URQL - GraphQL client for Svelte
  • CSS3 - Custom styling with modern design

Backend

  • ASP.NET Core 8 - Web API framework
  • HotChocolate - GraphQL server
  • Entity Framework Core - ORM for database operations
  • SQLite - Lightweight database

Machine Learning

  • Python 3.8+ - ML runtime
  • scikit-learn - ML library (SVM classifier)
  • pandas - Data processing
  • joblib - Model serialization

Project Structure

numerology-classifier-graphql/
├── backend/
│   └── NumerologyAPI/
│       ├── Program.cs                 # Main entry point
│       ├── appsettings.json          # Configuration
│       ├── NumerologyAPI.csproj      # Project file
│       ├── Models/
│       │   └── Prediction.cs         # Data model
│       ├── Data/
│       │   └── AppDbContext.cs       # Database context
│       ├── GraphQL/
│       │   ├── Query.cs              # GraphQL queries
│       │   ├── Mutation.cs           # GraphQL mutations
│       │   └── Types/
│       │       └── PredictionType.cs # GraphQL type
│       └── Services/
│           └── PredictionService.cs  # ML service
├── frontend/
│   ├── package.json                  # Dependencies
│   ├── vite.config.js               # Vite config
│   ├── svelte.config.js             # Svelte config
│   └── src/
│       ├── main.js                  # Entry point
│       ├── app.css                  # Global styles
│       ├── App.svelte               # Root component
│       ├── routes/
│       │   ├── Predict.svelte       # Prediction page
│       │   └── History.svelte       # History page
│       └── lib/
│           └── graphql.js           # GraphQL client
├── ML/
│   ├── train_data.csv               # Training dataset (330 rows)
│   ├── train_model.py               # Model training script
│   ├── predict.py                   # Prediction script
│   ├── requirements.txt             # Python dependencies
│   └── numerology_model.pkl         # Trained model (generated)
└── README.md                        # This file

Prerequisites

  • .NET 8.0 SDK - Download
  • Node.js 18+ and npm - Download
  • Python 3.8+ - Download
  • Git (optional) - For cloning the repository

Installation

1. Clone or Download the Project

git clone <repository-url>
cd numerology-classifier-graphql

2. Setup Python ML Environment

cd ML
pip install -r requirements.txt
python3 train_model.py

Expected output:

Loading training data...
Training samples: 330
Unique numbers: [1, 2, 3, ..., 33]
...
Training accuracy: 0.xxx
Model saved to .../numerology_model.pkl

3. Setup Backend (ASP.NET)

cd ../backend
dotnet new webapi -n NumerologyAPI
cd NumerologyAPI

Install required NuGet packages:

dotnet add package Microsoft.EntityFrameworkCore.Sqlite
dotnet add package Microsoft.EntityFrameworkCore.Design
dotnet add package HotChocolate.AspNetCore
dotnet add package HotChocolate.Data.EntityFramework

Copy all the C# files into their respective folders, then:

dotnet restore
dotnet build

4. Setup Frontend (Svelte)

cd ../../frontend
npm install

Running the Application

Start the Backend

cd backend/NumerologyAPI
dotnet run --urls "http://localhost:5000"

The backend will start at: http://localhost:5000

You should see:

info: Microsoft.Hosting.Lifetime[14]
      Now listening on: http://localhost:5000

Start the Frontend (in a new terminal)

cd frontend
npm run dev

The frontend will start at: http://localhost:5173

Usage

  1. Open your browser to http://localhost:5173
  2. Predict Page:
    • Enter keywords or text (e.g., "spiritual wisdom intuition")
    • Click "Predict Number"
    • View your predicted numerology number (1-33)
  3. History Page:
    • View all past predictions
    • Click "Refresh" to reload history
    • See timestamps and input text for each prediction

GraphQL Playground

Access the GraphQL API directly at: http://localhost:5000/graphql

Example Queries

Get all predictions:

query {
  predictions {
    id
    inputText
    predictedNumber
    timestamp
  }
}

Create a new prediction:

mutation {
  addPrediction(inputText: "love harmony balance") {
    id
    inputText
    predictedNumber
    timestamp
  }
}

API Documentation

GraphQL Schema

Query Type

type Query {
  predictions: [Prediction!]!
  prediction(id: Int!): Prediction
}

Mutation Type

type Mutation {
  addPrediction(inputText: String!): Prediction!
}

Prediction Type

type Prediction {
  id: Int!
  inputText: String!
  predictedNumber: Int!
  timestamp: DateTime!
}

Machine Learning Model

Algorithm

  • SVM (Support Vector Machine) with RBF kernel
  • TF-IDF Vectorization for text feature extraction
  • Class balancing to prevent bias

Training Data

  • 330 samples across 33 numerology numbers
  • 10 variations per number for better accuracy
  • Includes synonyms, related concepts, and varied phrasings

Model Performance

  • Training accuracy typically 95%+
  • Good prediction diversity across all 33 numbers
  • Handles both single words and phrases

Retraining the Model

If you modify train_data.csv:

cd ML
python3 train_model.py

Then restart the backend to use the new model.

Configuration

Backend Port Configuration

Edit backend/NumerologyAPI/appsettings.json:

{
  "Kestrel": {
    "Endpoints": {
      "Http": {
        "Url": "http://localhost:5000"
      }
    }
  }
}

Frontend API URL

Edit frontend/src/lib/graphql.js:

export const client = new Client({
  url: 'http://localhost:5000/graphql',
  // ...
});

CORS Configuration

Edit backend/NumerologyAPI/Program.cs:

builder.Services.AddCors(options =>
{
    options.AddDefaultPolicy(policy =>
    {
        policy.WithOrigins("http://localhost:5173") // Frontend URL
              .AllowAnyHeader()
              .AllowAnyMethod();
    });
});

Troubleshooting

Python Command Not Found

Error: python: command not found

Solution: Use python3 instead. Update PredictionService.cs:

FileName = "python3",  // Change from "python" to "python3"

Model File Not Found

Error: Model not found. Please train the model first.

Solution:

cd ML
python3 train_model.py

Port Already in Use

Error: Address already in use

Solution: Change the port in appsettings.json or kill the process:

# Find the process
lsof -i :5000
# Kill it
kill -9 <PID>

CORS Errors

Error: Access to fetch at 'http://localhost:5000/graphql' from origin 'http://localhost:5173' has been blocked by CORS policy

Solution: Verify CORS configuration in Program.cs matches your frontend URL.

Database Locked Error

Error: SQLite Error: database is locked

Solution: Close any database connections and restart the backend:

rm backend/NumerologyAPI/numerology.db
dotnet run --urls "http://localhost:5000"

Low Prediction Diversity

Problem: Model predicts mostly the same numbers

Solution: Add more varied training data to train_data.csv and retrain:

cd ML
python3 train_model.py

Development

Adding New Training Data

  1. Edit ML/train_data.csv
  2. Add rows in format: number,keywords
  3. Retrain: python3 train_model.py
  4. Restart backend

Modifying the UI

Frontend files are in frontend/src/:

  • App.svelte - Main layout and navigation
  • routes/Predict.svelte - Prediction interface
  • routes/History.svelte - History display
  • app.css - Global styles

After changes, Vite will hot-reload automatically.

Database Management

The SQLite database is created automatically at:

backend/NumerologyAPI/numerology.db

To reset the database:

rm backend/NumerologyAPI/numerology.db
dotnet run

Building for Production

Frontend

cd frontend
npm run build

Output will be in frontend/dist/

Backend

cd backend/NumerologyAPI
dotnet publish -c Release -o ./publish

Output will be in backend/NumerologyAPI/publish/

License

This project is provided as-is for educational and personal use.

Acknowledgments

  • Svelte - Frontend framework
  • ASP.NET Core - Backend framework
  • HotChocolate - GraphQL implementation
  • scikit-learn - Machine learning library
  • URQL - GraphQL client

Top categories

tailwinddaisyuiadmin templatepopupmdsvexportfolioblogformecommerceuicarouselauthdarkseoimagerouting
Loading Svelte Themes