TensorForce - An ML-based automated stock trading platform.

---

Table of Contents

• Description
• Links
• Tech Stack
• Future Scope
• Applications
• Project Setup
• Usage
• Team Members
• Mentors
• Screenshots

📝 Description

Are you a trader who doesn't get the time to actively execute each trade or an ML programmer with a knack for building stock trading models but don't have the capabilities to deploy them? Fret not, for TensorForce has got your back! TensorForce is an ML-based web application designed to let users invest in various stocks using Machine Learning models to execute trades on their behalf without human intervention. Along with using the in-house built models, it also is a marketplace from which users can purchase other models from developers, which will cater to users with varied trading preferences and strategies, expanding the platform's appeal. The cutting-edge Dashboard UI and the auxiliary features such as the sentiment analysis and the Stock charts viewer provide a seamless experience for the users to monitor their portfolios and trades and analyze the current market sentiment effectively, making TensorForce a one-stop shop for users fascinated with smart trading and investing.

Features

1. Dashboard
2. Sentiment Analysis
3. Heatmap
4. Shop
5. News
6. Stock Charts

🎥 Demo Video

Inheritance.Tensorforce.Demo.Video.mp4

🔗 Links

• GitHub Repository
• Demo Video
• Drive Link to Screenshots of the project
• Hosted Website Link
• Hosted Backend Link

🤖 Tech-Stack

Front-end

• HTML
• CSS
• JavaScript
• React
• Tailwind CSS

Back-end

• NodeJS
• Flask

Database

• MongoDB

Machine Learning

• Python
• Jupyter Notebook
• TensorFlow
• PyTorch

🔮 Future Scope

• An intraday trading bot which is trained on the historical intraday data of various stocks and will execute trades based on daily as well as intraday signals.
• An LLM chatbot fine-tuned to recent financial and economic news which can help users with the basics of investing and trading and also help them assess the performance of a particular company over the previous quarter or financial year.

💸 Applications

TensorForce has the potential to address several real-life problems and offer valuable applications in the field of stock trading and investing. Some of its possible applications include:

1. Automated Trading for Busy Individuals: TensorForce allows busy individuals who are interested in trading but lack the time to actively manage their portfolios to still participate in the stock market. By utilizing machine learning models to execute trades on behalf of users, TensorForce enables them to invest without needing to dedicate significant time to monitoring the market.

2. Algorithmic Trading for Efficient Execution: The use of machine learning models in TensorForce enables algorithmic trading strategies to be employed, which can execute trades with greater efficiency and speed than manual trading. This can potentially lead to improved trading performance and better capital utilization.

3. Access to Diverse Trading Strategies: The marketplace feature of TensorForce provides users with access to a variety of trading models developed by different developers. This allows users to explore and utilize diverse trading strategies tailored to their preferences and risk tolerance, enhancing their investment experience.

4. Portfolio Monitoring and Analysis: The dashboard UI of TensorForce offers users a comprehensive view of their portfolios, including performance metrics, asset allocation, and historical data. This enables users to monitor their investments effectively, identify trends, and adjust their strategies accordingly.

💲 Monetization Opportunities for Developers:

• Model Marketplace Fees: Developers who list their trading models on the marketplace can pay fees or commissions for each sale made through the platform.
• Premium Features: Certain training models or portfolio analysis tools can be offered as premium upgrades for users willing to pay an additional fee.

Project Setup:

1. Clone the entire repository from GitHub:

   git clone https://github.com/manascb1344/Inheritance

Frontend Setup

1. Navigate to the WebD directory:

   cd Inheritance/WebD

2. Install frontend dependencies:

   yarn

Backend Setup

1. Navigate to the backend directory:

   cd ../backend

2. Install backend dependencies:

   yarn

Machine Learning Setup

1. Navigate to the ML directory:

   cd ../ML

2. Install machine learning dependencies:

   pip install -r requirements.txt

Usage

Frontend

1. After completing the frontend setup, start the frontend server:

   yarn run dev

2. Access the frontend application in your browser at http://localhost:3000.

Backend

1. After completing the backend setup, start the backend server:

   yarn run dev

2. The backend server will be running at https://tensorforce-backend.onrender.com.

Machine Learning

1. After completing the machine learning setup, you can run the machine learning scripts as needed by navigating to the ML directory and executing the Python scripts.

👨‍💻 Team Members

• Manas Bavaskar: Email: [email protected]
• Yash Kadam: Email: [email protected]
• Soham Mukane: Email: [email protected]
• Aarya Bodas: Email: [email protected]

📱 Screenshots

![order history](https![model shop]

🤝 Want to Contribute?

We welcome contributions to TensorForce! If you're interested in helping out, please check out our Contributing Guide for more information on how to get started.