Leci37/TensorFlow-stocks-prediction-Machine-learning-RealTime

Predict operation stocks points (buy-sell) with past technical patterns, and powerful machine-learning libraries such as: Sklearn.RandomForest , Sklearn.GradientBoosting, XGBoost, Google TensorFlow and Google TensorFlow LSTM..Real time Twitter:

What it solves

This project provides a system for predicting stock market movements using machine learning. It aims to move beyond simple price prediction by providing categorical targets (buy, sell, or do nothing) based on technical indicators, and it includes a real-time evaluation system to send alerts via Telegram, Twitter, or Email.

How it works

The system follows a pipeline: it collects historical OHLCV (Open, High, Low, Close, Volume) data via APIs like Alpha Vantage or Alpaca, extracts over 600 technical indicators, and filters out noisy patterns. It then creates time-based windows of data and trains models using TensorFlow, XGBoost, or Scikit-Learn. The models are trained to recognize "ground true" points—significant price changes (e.g., 2.5% over 3 hours)—to categorize future movements.

Who it’s for

This tool is designed for traders and developers interested in applying machine learning to financial markets who have a basic understanding of Python and technical analysis.

Highlights

  • Extensive Model Testing: Supports testing across 36 different model combinations, including LSTM, GRU, and Dense networks.
  • Real-time Alerts: Integrates with Telegram, Twitter, and Mail for live market notifications.
  • Broad Indicator Set: Utilizes 637 technical stock indicators to inform predictions.
  • Categorical Output: Predicts specific actions (Buy/Sell/Hold) rather than attempting to predict a precise continuous price value.

Related

  • Project
    InfuseAI/ArtiVC

    A command-line tool for data versioning on cloud storage that allows users to snapshot and switch between data versions without requiring a separate server.

  • Project
    openucx/ucc

    A unified collective communication library designed for high-performance AI/ML and HPC workloads across CPU, GPU, and DPU hardware.

  • Project
    ThinkInAIXYZ/clawdhome

    A multi-agent security console for macOS that runs isolated AI agent instances in separate macOS user accounts to ensure data privacy and system security.

  • Project
    ViciousSquid/Dosidicus

    A transparent cognitive sandbox and digital pet simulation that allows users to learn neuroscience by raising a neural network they can see and trace in real time.