Summary

This repository contains the codebase for the "Make Money with Tensorflow 2.0" YouTube tutorial by Siraj Raval. It demonstrates the development of NeuralFund, an application leveraging deep learning for automated investment decisions, targeting users interested in applying AI to financial markets.

How It Works

The project builds upon a TensorFlow Serving web app skeleton, integrating TF Serving with Flask to establish a continuous training pipeline. It merges this with a Flask boilerplate that includes user authentication and MySQL database integration. The core approach involves training models on historical stock data and news sentiment to predict optimal investment strategies.

Quick Start & Requirements

• Install/Run: Requires downloading and running a separate TF Serving web app skeleton, then merging code from this repository's Flask boilerplate (user auth + MySQL). Specific execution commands are not provided.
• Prerequisites: TensorFlow 2.0, Flask, TensorFlow Serving.
• Setup: Described as a "work in progress" with several manual integration steps (TODOs), indicating a complex setup process.

Highlighted Details

• Features a continuous training pipeline for financial market analysis.
• Incorporates user authentication and MySQL database connectivity.
• Outlines future enhancements: real-time stock data fetching, multi-model stock prediction, sentiment analysis using BERT on news data, and Deep Reinforcement Learning for trading strategies.

Maintenance & Community

The README encourages pull requests for code improvement, suggesting community involvement is welcomed. However, it does not detail specific community channels, active contributors, or a formal roadmap.

Licensing & Compatibility

No license information is specified in the provided README.

Limitations & Caveats

This project is explicitly labeled a "work in progress" with significant "TODO" items. Key functionalities like real-time data integration, advanced prediction models, sentiment analysis, and reinforcement learning are yet to be implemented. The setup involves manual merging of multiple codebases, and the current state is experimental.