AaravMehta-07/LSTM-Random-Forest-XGBoost-Stock-Predictor-with-Optuna — reverse-engineered prompt

Build me a Python command line stock prediction tool that can download real stock price data from Yahoo Finance, create useful indicators like RSI, moving averages, volatility, and volume trends, then train three models to predict whether the stock looks like a buy or sell.

I want it to combine an LSTM model, a Random Forest model, and an XGBoost model, then use soft voting or a similar ensemble idea to give one final trade signal. Add an optional mode that uses Optuna to tune the models for better F1 score and accuracy, but also make it possible to run without tuning for faster testing.

Please include model saving and loading so I don’t have to retrain every time. Also add a simple backtesting script that simulates the strategy on historical data and prints useful results. Keep it CLI based, clean, and easy to run. Include setup instructions and a requirements file. Look up current docs online if you need to.