deshpandenu/Time-Series-Forecasting-of-Amazon-Stock-Prices-using-Neural-Networks-LSTM-and-GAN- — reverse-engineered prompt

Build me a Python Jupyter notebook project that studies and predicts Amazon stock prices from the included data files.

I want it to feel like a complete stock forecasting experiment, not just one model. Load and clean the Amazon price data, create useful features, show the trends with clear charts, then try classic time series methods like ARIMA and Fourier style forecasting. After that, build neural network models that use multiple features to predict price movement, including LSTM and GRU models, and show how well they do.

Also include a notebook section for sentiment analysis using Amazon related news and Reddit text if sample data or an API is available, then connect that sentiment back to the stock prediction work. Add a GAN based experiment too, with a generator using stock data and a CNN style discriminator.

Please make the notebooks easy to follow, with explanations, plots, model comparisons, and simple instructions so I can run everything end to end.