How sifat192/amazon-stock-forecasting-pytorch works
This is an educational stock-price forecasting workflow, not a reusable application or service. A user runs the EDA notebook to inspect historical prices and export the closing-price series, then runs the forecasting notebook to prepare chronological train/test sequences, fit an LSTM, and evaluate it against a previous-close baseline. The README documents these responsibilities and file handoffs; notebook source was not available for inspection, so internal implementation details are limited to those documented stages.
Main entry points: 01_eda.ipynb Explore and export · 02_lstm_forecasting.ipynb Sequence preparation · Amazon_stock_closing_prices.csv Closing-price series · AMZN.csv Historical prices
Components 8
Data Exploration
- Explore and export
01_eda.ipynb - Historical prices
AMZN.csv - Closing-price series
Amazon_stock_closing_prices.csv
Forecasting
- Sequence preparation
02_lstm_forecasting.ipynb - Chronological split and scaling
02_lstm_forecasting.ipynb - LSTM training
02_lstm_forecasting.ipynb
Evaluation
- Prediction evaluation
02_lstm_forecasting.ipynb
Other components
- Notebook user
Connections 8
- Notebook user → Explore and export: runs first
README.md - Explore and export → Historical prices: reads
README.md - Explore and export → Closing-price series: exports
README.md - Notebook user → Sequence preparation: runs next
README.md - Closing-price series → Sequence preparation: provides prices
README.md - Sequence preparation → Chronological split and scaling: builds examples
README.md - Chronological split and scaling → LSTM training: provides scaled data
README.md - LSTM training → Prediction evaluation: provides predictions
README.md