Update README.md

README.md

@@ -1,12 +1,32 @@
-# DataSynthis_ML_JobTask - Stock Price Forecasting (Deliverables)
-This bundle contains a ready-to-run Jupyter notebook, a small synthetic sample dataset, and helper scripts for training ARIMA, LSTM, and Prophet models; evaluating them with rolling-window evaluation; and steps to deploy model artifacts to Hugging Face Hub (you must run that step locally or from an environment with internet access).
+# 📈 Stock Price Forecasting - DataSynthis ML Job Task
 
-## Files included
-- `stock_forecasting_notebook.ipynb` — The full notebook with preprocessing, ARIMA, LSTM, Prophet, rolling-window evaluation, plots, and saving models.
-- `sample_stock.csv` — A synthetic daily closing-price CSV (2020-01-01 to 2021-12-31, business days) to run the notebook offline.
-- `requirements.txt` — Python dependencies.
-- `upload_to_hf.py` — Example script to upload model files to Hugging Face Hub (requires `huggingface_hub` and HF token).
-- `README.md` — This file.
+This repository contains implementations of **time-series forecasting** for stock prices using both **traditional statistical models (ARIMA, Prophet)** and **deep learning (LSTM)**.  
+The project demonstrates model comparison, rolling-window evaluation, and deployment to Hugging Face Hub.
+
+---
+
+##  Project Overview
+- **Dataset**: Daily stock price dataset (closing prices).  
+- **Models Implemented**:
+  - ARIMA (AutoRegressive Integrated Moving Average)
+  - Prophet (Additive Time Series Forecasting by Meta)
+  - LSTM (Long Short-Term Memory Neural Network)
+- **Evaluation**:
+  - Rolling-window forecasts
+  - Metrics: RMSE, MAPE
+- **Deployment**:
+  - Models and results shared on Hugging Face Hub.
+
+---
+
+## Repository Contents
+- `lstm_model.h5` – Trained LSTM model  
+- `scaler.pkl` – Scaler used for preprocessing  
+- `performance_summary.csv` – Comparison of ARIMA, Prophet, and LSTM performance  
+- `stock_forecasting_notebook.ipynb` – Full notebook with preprocessing, training, evaluation, and plots  
+- `upload_to_hf.py` – Script for uploading to Hugging Face Hub  
+
+---
 
 ## Quick start
 1. Create and activate a python environment (recommended: conda or venv)
@@ -23,11 +43,22 @@ This bundle contains a ready-to-run Jupyter notebook, a small synthetic sample d
 3. The notebook contains cells to download real stock data via `yfinance` (if you have internet) or use the included `sample_stock.csv` for an offline demo.
 
 ## Hugging Face deployment (notes)
-- You cannot deploy directly from this notebook in an environment without internet.
 - Use `upload_to_hf.py` to push saved model files to the HF repo `DataSynthis_ML_JobTask` after creating it on the Hugging Face website (or the script will create the repo for you if you provide a valid token).
 - Create a HF token at https://huggingface.co/settings/tokens and set environment variable `HF_TOKEN` or pass `--token` to the script.
 
-## About results
-- The notebook runs quick examples and shows how to compute RMSE and MAPE, and how to perform rolling-window evaluation.
-- For production-quality training and evaluation (e.g., longer LSTM training), run on a machine with a GPU and more data.
+## Results
+
+The performance of the three models on stock price forecasting is summarized below:
+
+| Model   | RMSE   | MAPE (%) |
+|---------|--------|----------|
+| ARIMA   | 3.3748 | 1.8973   |
+| Prophet | 4.7650 | 3.1859   |
+| LSTM    | 2.0890 | 1.2516   |
+
+### Key Insights
+- **LSTM** achieved the **lowest RMSE and MAPE**, showing the best accuracy.  
+- **ARIMA** performed reasonably well, but less effective with non-linear trends.  
+- **Prophet** captured trends and seasonality but had higher errors.  
+- Overall, **LSTM is the most reliable model** for this task.