Update README.md

README.md

@@ -1,12 +1,56 @@
 ---
-title: Laptop Specs
+title: Laptop Price Estimator
+tags:
+  - tabular-data
+  - xgboost
+  - onnx
+  - pricing
+  - algeria
 emoji: 💻
 colorFrom: indigo
-colorTo: red
+colorTo: blue
 sdk: static
 pinned: false
 license: apache-2.0
-short_description: 'Client web app for laptop price estimation for the Algerian '
+short_description: 'Client web app for laptop price estimation for the Algeria'
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+# Laptop Price Estimator
+
+Client web app for laptop price estimation for the Algerian from multi-market public dataset. The XGBoost model runs directly in your browser via ONNX WebAssembly — no backend, no API calls, no data leaves your machine.
+
+## How it works
+
+1. **ONNX Runtime Web** loads the 2.3 MB model via WebAssembly
+2. You tweak laptop specs (brand, CPU, RAM, GPU, etc.)
+3. Feature engineering + inference happens entirely in JavaScript
+4. Price prediction + per-feature contribution hints appear instantly
+
+## Model
+
+| Metric | Value |
+|--------|-------|
+| R² | 0.8665 |
+| MAPE | 17.8% |
+| Within ±10% | 47.7% |
+| Within ±20% | 73.9% |
+| Within ±30% | 86.0% |
+| Training Samples | 14,815 |
+| Features | 32 |
+
+## Tech
+
+- **Model**: XGBoost v4 → ONNX (no Python at runtime)
+- **Inference**: [onnxruntime-web](https://github.com/microsoft/onnxruntime) (WASM)
+- **UI**: Vanilla HTML/CSS/JS (zero dependencies beyond fonts)
+- **All encoders & metadata** embedded directly in `index.html`
+
+## Files
+
+- `index.html` — Self-contained frontend with embedded artefacts + all logic
+- `model.onnx` — XGBoost v4 exported to ONNX format
+
+## Dataset
+
+Data can be found at: [bitsnaps/laptops_specs](https://huggingface.co/datasets/bitsnaps/laptops_specs).