Update README.md

README.md

@@ -29,37 +29,62 @@ model-index:
       value: 0.96
 ---
 
-# Lego Brick Classification (Classical AutoML)
+# Model Card for Lego Brick Classification (Classical AutoML)
 
-This model was trained using **AutoGluon Tabular (Classical AutoML)** on the dataset [aedupuga/lego-sizes](https://huggingface.co/datasets/aedupuga/lego-sizes).  
-The task is to classify LEGO bricks into three categories: **Standard, Flat, Sloped**, given their measured dimensions (length, height, width, studs).  
+This model classifies LEGO pieces into three types — **Standard**, **Flat**, and **Sloped** — using their dimensions (Length, Height, Width, Studs).  
+It was trained with **AutoGluon Tabular AutoML**, selecting the best-performing algorithm (LightGBM).
+
+---
 
 ## Model Details
-- **Framework**: AutoGluon (TabularPredictor)
-- **Algorithms searched**: Random Forest, Gradient Boosting Trees, XGBoost, LightGBM, CatBoost
-- **Best model**: LightGBM (selected by AutoML)
-- **Training data**: 300 augmented + 30 original samples
-- **Evaluation metric**: Accuracy, Weighted F1
-
-## Results
-- Accuracy: **0.97**
-- Weighted F1: **0.96**
-
-## Dataset
-- Name: [aedupuga/lego-sizes](https://huggingface.co/datasets/aedupuga/lego-sizes)  
-- Original: 30 manually measured LEGO bricks  
-- Augmented: 300 synthetically generated samples (jittered dimensions)  
-- Features: `Length`, `Height`, `Width`, `Studs`  
-- Target: `Type (Standard / Flat / Sloped)`
-
-## Intended Use
-This model is intended for **educational practice** in AutoML and tabular classification.  
-It is **not suitable for industrial use**, given the small sample size and synthetic data.  
-
-## Limitations
-- Very small real dataset (30 samples)  
-- Synthetic augmentation may not capture all real-world variations  
-
-## Contact
-Dataset by: Anuhya Edupuganti (CMU)  
-Model by: Xinxuan Tang (CMU) - xinxuant@andrew.cmu.edu 
+
+### Model Description
+- **Developed by:** Xinxuan Tang (CMU)  
+- **Dataset curated by:** Anuhya Edupuganti (CMU)  
+- **Model type:** AutoML ensemble (best model = LightGBM)  
+- **Language(s):** N/A (tabular data)  
+- **License:** MIT  
+- **Finetuned from:** Not applicable  
+
+### Model Sources
+- **Repository:** [Hugging Face Model Repo](https://huggingface.co/)  
+- **Dataset:** [aedupuga/lego-sizes](https://huggingface.co/datasets/aedupuga/lego-sizes)
+
+---
+
+## Uses
+
+### Direct Use
+- Educational practice in **tabular classification**.  
+- Experimenting with AutoML search and hyperparameter tuning.
+
+### Downstream Use
+- Could be used as a **teaching example** for AutoML pipelines on small tabular datasets.  
+
+### Out-of-Scope Use
+- **Not suitable for industrial LEGO quality control**, since dataset is synthetic and small.
+
+---
+
+## Bias, Risks, and Limitations
+
+- **Small dataset**: only 30 original bricks, augmented to 300 synthetic samples.  
+- **Synthetic data bias**: jitter augmentation may not reflect real-world LEGO variations.  
+
+### Recommendations
+Users should treat results as **proof-of-concept** and not deploy in production.
+
+---
+
+## How to Get Started with the Model
+
+```python
+from autogluon.tabular import TabularPredictor
+import pandas as pd
+
+# Load trained predictor
+predictor = TabularPredictor.load("autogluon_model/")
+
+# Run inference
+test_data = pd.DataFrame([{"Length": 4, "Height": 1.2, "Width": 2, "Studs": 4}])
+print(predictor.predict(test_data))