Update README.md

README.md

@@ -1,11 +1,11 @@
 ---
 license: mit
 ---
-# Model card - Ikimina Reliability Index
+# Model card — Ikimina Reliability Index
 
-**Artifacts:** On the Hugging Face repo, open the **Files** tab. You should see **`calibrated_model.joblib`** and **`model_meta.json`** (required), and optionally **`month_imputes.json`**, **`group_reliability.json`**.
+**Artifacts:** Use the **Files** tab on this repository. Required: **`calibrated_model.joblib`**, **`model_meta.json`**. Optional: **`month_imputes.json`**, **`group_reliability.json`**.
 
-**Reproducibility:** Full pipeline code is in your submitted project (`generate_data.py`, `features.py`, `train_model.py`, `scorer.py`, `train.ipynb`). If the brief requires a public Git URL, paste it on this line: _[repository URL]_.
+**Source code:** This repository stores checkpoints only. The training and scoring pipeline is implemented in **`generate_data.py`**, **`features.py`**, **`train_model.py`**, **`scorer.py`**, and **`train.ipynb`**, supplied alongside this model as part of the same deliverable.
 
 **What this is:** CPU **XGBoost** (`tree_method="hist"`) + **`CalibratedClassifierCV`** (sigmoid, `cv=4`), saved as **`calibrated_model.joblib`** (scikit-learn / **joblib** — not Transformers). Load in Python with **`joblib.load`**. Use **`model_meta.json`** with it: **14** feature names in order (`feature_columns`), score formula, holdout **metrics**, tier labels.
 
@@ -20,3 +20,4 @@ license: mit
 **Use / limits:** Prioritization / exploration only — **not** an automatic lending decision. Simulated labels; **no** fairness audit on protected attributes (not in schema). Hyperparameters **hand-tuned**, not full HPO.
 
 **Privacy:** Scores only with **member consent** and **minimum necessary** identifiers.
+