Solves issue #2

Author: lopezco

README.md

@@ -33,9 +33,10 @@ or one of the scripts in the `python/example` folder
     ```javascript
     {
         "model": trained_model,
-        "metadata": {"features": [{"name": "feature1", "type": "numeric"},
-                                  {"name": "feature2", "type": "numeric", "default": -1},
-                                  {"name": "feature3", "type": "numeric"}]}
+        "metadata": {"features": [
+            {"name": "feature1", "type": "numeric"},
+            {"name": "feature2", "type": "numeric", "default": -1},
+            {"name": "feature3", "type": "category", "categories": ["A", "B"]}]}
     }
     ```
 

python/example/build_lgbm_binary.py

@@ -0,0 +1,20 @@
+from lightgbm import LGBMClassifier
+import pandas as pd
+import joblib
+
+
+df = pd.DataFrame({"feature1": [1.,1.,5.], "feature2": [2.,2.,5.], "feature3": ["B","B","A"]})
+df['feature3'] = df['feature3'].astype('category')
+
+model_path = './model.joblib'
+
+model = LGBMClassifier()
+model.fit(df, [0, 0, 1])
+to_save = dict(model=model,
+               metadata={"features": [
+                   {"name": "feature1", "type": "numeric"},
+                   {"name": "feature2", "type": "numeric", "default": -1},
+                   {"name": "feature3", "type": "category", "categories": ["A", "B"]}]})
+
+with open(model_path, 'wb') as fo:
+    joblib.dump(to_save, fo)