diff --git a/detectree2/models/predict.py b/detectree2/models/predict.py
index 4c774fb9..648e002a 100644
--- a/detectree2/models/predict.py
+++ b/detectree2/models/predict.py
@@ -20,6 +20,7 @@
 
 def predict_on_data(
     directory: str = "./",
+    out_folder: str = "predictions",
     predictor=DefaultPredictor,
     eval=False,
     save: bool = True,
@@ -30,7 +31,7 @@ def predict_on_data(
     Predicts crowns for all png images present in a directory and outputs masks as jsons.
     """
 
-    pred_dir = os.path.join(directory, "predictions")
+    pred_dir = os.path.join(directory, out_folder)
 
     Path(pred_dir).mkdir(parents=True, exist_ok=True)
 
diff --git a/docs/source/tutorial.rst b/docs/source/tutorial.rst
index c7610818..c74dc6b2 100644
--- a/docs/source/tutorial.rst
+++ b/docs/source/tutorial.rst
@@ -331,7 +331,7 @@ Point to a trained model, set up the configuration state and make predictions on
    
    trained_model = "./230103_randresize_full.pth"
    cfg = setup_cfg(update_model=trained_model)
-   predict_on_data(tiles_path, DefaultPredictor(cfg))
+   predict_on_data(tiles_path, predictor=DefaultPredictor(cfg))
 
 Once the predictions have been made on the tiles, it is necessary to project them back into geographic space.