diff --git a/src/deepforest/dataset.py b/src/deepforest/dataset.py
index 1cc595ce..f0421465 100644
--- a/src/deepforest/dataset.py
+++ b/src/deepforest/dataset.py
@@ -66,7 +66,11 @@ def __init__(self,
         Returns:
             If train, path, image, targets else image
         """
-        self.annotations = pd.read_csv(csv_file)
+        # Check if csv_file is a DataFrame or a file path
+        if isinstance(csv_file, pd.DataFrame):
+            self.annotations = csv_file
+        else:
+            self.annotations = pd.read_csv(csv_file)
         self.root_dir = root_dir
         if transforms is None:
             self.transform = get_transform(augment=train)
diff --git a/src/deepforest/main.py b/src/deepforest/main.py
index 320beca0..82ca4f9e 100644
--- a/src/deepforest/main.py
+++ b/src/deepforest/main.py
@@ -420,6 +420,7 @@ def predict_image(self,
             result = utilities.read_file(result, root_dir=root_dir)
 
         return result
+    
 
     def predict_file(self, csv_file, root_dir, savedir=None, color=None, thickness=1):
         """Create a dataset and predict entire annotation file Csv file format
@@ -431,7 +432,7 @@ def predict_file(self, csv_file, root_dir, savedir=None, color=None, thickness=1
         Deprecation warning: The return_plot argument is deprecated and will be removed in 2.0. Use visualize.plot_results on the result instead.
 
         Args:
-            csv_file: path to csv file
+            csv_file (str or pd.DataFrame): Path to a CSV file or a DataFrame with annotations.
             root_dir: directory of images. If none, uses "image_dir" in config
             (deprecated) savedir: directory to save images with bounding boxes
             (deprecated) color: color of the bounding box as a tuple of BGR color, e.g. orange annotations is (0, 165, 255)
@@ -441,11 +442,17 @@ def predict_file(self, csv_file, root_dir, savedir=None, color=None, thickness=1
             df: pandas dataframe with bounding boxes, label and scores for each image in the csv file
         """
 
-        df = utilities.read_file(csv_file)
-        ds = dataset.TreeDataset(csv_file=csv_file,
+        # Use DataFrame directly if provided, otherwise treat as file path
+        if isinstance(csv_file, pd.DataFrame):
+            df = csv_file
+        else:
+            df = utilities.read_file(csv_file)
+
+        ds = dataset.TreeDataset(csv_file=df,
                                  root_dir=root_dir,
                                  transforms=None,
                                  train=False)
+
         dataloader = self.predict_dataloader(ds)
 
         results = predict._dataloader_wrapper_(model=self,
diff --git a/tests/test_main.py b/tests/test_main.py
index a8c3543e..f4019335 100644
--- a/tests/test_main.py
+++ b/tests/test_main.py
@@ -649,8 +649,7 @@ def test_predict_tile_with_crop_model(m, config):
         "xmin", "ymin", "xmax", "ymax", "label", "score", "cropmodel_label", "geometry",
         "cropmodel_score", "image_path"
     }
-
-
+    
 def test_predict_tile_with_crop_model_empty():
     """If the model return is empty, the crop model should return an empty dataframe"""
     raster_path = get_data("SOAP_061.png")
@@ -673,4 +672,4 @@ def test_predict_tile_with_crop_model_empty():
                             crop_model=crop_model)
 
     # Assert the result
-    assert result is None
+    assert result is None
\ No newline at end of file