diff --git a/misinformation/multimodal_search.py b/misinformation/multimodal_search.py
index 678bf91..e7d7cdf 100644
--- a/misinformation/multimodal_search.py
+++ b/misinformation/multimodal_search.py
@@ -154,17 +154,12 @@ class MultimodalSearch(AnalysisMethod):
 
         return features_text
 
-
     def parsing_images(
         self,
         model_type,
         path_to_saved_tensors="./saved_tensors/",
         path_to_load_tensors=None,
     ):
-
-
-
-
         if model_type in ("clip_base", "clip_vitl14_336", "clip_vitl14"):
             path_to_lib = lavis.__file__[:-11] + "models/clip_models/"
             url = "https://raw.githubusercontent.com/salesforce/LAVIS/main/lavis/models/clip_models/bpe_simple_vocab_16e6.txt.gz"
@@ -470,7 +465,6 @@ class MultimodalSearch(AnalysisMethod):
         return att_map
 
     def upload_model_blip2_coco(self):
-
         itm_model = load_model(
             "blip2_image_text_matching",
             "coco",
@@ -481,7 +475,6 @@ class MultimodalSearch(AnalysisMethod):
         return itm_model, vis_processor
 
     def upload_model_blip_base(self):
-
         itm_model = load_model(
             "blip_image_text_matching",
             "base",
@@ -492,7 +485,6 @@ class MultimodalSearch(AnalysisMethod):
         return itm_model, vis_processor
 
     def upload_model_blip_large(self):
-
         itm_model = load_model(
             "blip_image_text_matching",
             "large",