Update app.py

app.py

@@ -16,23 +16,33 @@ device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 # Load PathDino model and image transforms
 model, image_transforms = get_pathDino_model("PathDino512.pth")
 
+# Increase the width of the sidebar
+st.write(
+    f"""
+    <style>
+    .sidebar .sidebar-content {{
+        width: 420px;
+    }}
+    </style>
+    """
+)
 
 st.sidebar.markdown("### PathDino")
 st.sidebar.markdown(
-    "PathDino is a lightweight histology transformer consisting of just five small vision transformer blocks. "
-    "PathDino is a customized ViT architecture, finely tuned to the nuances of histological images. It not only exhibits "
+    "PathDino is a lightweight Histopathology transformer consisting of just five small vision transformer blocks. "
+    "PathDino is a customized ViT architecture, finely tuned to the nuances of histology images. It not only exhibits "
     "superior performance but also effectively reduces susceptibility to overfitting, a common challenge in histology "
     "image analysis.\n\n"
 )
 
 default_image_url_compare = "images/HistRotate.png"
-st.sidebar.image(default_image_url_compare, caption='A 360 rotation augmentation for training models on histopathology images. Unlike training on natural images where the rotation may change the context of the visual data, rotating a histopathology patch does not change the context and it improves the learning process for better reliable embedding learning.', width=500)
+st.sidebar.image(default_image_url_compare, caption='A 360 rotation augmentation for training models on histopathology images. Unlike training on natural images where the rotation may change the context of the visual data, rotating a histopathology patch does not change the context and it improves the learning process for better reliable embedding learning.', width=300)
 
 default_image_url_compare = "images/FigPathDino_parameters_FLOPs_compare.png"
-st.sidebar.image(default_image_url_compare, caption='PathDino Vs its counterparts. Number of Parameters (Millions) vs the patch-level retrieval with macro avg F-score of majority vote (MV@5) on CAMELYON16 dataset. The bubble size represents the FLOPs.', width=500)
+st.sidebar.image(default_image_url_compare, caption='PathDino Vs its counterparts. Number of Parameters (Millions) vs the patch-level retrieval with macro avg F-score of majority vote (MV@5) on CAMELYON16 dataset. The bubble size represents the FLOPs.', width=300)
 
 default_image_url_compare = "images/ActivationMap.png"
-st.sidebar.image(default_image_url_compare, caption='Attention Visualization. When visualizing attention patterns, our PathDino transformer outperforms HIPT-small and DinoSSLPath, despite being trained on a smaller dataset of 6 million TCGA patches. In contrast, DinoSSLPath and HIPT were trained on much larger datasets, with 19 million and 104 million TCGA patches, respectively.', width=500)
+st.sidebar.image(default_image_url_compare, caption='Attention Visualization. When visualizing attention patterns, our PathDino transformer outperforms HIPT-small and DinoSSLPath, despite being trained on a smaller dataset of 6 million TCGA patches. In contrast, DinoSSLPath and HIPT were trained on much larger datasets, with 19 million and 104 million TCGA patches, respectively.', width=300)
 
 
 
@@ -67,7 +77,7 @@ def generate_activation_maps(image):
     return attention_list
 
 # Streamlit UI
-st.title("PathDino - Compact ViT for Histolopathology Image Analysis")
+st.title("PathDino - Compact ViT for Histopathology Image Analysis")
 st.write("Upload a histology image to view the activation maps.")
 
 # uploaded_image = st.file_uploader("Upload an image", type=["jpg", "png", "jpeg"])