Update app.py

app.py

@@ -611,14 +611,16 @@ with gr.Blocks(css="""
               <h3 style="color: var(--text-primary);">🔍 <b>LoS/NLoS Classification Task</b></h3>
               <ul style="padding-left: 20px;">
                 <li><b>🎯 Goal</b>: Classify whether a channel is <b>LoS</b> (Line-of-Sight) or <b>NLoS</b> (Non-Line-of-Sight) with very small LWM CLS embeddings.</li>
-                <li><b>📂 Dataset</b>: Use the default dataset (a combination of six scenarios from the DeepMIMO dataset) or upload your own dataset in <b>h5py</b> format.</li>
+                <li><b>📂 Dataset</b>: Use the default dataset (a combination of six scenarios from the DeepMIMO dataset) or upload your own dataset in <b>h5</b> format.</li>
                 <li><b>💡 Custom Dataset Requirements:</b>
                   <ul>
-                    <li>🛢 <b>channels</b> array: Shape (N,32,32)</li>
+                    <li>🎛 <b>channels</b> array: Shape (N,32,32), rows: 32 antennas at BS, columns: 32 antennas at UEs</li>
                     <li>🏷️ <b>labels</b> array: Binary LoS/NLoS values (1/0)</li>
                   </ul>
                 </li>
-                <li><b>🔗 Tip</b>: Instructions for organizing your dataset are available at the bottom of the page.</li>
+                <li><b>🔗 Tip 1</b>: Instructions for organizing your dataset are available at the bottom of the page.</li>
+                <li><b>🔗 Tip 2</b>: As the computations and inference are performed on HuggingFace CPUs, please use small datasets for faster demo experience (say <500 samples). </li>
+                <li><b>🔗 Tip 3</b>: Your dataset will be normalized automatically based on outdoor environments. </li>
                 <li><b>💼 No Downstream Model</b>: Instead of a complex downstream model, we classify each sample based on its distance to the centroid of training samples from each class (LoS/NLoS).</il>
               </ul>
             </div>