Update app.py

app.py

@@ -24,18 +24,24 @@ css = """
 .manuscript {
     display: flex;
     margin-bottom: 20px;
+    align-items: flex-start;  /* Align items to the top */
 }
 .annotation {
-    width: 30%;
+    width: 10%;
     padding-right: 20px;
     color: grey;
     font-style: italic;
+    padding-top: 5px;  /* Add some top padding to align with the title */
 }
 .content {
     width: 70%;
 }
-h3 {
+h2 {
     margin-top: 0;
+    margin-bottom: 10px;  /* Add some bottom margin for spacing */
+}
+.title-content {
+    margin-top: -5px;  /* Negative margin to offset the padding of annotation */
 }
 </style>
 """
@@ -104,11 +110,11 @@ def transform_chunks(marianne_segmentation):
 
     html_output = []
     for _, row in marianne_segmentation.iterrows():
-        entity_group = row['entity_group']
+        entity_group = "<" + capitalize(row['entity_group']) + ">"
         word = row['word']
         
         if entity_group == 'title':
-            html_output.append(f'<div class="manuscript"><div class="annotation">{entity_group}</div><div class="content"><h3>{word}</h3></div></div>')
+            html_output.append(f'<div class="manuscript"><div class="annotation">{entity_group}</div><div class="content title-content"><h2>{word}</h2></div></div>')
         else:
             html_output.append(f'<div class="manuscript"><div class="annotation">{entity_group}</div><div class="content">{word}</div></div>')
 
@@ -131,7 +137,7 @@ class MistralChatBot:
             batch_prompts = [editorial_text]
     
         out = token_classifier(batch_prompts)
-        out = transform_chunks(out[0])
+        out = transform_chunks(pd.concat(out))
         generated_text = f'{css}<h2 style="text-align:center">Réponse</h2>\n<div class="generation">{out}</div>'
         return generated_text