Update app.py

app.py

@@ -30,15 +30,13 @@ def load_dataset():
     dataset = datasets.load_dataset('rjadr/ditaduranuncamais', split='train', use_auth_token=token)
     dataset.add_faiss_index(column="txt_embs")
     dataset.add_faiss_index(column="img_embs")
-    dataset = dataset.remove_columns(['Post Created','Post Created Time','Like and View Counts Disabled','Link','Photo','Title','Sponsor Id','Sponsor Name','Download URL', 'image', 'Views', 'text_full'])
+    dataset = dataset.remove_columns(['Post Created','Like and View Counts Disabled','Link','Download URL','Views'])
     return dataset
 
 @st.cache_data(show_spinner=False)
 def load_dataframe(_dataset):
     dataframe = _dataset.remove_columns(['txt_embs', 'img_embs']).to_pandas()
     dataframe['image_base64'] = dataframe['image_base64'].str.decode('utf-8')
-    dataframe['Overperforming Score (weighted  —  Likes 1x Comments 1x )'] = dataframe['Overperforming Score (weighted  —  Likes 1x Comments 1x )'].str.replace(',','').astype(float)
-    dataframe['Total Interactions'] = dataframe['Total Interactions'].str.replace(',','').astype(int)
     return dataframe
 
 @st.cache_resource(show_spinner=True)
@@ -166,7 +164,7 @@ def postprocess_results(scores, samples):
             samples_df["score"].max() - samples_df["score"].min())) * 100
     samples_df["score"] = samples_df["score"].astype(int)
     samples_df.reset_index(inplace=True, drop=True)
-    samples_df = samples_df[['Post Created Date', 'image_base64', 'Description', 'Image Text', 'Account', 'User Name'] + [col for col in samples_df.columns if col not in ['Post Created Date', 'image_base64', 'Description', 'Image Text', 'Account', 'User Name']]]  
+    samples_df = samples_df[['Post Created', 'image', 'Description', 'Image Text', 'Account', 'User Name'] + [col for col in samples_df.columns if col not in ['Post Created', 'image', 'Description', 'Image Text', 'Account', 'User Name']]]  
     return samples_df.drop(columns=['txt_embs', 'img_embs'])
 
 @st.cache_data
@@ -268,76 +266,36 @@ if check_password():
     tab1, tab2, tab3 = st.tabs(["Data exploration", "Semantic search", "Stats"])
     
     with tab1:
-        # Initialization
-        if 'rows_per_page' not in st.session_state:
-            st.session_state['rows_per_page'] = 25
-        if 'page_number' not in st.session_state:
-            st.session_state['page_number'] = 1
-    
-        filtered_df = filter_dataframe(df)
-        max_page = -(-len(filtered_df) // st.session_state['rows_per_page'])  # ceiling division
-    
-        start_index = st.session_state['rows_per_page'] * (st.session_state['page_number'] - 1)
-        end_index = start_index + st.session_state['rows_per_page']
-        sub_df = filtered_df.iloc[start_index:end_index]
-        # sort columms order: Post Created Date, image_base64, Description, Image Text, Account, User Name and then the rest
-        sub_df = sub_df[['Post Created Date', 'image_base64', 'Description', 'Image Text', 'Account', 'User Name'] + [col for col in sub_df.columns if col not in ['Post Created Date', 'image_base64', 'Description', 'Image Text', 'Account', 'User Name']]]
-    
-        col1, col2, col3, col4 = st.columns(4)
-    
-        with col4:
-            rows_per_page = st.selectbox('Rows per page', [25, 50, 75, 100, 150, 200], index=0, key='rows_per_page_select')
-            if rows_per_page != st.session_state['rows_per_page']:
-                st.session_state['rows_per_page'] = rows_per_page
-                st.session_state['page_number'] = 1  # Reset page number when rows per page changes
-                st.experimental_rerun()
-    
-        with col2:
-            page_select = st.selectbox('Jump to page', options=range(1, max_page + 1), index=st.session_state['page_number']-1, key='page_number_select')
-            if page_select != st.session_state['page_number']:
-                st.session_state['page_number'] = page_select
-                st.experimental_rerun()
-    
-        with col1:
-            if st.button('Previous'):
-                st.session_state['page_number'] = max(1, st.session_state['page_number'] - 1)
-                st.experimental_rerun()
-    
-        with col3:
-            if st.button('Next'):
-                st.session_state['page_number'] = min(max_page, st.session_state['page_number'] + 1)
-                st.experimental_rerun()
-    
         st.dataframe(
-            data=sub_df,
+            data=filter_dataframe(df),
+        # use_container_width=True,
             column_config={
-                "image_base64": st.column_config.ImageColumn(
-                    "image", help="Instagram image"
+                "image": st.column_config.ImageColumn(
+                    "Image", help="Instagram image"
                 ),
                 "URL": st.column_config.LinkColumn(
-                    "link", help="Instagram link", width="small"
+                    "Link", help="Instagram link", width="small"
                 )
             },
-            # hide_index=True,
+            hide_index=True,
         )
     
-    
     with tab2:
         tabs = ["Text to Text", "Text to Image", "Image to Image", "Image to Text"]
         selected_tab = st.radio("Select a search type", tabs)
     
         if selected_tab == "Text to Text":
             text_to_text_input = st.text_input("Enter text")
-            text_to_text_k_top = st.slider("Number of results", 1, 20, 8)
+            text_to_text_k_top = st.slider("Number of results", 1, 60, 8)
             if st.button("Search"):
                 st.dataframe(
                     data=text_to_text(text_to_text_input, text_to_text_k_top),
                     column_config={
-                    "image_base64": st.column_config.ImageColumn(
-                        "image", help="Instagram image"
+                    "image": st.column_config.ImageColumn(
+                        "Image", help="Instagram image"
                     ),
                     "URL": st.column_config.LinkColumn(
-                        "link", help="Instagram link", width="small"
+                        "Link", help="Instagram link", width="small"
                     )
                     },
                     hide_index=True,
@@ -345,50 +303,50 @@ if check_password():
                 
         elif selected_tab == "Text to Image":
             text_to_image_input = st.text_input("Enter text")
-            text_to_image_k_top = st.slider("Number of results", 1, 20, 8)
+            text_to_image_k_top = st.slider("Number of results", 1, 60, 8)
             if st.button("Search"):
                 st.dataframe(
                     data=text_to_image(text_to_image_input, text_to_image_k_top),
                     column_config={
                         "image_base64": st.column_config.ImageColumn(
-                            "image", help="Instagram image"
+                            "Image", help="Instagram image"
                         ),
                         "URL": st.column_config.LinkColumn(
-                            "link", help="Instagram link", width="small"
+                            "Link", help="Instagram link", width="small"
                         )
                     },
                     hide_index=True,
                 )
     
         elif selected_tab == "Image to Image":
-            image_to_image_k_top = st.slider("Number of results", 1, 20, 8)
+            image_to_image_k_top = st.slider("Number of results", 1, 60, 8)
             image_to_image_input = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
             if st.button("Search"):
                 st.dataframe(
                     data=image_to_image(image_to_image_input, image_to_image_k_top),
                     column_config={
                         "image_base64": st.column_config.ImageColumn(
-                            "image", help="Instagram image"
+                            "Image", help="Instagram image"
                         ),
                         "URL": st.column_config.LinkColumn(
-                            "link", help="Instagram link", width="small"
+                            "Link", help="Instagram link", width="small"
                         )
                     },
                     hide_index=True,
                 )
     
         elif selected_tab == "Image to Text":
-            image_to_text_k_top = st.slider("Number of results", 1, 20, 8)
+            image_to_text_k_top = st.slider("Number of results", 1, 60, 8)
             image_to_text_input = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
             if st.button("Search"):
                 st.dataframe(
                     data=image_to_text(image_to_text_input, image_to_text_k_top),
                     column_config={
                         "image_base64": st.column_config.ImageColumn(
-                            "image", help="Instagram image"
+                            "Image", help="Instagram image"
                         ),
                         "URL": st.column_config.LinkColumn(
-                            "link", help="Instagram link", width="small"
+                            "Link", help="Instagram link", width="small"
                         )
                     },
                     hide_index=True,
@@ -413,7 +371,7 @@ if check_password():
         # Dropdown to select time resampling
         resample_time = st.selectbox('Select Time Resampling', list(resample_dict.keys()))
     
-        df_filtered = df.set_index('Post Created Date')
+        df_filtered = df.set_index('Post Created')
     
         # Slider for date range selection
         min_date = df_filtered.index.min().date()