bring up to date

app.py

@@ -9,6 +9,7 @@ import tensorflow as tf
 import tensorflow_hub as hub
 from sklearn.metrics.pairwise import cosine_similarity
 
+
 # Import logging module
 import logging
 
@@ -87,13 +88,13 @@ def process_images_and_statements(image):
     # Generate image caption for the uploaded image using git-large-r-textcaps
     caption = generate_caption(git_processor_large_textcaps, git_model_large_textcaps, image)
 
-    # Initialize an empty list to store the results
-    results = []
-
     # Define weights for combining textual similarity score and image-statement ITM score (adjust as needed)
     weight_textual_similarity = 0.5
     weight_statement = 0.5
 
+    # Initialize an empty DataFrame with column names
+    results_df = pd.DataFrame(columns=['Statement', 'Textual Similarity Score', 'ITM Score', 'Final Combined Score'])
+
     # Loop through each predefined statement
     for statement in statements:
         # Compute textual similarity between caption and statement
@@ -105,21 +106,30 @@ def process_images_and_statements(image):
         # Combine the two scores using a weighted average
         final_score = (weight_textual_similarity * textual_similarity_score) + (weight_statement * itm_score_statement)
 
-        # Store the result
-        result_text = (f'Textual similarity between caption ("{caption}") and statement ("{statement}") is {textual_similarity_score:.3f}\n'
-                       f'The image-statement pair ("{statement}") is matched with a probability of {itm_score_statement:.3%}\n'
-                       f'The final combined score is {final_score:.3%}')
-        results.append(result_text)
+        # Append the result to the DataFrame
+        results_df = results_df.append({
+            'Statement': statement,
+            'Textual Similarity Score': textual_similarity_score,
+            'ITM Score': itm_score_statement,
+            'Final Combined Score': final_score
+        }, ignore_index=True)
 
     logging.info('Finished process_images_and_statements')
 
-        # Combine the results and return them
-    output = "\n\n".join(results)
-    return output
+    # Return the DataFrame directly as output (no need to convert to HTML)
+    return results_df  # <--- Return results_df directly
 
 # Gradio interface
 image_input = gr.inputs.Image()
-output = gr.outputs.Textbox(label="Results")
+output = gr.outputs.Dataframe(type="pandas", label="Results")  # <--- Use "pandas" type for DataFrame output
+
+iface = gr.Interface(
+    fn=process_images_and_statements,
+    inputs=image_input,
+    outputs=output,
+    title="Image Captioning and Image-Text Matching",
+    theme='freddyaboulton/dracula_revamped',
+    css=".output { flex-direction: column; } .output .outputs { width: 100%; }"  # Custom CSS
+)
 
-iface = gr.Interface(fn=process_images_and_statements, inputs=image_input, outputs=output, title="Image Captioning and Image-Text Matching")
 iface.launch()