Update app.py

app.py

@@ -30,19 +30,10 @@ def aggregate_predictions(predicted_probs, taxonomic_level, class_names):
         species = row['species']
         higher_level = row[taxonomic_level]
         
-        if species in class_names:  # Check if species exists in class names
-            species_index = class_names.index(species)  # Index of the species in the prediction array
-
-            if higher_level in unique_labels:  # Check if higher level exists
-                higher_level_index = unique_labels.index(higher_level)
-                
-                # Only update if indices are valid
-                if species_index < predicted_probs.shape[1] and higher_level_index < aggregated_predictions.shape[1]:
-                    if predicted_probs[:, species_index].max() >= 0.80:  # Check confidence level
-                        aggregated_predictions[:, higher_level_index] += predicted_probs[:, species_index]
-                    else:
-                        # Stop aggregation at the current level if confidence is below 0.80
-                        break
+        species_index = class_names.index(species)  # Index of the species in the prediction array
+        higher_level_index = unique_labels.index(higher_level)
+        
+        aggregated_predictions[:, higher_level_index] += predicted_probs[:, species_index]
     
     return aggregated_predictions, unique_labels
 
@@ -57,51 +48,26 @@ def load_and_preprocess_image(image, target_size=(224, 224)):
     return img_array
 
 # Function to make predictions
-def make_prediction(image, taxonomic_resolution_choice, taxonomic_level):
+def make_prediction(image, taxonomic_level):
     # Preprocess the image
     img_array = load_and_preprocess_image(image)
-
+    
     # Get the class names from the 'species' column
-    class_names = sorted(taxo_df['species'].unique())
-
+    class_names = sorted(taxo_df['species'].unique())  # Add this line to define class_names
+    
     # Make a prediction
     prediction = model.predict(img_array)
-
-    # Initialize the current taxonomic level index based on the user selection
-    current_taxonomic_level_index = taxonomic_levels.index(taxonomic_level)
-
-    # If the user chose to let the model decide, check the confidence levels
-    if taxonomic_resolution_choice == "No, I will let the model decide":
-        aggregated_predictions = prediction
-        while current_taxonomic_level_index < len(taxonomic_levels):
-            # Aggregate predictions based on the current taxonomic level
-            aggregated_predictions, aggregated_class_labels = aggregate_predictions(
-                aggregated_predictions, taxonomic_levels[current_taxonomic_level_index], class_names
-            )
-
-            # Check if the max confidence in the aggregated predictions is >= 0.80
-            if np.max(aggregated_predictions) >= 0.80:
-                break
-
-            # Move to the next higher taxonomic level
-            current_taxonomic_level_index += 1
-
-        # Ensure we don't go out of bounds
-        if current_taxonomic_level_index < len(taxonomic_levels):
-            taxonomic_level = taxonomic_levels[current_taxonomic_level_index]
-        else:
-            taxonomic_level = taxonomic_levels[-1]  # fallback to the highest taxonomic level
-    else:
-        # Aggregate predictions based on the selected taxonomic level
-        aggregated_predictions, aggregated_class_labels = aggregate_predictions(prediction, taxonomic_level, class_names)
-
+    
+    # Aggregate predictions based on the selected taxonomic level
+    aggregated_predictions, aggregated_class_labels = aggregate_predictions(prediction, taxonomic_level, class_names)
+    
     # Get the top 5 predictions
     top_indices = np.argsort(aggregated_predictions[0])[-5:][::-1]
-
+    
     # Get predicted class for the top prediction
     predicted_class_index = np.argmax(aggregated_predictions)
     predicted_class_name = aggregated_class_labels[predicted_class_index]
-
+    
     # Check if common name should be displayed (only at species level)
     if taxonomic_level == "species":
         predicted_common_name = taxo_df[taxo_df[taxonomic_level] == predicted_class_name]['common_name'].values[0]
@@ -111,10 +77,10 @@ def make_prediction(image, taxonomic_resolution_choice, taxonomic_level):
 
     # Add the top 5 predictions
     output_text += "<h4 style='font-weight: bold; font-size: 1.2em;'>Top 5 Predictions:</h4>"
-
+    
     for i in top_indices:
         class_name = aggregated_class_labels[i]
-
+        
         if taxonomic_level == "species":
             # Display common names only at species level and make it italic
             common_name = taxo_df[taxo_df[taxonomic_level] == class_name]['common_name'].values[0]
@@ -128,7 +94,7 @@ def make_prediction(image, taxonomic_resolution_choice, taxonomic_level):
             output_text += f"<div style='display: flex; justify-content: space-between;'>" \
                            f"<span>{class_name}</span>" \
                            f"<span style='margin-left: auto;'>{confidence_percentage:.2f}%</span></div>"
-
+    
     return output_text
 
 # Define the Gradio interface
@@ -143,7 +109,7 @@ interface = gr.Interface(
     ],
     outputs="html",  # Output type: HTML for formatting
     title="Amazon arboreal species classification",
-    description="Upload an image and select the taxonomic level to classify the species."
+    description="Upload an image and select the taxonomic level (optional) to classify the species."
 )
 
 # Launch the Gradio interface with authentication for the specified users