Update app.py

app.py

@@ -1,13 +1,13 @@
 # Import the libraries
 import numpy as np
 import pandas as pd
-from tensorflow.keras.layers import TFSMLayer  # type: ignore
-from tensorflow.keras.preprocessing.image import load_img, img_to_array  # type: ignore
-from tensorflow.keras.applications.convnext import preprocess_input  # type: ignore
+from tensorflow.keras.models import load_model # type: ignore
+from tensorflow.keras.preprocessing.image import load_img, img_to_array # type: ignore
+from tensorflow.keras.applications.convnext import preprocess_input # type: ignore
 import gradio as gr
 
 # Load the model
-model = TFSMLayer('models/ConvNeXtBase_80_tresh_spp.tf', call_endpoint='serving_default')
+model = load_model('models/ConvNeXtBase_80_tresh_spp.tf', call_endpoint='serving_default')
 
 # Load the taxonomy .csv
 taxo_df = pd.read_csv('taxonomy/taxonomy_mapping.csv', sep=';')
@@ -26,7 +26,7 @@ def load_and_preprocess_image(image, target_size=(224, 224)):
     img_array = img_to_array(image.resize(target_size))
     # Expand the dimensions of the array to match model input
     img_array = np.expand_dims(img_array, axis=0)
-    # Preprocess using the appropriate function (for example, ConvNeXt)
+    # Preprocess using the appropriate function (for example, ResNet50)
     img_array = preprocess_input(img_array)
     return img_array
 
@@ -34,17 +34,17 @@ def load_and_preprocess_image(image, target_size=(224, 224)):
 def make_prediction(image):
     # Preprocess the image
     img_array = load_and_preprocess_image(image)
-    # Make a prediction by calling the model directly
-    prediction = model(img_array)  # Call the model instead of using predict()
+    # Make a prediction
+    prediction = model.predict(img_array)
     
     # Get the top 5 predictions
     top_indices = np.argsort(prediction[0])[-5:][::-1]  # Get indices of top 5 classes
     
     # Get predicted class and common name for the top prediction
-    predicted_class_index = np.argmax(prediction.numpy())  # Convert to numpy before processing
+    predicted_class_index = np.argmax(prediction)
     predicted_class_name = get_class_name(predicted_class_index)
     predicted_common_name = taxo_df[taxo_df['species'] == predicted_class_name]['common_name'].values[0]  # Get common name
-    confidence = prediction[0][predicted_class_index].numpy() * 100  # Confidence of the predicted class
+    confidence = prediction[0][predicted_class_index] * 100  # Confidence of the predicted class
     
     # Create output text with HTML formatting
     output_text = f"<h1 style='font-weight: bold;'><span style='font-style: italic;'>{predicted_class_name}</span> ({predicted_common_name})</h1>"  # Large bold for predicted class, italic for class name
@@ -53,7 +53,7 @@ def make_prediction(image):
     for i in top_indices:
         class_name = get_class_name(i)
         common_name = taxo_df[taxo_df['species'] == class_name]['common_name'].values[0]  # Get common name from CSV
-        confidence_percentage = prediction[0][i].numpy() * 100  # Convert tensor to numpy for indexing
+        confidence_percentage = prediction[0][i] * 100
         
         # Format the output with space between class name and common name
         output_text += f"<div style='display: flex; justify-content: space-between;'>" \