Update app.py

app.py

@@ -1,43 +1,51 @@
 import numpy as np
-import pandas as pd 
-import tensorflow as tf 
+import tensorflow as tf
 import streamlit as st
-
 from PIL import Image
 
-model = tf.keras.models.load_model(r"C:\Dhruv\potato diseases\my_model.h5")
+# Custom deserialization function for SparseCategoricalCrossentropy
+def custom_sparse_categorical_crossentropy_from_config(config):
+    # Remove the 'fn' key from config if it exists
+    config.pop('fn', None)
+    return tf.keras.losses.SparseCategoricalCrossentropy(**config)
+
+# Load the model with custom objects
+custom_objects = {'SparseCategoricalCrossentropy': custom_sparse_categorical_crossentropy_from_config}
 
+model_path = r"C:\Dhruv\potato diseases\my_model.h5"
+model = tf.keras.models.load_model(model_path, custom_objects=custom_objects)
 
+# Image preprocessing function
 resize_and_rescale = tf.keras.Sequential([
-    tf.keras.layers.experimental.preprocessing.Resizing(256,256),
-    tf.keras.layers.experimental.preprocessing.Rescaling(1.0/255)
+    tf.keras.layers.Resizing(256, 256),
+    tf.keras.layers.Rescaling(1.0 / 255)
 ])
 
 def predict(model, img):
-
     class_names = ['Potato___Early_blight', 'Potato___Late_blight', 'Potato___healthy']
-
+    
+    # Convert the image to array
     img_array = tf.keras.preprocessing.image.img_to_array(np.array(img))
-    img_array = tf.expand_dims(img_array, 0)
+    img_array = tf.expand_dims(img_array, 0)  # Create batch axis
 
+    # Preprocess the image
     img_array = resize_and_rescale(img_array)
 
+    # Make prediction
     predictions = model.predict(img_array)
-
     predicted_class = class_names[np.argmax(predictions[0])]
-    confidence = round(100 * (np.max(predictions[0])), 2)
+    confidence = round(100 * np.max(predictions[0]), 2)
     return predicted_class, confidence
 
+# Streamlit app
 st.title("Potato Disease Classification")
 
 uploaded_image = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
 
-if uploaded_image:
-
-    st.image(uploaded_image)
+if uploaded_image is not None:
+    st.image(uploaded_image, caption='Uploaded Image', use_column_width=True)
     img = Image.open(uploaded_image)
     
     predicted_class, confidence = predict(model=model, img=img)
-
+    
     st.markdown(f"The predicted class is **{predicted_class}** with **{confidence}%** confidence")
-