Update app.py

app.py

@@ -1,3 +1,5 @@
+# app.py
+
 import gradio as gr
 import tensorflow as tf
 import numpy as np
@@ -7,48 +9,40 @@ from PIL import Image
 model = tf.keras.models.load_model('Adam_8_1000_Acc 0.88_Nutrient-Model.h5')
 
 # Define the class names
-class_names = ['Calcium', 'Magnesium', 'Nitrogen', 'Phosphorus', 'Potassium', 'Sulfur']
-
+class_names = ['Calcium','Magnesium','Nitrogen','Phosphorus','Potassium','Sulfur']
 # Function to classify the image
-def classify_image(images):
-    predictions = []
-    for image in images:
-        # Convert the numpy array to a PIL Image object
-        pil_image = Image.fromarray(np.uint8(image)).convert('RGB')
-
-        # Resize the image
-        pil_image = pil_image.resize((224, 224))
+def classify_image(image):
+    # Convert the numpy array to a PIL Image object
+    pil_image = Image.fromarray(np.uint8(image)).convert('RGB')
 
-        # Convert the PIL Image object to a numpy array
-        image_array = np.array(pil_image)
+    # Resize the image
+    pil_image = pil_image.resize((224, 224))
 
-        # Normalize the image
-        normalized_image_array = (image_array.astype(np.float32) / 255.0)
+    # Convert the PIL Image object to a numpy array
+    image_array = np.array(pil_image)
 
-        # Reshape the image
-        data = normalized_image_array.reshape((1, 224, 224, 3))
+    # Normalize the image
+    normalized_image_array = (image_array.astype(np.float32) / 255.0)
 
-        # Make the prediction
-        prediction = model.predict(data)[0]
+    # Reshape the image
+    data = normalized_image_array.reshape((1, 224, 224, 3))
 
-        # Get the predicted class name
-        predicted_class = class_names[np.argmax(prediction)]
+    # Make the prediction
+    prediction = model.predict(data)[0]
 
-        # Get the confidence score for the predicted class
-        confidence_score = np.max(prediction)
+    # Get the predicted class name
+    predicted_class = class_names[np.argmax(prediction)]
 
-        # Append the predicted class and confidence score to the predictions list
-        predictions.append(f"{predicted_class} ({confidence_score*100:.2f}%)")
+    # Get the confidence score for the predicted class
+    confidence_score = np.max(prediction)
 
-    # Return the list of predictions
-    return predictions
+    # Return the predicted class and confidence score
+    return f"{predicted_class} ({confidence_score*100:.2f}%)"
 
 # Define the Gradio interface
-inputs = gr.inputs.Image(type='numpy', label="Upload or Select Images", multiple=True)
+inputs = gr.inputs.Image()
 outputs = gr.outputs.Textbox()
-interface = gr.Interface(fn=classify_image, inputs=inputs, outputs=outputs,
-                        title="Image Classification",
-                        description="Classify multiple images into one of six classes: Phosphorus, Magnesium, Nitrogen, Potassium, Calcium, Sulfur.")
+interface = gr.Interface(fn=classify_image, inputs=inputs, outputs=outputs, title="Image Classification", description="Classify an image into one of six classes: Phosphorus, Magnesium, Nitrogen,Potassium, Calcium, Sulfur.")
 
 # Launch the interface
 interface.launch()