Update app.py

app.py

@@ -1,9 +1,8 @@
-import json
+from PIL import Image
+import tensorflow as tf
 import numpy as np
 import gradio as gr
-import tensorflow as tf
-from PIL import Image
-from tensorflow.keras.models import load_model
+import io
 
 # Load the model
 model_path = 'final_teath_classifier.h5'
@@ -12,7 +11,6 @@ model = tf.keras.models.load_model(model_path)
 # Define preprocessing function
 def preprocess_image(image):
     # Resize the image to match input size
-    image = Image.fromarray(image)
     image = image.resize((256, 256))
     # Convert image to array and preprocess input
     img_array = np.array(image) / 255.0
@@ -22,6 +20,14 @@ def preprocess_image(image):
 
 # Define prediction function
 def predict_image(image):
+    # Save the image to a file-like object
+    image_bytes = io.BytesIO()
+    image.save(image_bytes, format="jpg")
+    image_bytes.seek(0)  # Reset file pointer to start
+
+    # Load the image from the file-like object
+    image = tf.keras.preprocessing.image.load_img(image_bytes, target_size=(256, 256))
+
     img_array = preprocess_image(image)
     outputs = model(img_array)
     predictions = tf.nn.softmax(outputs.logits, axis=-1)
@@ -32,16 +38,16 @@ def predict_image(image):
         predict_label = "Carries"
     confidence = float(np.max(predictions))
     prediction_dict = {"prediction": predict_label, "confidence": confidence}
-    return prediction_dict
+    return json.dumps(prediction_dict, indent=2)
 
 # Create the interface
-input_interface = gr.Image()
-output_interface = gr.JSON()
+input_interface = gr.inputs.Image(shape=(256, 256), image_mode='RGB')
+output_interface = "json"
 
 iface = gr.Interface(
     fn=predict_image,
     inputs=input_interface,
-    outputs=output_interface)
+    outputs=gr.Textbox("output"))
 
 # Launch the interface
 iface.launch(share=True)