Update app.py

app.py

@@ -3,33 +3,27 @@ from PIL import Image
 import tensorflow as tf
 import numpy as np
 import os
-from tensorflow.keras.layers import LSTM
-from keras.saving import register_keras_serializable
-
-# Kelas LSTM Kustom
-@register_keras_serializable()
-class CustomLSTM(LSTM):
-    pass
 
 # Caching the model loading function to optimize performance
 @st.cache_resource
-def load_captcha_model():
-    model_path = "model1.keras"  # Update with the actual CAPTCHA model path
-    return tf.keras.models.load_model(model_path, custom_objects={'CustomLSTM': CustomLSTM})
+def load_model():
+    model_path = "captcha.keras"  # Update with the actual model path
+    return tf.keras.models.load_model(model_path)
 
 # Load the model
-model = load_captcha_model()
+model = load_model()
 
 # Function to prepare the image for model prediction
-def prepare_captcha_image(img):
+def prepare_image(img):
     try:
-        # Resize image to the input shape required by the CAPTCHA model
+        # Resize image to the input shape required by the model
         img = img.resize((200, 50))  # Adjust size according to the trained model
         img_array = np.array(img.convert('L'))  # Convert to grayscale if necessary
         img_array = img_array / 255.0  # Normalize image
+        img_array = np.expand_dims(img_array, axis=-1)  # Add channel dimension
         img_array = np.expand_dims(img_array, axis=0)  # Add batch dimension
 
-        # Predict the CAPTCHA characters
+        # Predict the output using the loaded model
         predictions = model.predict(img_array)
 
         # Decode predictions assuming the model outputs probabilities
@@ -60,7 +54,7 @@ def run():
             f.write(img_file.getbuffer())
 
         # Predict the CAPTCHA
-        predicted_captcha, score = prepare_captcha_image(img)
+        predicted_captcha, score = prepare_image(img)
         if predicted_captcha:
             st.success(f"**Predicted CAPTCHA: {predicted_captcha}**")
         else: