Update app.py

app.py

@@ -1,17 +1,66 @@
 import numpy as np
 import tensorflow as tf
 from tensorflow.keras.utils import to_categorical
+from tensorflow.keras.layers import Layer, Dense, Dropout, LayerNormalization, MultiHeadAttention
+from tensorflow.keras.models import Sequential
 from sklearn.preprocessing import LabelEncoder
 import gradio as gr
-
-# Load the saved model
-model = tf.keras.models.load_model("enhanced_adaptive_model.keras", custom_objects={'EnhancedTransformerBlock': EnhancedTransformerBlock})
+import os
+
+# Define the EnhancedTransformerBlock class
+class EnhancedTransformerBlock(Layer):
+    def __init__(self, embed_dim, num_heads, ff_dim, rate=0.1, **kwargs):
+        super(EnhancedTransformerBlock, self).__init__(**kwargs)
+        self.embed_dim = embed_dim
+        self.num_heads = num_heads
+        self.ff_dim = ff_dim
+        self.rate = rate
+        
+        self.att = MultiHeadAttention(num_heads=num_heads, key_dim=embed_dim)
+        self.ffn = Sequential([
+            Dense(ff_dim, activation="relu"),
+            Dense(embed_dim),
+        ])
+        self.layernorm1 = LayerNormalization(epsilon=1e-6)
+        self.layernorm2 = LayerNormalization(epsilon=1e-6)
+        self.dropout1 = Dropout(rate)
+        self.dropout2 = Dropout(rate)
+        self.self_attention = MultiHeadAttention(num_heads=1, key_dim=embed_dim)
+
+    def call(self, inputs, training=False):
+        attn_output = self.att(inputs, inputs)
+        attn_output = self.dropout1(attn_output, training=training)
+        out1 = self.layernorm1(inputs + attn_output)
+        ffn_output = self.ffn(out1)
+        ffn_output = self.dropout2(ffn_output, training=training)
+        out2 = self.layernorm2(out1 + ffn_output)
+        self_attn_output = self.self_attention(out2, out2)
+        return self.layernorm2(out2 + self_attn_output)
+
+    def get_config(self):
+        config = super().get_config()
+        config.update({
+            "embed_dim": self.embed_dim,
+            "num_heads": self.num_heads,
+            "ff_dim": self.ff_dim,
+            "rate": self.rate,
+        })
+        return config
 
 # Initialize global variables
 sequence_length = 10
 data = []  # This will store the recent outcomes
 encoder = LabelEncoder()
-encoder.classes_ = np.load('label_encoder_classes.npy', allow_pickle=True)
+
+# Try to load the saved model and encoder classes
+try:
+    model = tf.keras.models.load_model("enhanced_adaptive_model.keras", custom_objects={'EnhancedTransformerBlock': EnhancedTransformerBlock})
+    encoder.classes_ = np.load('label_encoder_classes.npy', allow_pickle=True)
+except FileNotFoundError:
+    print("Model or encoder classes file not found. Using a dummy model and encoder for demonstration.")
+    # Create a dummy model and encoder for demonstration
+    model = tf.keras.Sequential([tf.keras.layers.Dense(10, input_shape=(sequence_length,), activation='softmax')])
+    encoder.classes_ = np.array(['Single small', 'Single big', 'Double small', 'Double big', 'Triple'])
 
 def update_data(data, new_outcome):
     data.append(new_outcome)