Update app.py

app.py

@@ -6,7 +6,7 @@ from sklearn.preprocessing import LabelEncoder, StandardScaler
 from sklearn.model_selection import train_test_split
 import tensorflow as tf
 from tensorflow.keras.models import Sequential, Model
-from tensorflow.keras.layers import LSTM, Dense, Input, MultiHeadAttention, LayerNormalization
+from tensorflow.keras.layers import LSTM, Dense, Input, MultiHeadAttention, LayerNormalization, GlobalAveragePooling1D
 from tensorflow.keras.optimizers import Adam
 import joblib
 import os
@@ -57,6 +57,13 @@ def create_lstm_model(input_shape):
     return model
 
 # Transformer Model
+def create_transformer_model(input_shape):
+    inputs = Input(shape=input_shape)
+    x = transformer_encoder(inputs, head_size=256, num_heads=4, ff_dim=4, dropout=0.1)
+    x = GlobalAveragePooling1D()(x)
+    outputs = Dense(1)(x)
+    return Model(inputs, outputs)
+
 def transformer_encoder(inputs, head_size, num_heads, ff_dim, dropout=0):
     x = MultiHeadAttention(key_dim=head_size, num_heads=num_heads, dropout=dropout)(inputs, inputs)
     x = LayerNormalization(epsilon=1e-6)(x)
@@ -65,13 +72,6 @@ def transformer_encoder(inputs, head_size, num_heads, ff_dim, dropout=0):
     x = Dense(inputs.shape[-1])(x)
     return LayerNormalization(epsilon=1e-6)(x + res)
 
-def create_transformer_model(input_shape):
-    inputs = Input(shape=input_shape)
-    x = transformer_encoder(inputs, head_size=256, num_heads=4, ff_dim=4, dropout=0.1)
-    x = GlobalAveragePooling1D()(x)
-    outputs = Dense(1)(x)
-    return Model(inputs, outputs)
-
 # RL Environment
 class PricingEnv(gym.Env):
     def __init__(self, data):