Work on optimizing the model

WebServer/AIPython/ai-trainer.py

@@ -5,7 +5,7 @@ import datapuller
 import os
 os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
 from sklearn.model_selection import train_test_split
-from keras import Sequential, layers, optimizers
+from keras import Sequential, layers, optimizers, losses
 
 def TrainAI():
     # Pull New Data
@@ -17,8 +17,9 @@ def TrainAI():
 
     # Load the dataset
     dataset = pd.read_parquet(os.path.join(DATA_DIR, "stocks.parquet"))
-    X = dataset.drop('Target_Close_Tomorrow', axis=1)
-    Y = dataset['Target_Close_Tomorrow']
+    X = dataset.drop('Target_Close', axis=1)
+    X = dataset.drop('Target_Direction', axis=1)
+    Y = dataset['Target_Close']
 
     # Show the datatypes
     print(dataset.dtypes)
@@ -34,15 +35,15 @@ def TrainAI():
     dnn_linear_model = Sequential([
         layers.Input(shape=(train_features.shape[1],)), # Load the feature count dynamically
         normalizer,
-        layers.Dense(64, activation='relu'),
-        layers.Dense(64, activation='relu'),
+        layers.Dense(64, activation='elu'),
+        layers.Dense(64, activation='elu'),
         layers.Dense(1)
     ])
 
     # Configure the model
     dnn_linear_model.compile(
-        optimizer=optimizers.Adam(learning_rate=0.001),
-        loss='mean_absolute_error'
+        optimizer=optimizers.Adam(learning_rate=0.0001, clipvalue=1.0),
+        loss=losses.Huber()
     )
 
     # Show the summary before training the model
@@ -52,7 +53,7 @@ def TrainAI():
     Training_Data = dnn_linear_model.fit( 
         train_features,
         train_labels,
-        batch_size=1024,
+        batch_size=64,
         epochs=100,
         # Show progress
         verbose=1,
@@ -76,5 +77,7 @@ def TrainAI():
         test_features, test_labels, verbose=0
     )
 
+    print(f"Test Results: {test_results}")
+
     # Save the model
     dnn_linear_model.save(os.path.join(DATA_DIR, "model.keras"))