Get tensorflow DNN layout

WebServer/AIPython/main.py

@@ -1,20 +1,72 @@
-import tensorflow as tf
+import pandas as pd
-import keras
+import numpy as np
-from keras.layers import Dense, Flatten, Conv2D
+import matplotlib.pyplot as plt
-from keras import Model
+import os
+from sklearn.model_selection import train_test_split
+from keras import Sequential, layers, optimizers
 
-def main():
+# Get the CWD for pathing due to being called from C# now
-    mnist = keras.datasets.mnist
+SCRIPT_DIR = os.path.dirname(os.path.abspath(__file__))
+DATA_DIR = os.path.join(SCRIPT_DIR, "data")
 
-    (x_train, y_train), (x_test, y_test) = mnist.load_data()
+# Load the dataset
-    x_train, x_test = x_train / 255.0, x_test / 255.0
+dataset = pd.read_parquet(os.path.join(DATA_DIR, "stocks.parquet"))
+X = dataset.drop('Volatility_5', axis=1)
+Y = dataset['Volatility_5']
 
-    # Add a channels dimension
+# Show the datatypes
-    x_train = x_train[..., tf.newaxis].astype("float32")
+print(dataset.dtypes)
-    x_test = x_test[..., tf.newaxis].astype("float32")
 
-    # batch and shuffle the dataset
+# Split out the test and train
-    train_ds = tf.data.Dataset.from_tensor_slices(
+train_features, test_features, train_labels, test_labels = train_test_split(X, Y, test_size=0.2)
-        (x_train, y_train)).shuffle(10000).batch(32)
 
-    test_ds = tf.data.Dataset.from_tensor_slices((x_test, y_test)).batch(32)
+# Create a normalizer to nomralize the data
+normalizer = layers.Normalization(axis=-1)
+normalizer.adapt(np.array(train_features))
+
+# Start with a linear model
+dnn_linear_model = Sequential([
+    layers.Input(shape=(9,)), # Explicitly tell Keras there are 9 features
+    normalizer,
+    layers.Dense(64, activation='relu'),
+    layers.Dense(64, activation='relu'),
+    layers.Dense(1)
+])
+
+# Configure the model
+dnn_linear_model.compile(
+    optimizer=optimizers.Adam(learning_rate=0.1),
+    loss='mean_absolute_error'
+)
+
+dnn_linear_model.summary()
+
+# Train the model
+Training_Data = dnn_linear_model.fit( 
+    train_features,
+    train_labels,
+    epochs=100,
+    # Show progress
+    verbose=1,
+    # Calculate validation results on 20% of the training data.
+    validation_split = 0.2
+)
+
+# Predict
+test_predictions = dnn_linear_model.predict(test_features).flatten()
+a = plt.axes(aspect='equal')
+plt.scatter(test_labels, test_predictions)
+plt.xlabel('True Values')
+plt.ylabel('Predictions')
+lims = [0, 50]
+plt.xlim(lims)
+plt.ylim(lims)
+_ = plt.plot(lims, lims)
+
+
+test_results = dnn_linear_model.evaluate(
+    test_features, test_labels, verbose=0
+)
+
+# Save the model
+dnn_linear_model.save(os.path.join(DATA_DIR, "model.keras"))