Initial setup for training

neuralnet/neuralnetwork.py

@@ -1,26 +1,40 @@
+import os
 import numpy as np
-from tensorflow.keras.models import Sequential
-from tensorflow.keras.layers import Dense
+import tensorflow as tf
 
 # create NeuralNetwork class
 class NeuralNetwork:
-    def __init__(self, inputs_len: int):
+    def __init__(self, input_length: int):
+        self.model = tf.keras.Sequential([
+            tf.keras.Input(shape=(input_length,), dtype=tf.int64),
+            tf.keras.layers.Dense(512, activation='relu'),
+            tf.keras.layers.Dropout(0.5),
+            tf.keras.layers.Dense(256, activation='relu'),
+            tf.keras.layers.Dropout(0.5),
+            tf.keras.layers.Dense(128, activation='relu'),
+            tf.keras.layers.Dense(1, activation='sigmoid'),
+        ])
+
+        self.model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
+
         # Setup checkpoint
-        self.checkpoint_path = "./training/cp.ckpt"
-        self.cp_callback = tf.keras.callbacks.ModelCheckpoint(filepath=checkpoint_path, save_weights_only=True, verbose=1)
+        self.checkpoint_path = "./training/cp.ckpt.weights.h5"
+        self.cp_callback = tf.keras.callbacks.ModelCheckpoint(
+            filepath=self.checkpoint_path, 
+            save_weights_only=True, 
+            save_best_only=True,
+            monitor='loss',
+            mode='min',
+        )
 
-        # Setup model
-        self.model = Sequential()
-        self.model.add(Dense(12, input_shape=(inputs_len,), activation='relu'))
-        self.model.add(Dense(8, activation='relu'))
-        self.model.add(Dense(1, activation='sigmoid'))
-        self.model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
+        #if os.path.isfile(self.checkpoint_path):
+        #    self.model.load_weights(self.checkpoint_path)
 
-        if os.path.isfile(self.checkpoint_path):
-            self.model.load_weights(self.checkpoint_path)
+    def train(self, inputs :list, outputs :list):
+        self.model.fit(inputs, outputs, epochs=100, batch_size=64, callbacks=[self.cp_callback])
 
-    def train(inputs :list, outputs :list):
-        self.model.fit(inputs, outputs, epochs=150, batch_size=10, callbacks=[self.cp_callback])
+    def summary(self):
+        print(self.model.summary())
 
     def predict(self, new_input):
         return self.model.predict(new_input)