Merge remote-tracking branch 'origin' into testing

mnk.py

@@ -78,6 +78,9 @@ def legal_moves(self):
                     moves.append((x, y))
         return moves
 
+    def num_legal_moves(self):
+        return len(self.legal_moves())
+
     # Reshapes board into the form needed for the model
     def get_board(self):
         if self.form == "flatten":
@@ -128,6 +131,9 @@ def __str__(self):
                     string += '\n'
         return string
 
+    def game_ongoing(self):
+        return not ( self.player_has_lost() or (self.num_legal_moves() == 0) )
+
     # returns True if the player whose turn it is has lost, False otherwise
     def player_has_lost(self):
         # check vertical line |

models/Pedro3HiddenLayersTanh50kGamesInterrupted/keras_metadata.pb

@@ -0,0 +1,7 @@
+
+�-root"_tf_keras_sequential*�-{"name": "sequential", "trainable": true, "expects_training_arg": true, "dtype": "float32", "batch_input_shape": null, "must_restore_from_config": false, "class_name": "Sequential", "config": {"name": "sequential", "layers": [{"class_name": "InputLayer", "config": {"batch_input_shape": {"class_name": "__tuple__", "items": [null, 1, 9]}, "dtype": "float32", "sparse": false, "ragged": false, "name": "dense_input"}}, {"class_name": "Dense", "config": {"name": "dense", "trainable": true, "batch_input_shape": {"class_name": "__tuple__", "items": [null, 1, 9]}, "dtype": "float32", "units": 27, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "Dense", "config": {"name": "dense_1", "trainable": true, "dtype": "float32", "units": 18, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "Dense", "config": {"name": "dense_2", "trainable": true, "dtype": "float32", "units": 9, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "Dense", "config": {"name": "dense_3", "trainable": true, "dtype": "float32", "units": 1, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}]}, "shared_object_id": 13, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": null, "max_ndim": null, "min_ndim": 2, "axes": {"-1": 9}}, "shared_object_id": 14}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 1, 9]}, "is_graph_network": true, "full_save_spec": {"class_name": "__tuple__", "items": [[{"class_name": "TypeSpec", "type_spec": "tf.TensorSpec", "serialized": [{"class_name": "TensorShape", "items": [null, 1, 9]}, "float32", "dense_input"]}], {}]}, "save_spec": {"class_name": "TypeSpec", "type_spec": "tf.TensorSpec", "serialized": [{"class_name": "TensorShape", "items": [null, 1, 9]}, "float32", "dense_input"]}, "keras_version": "2.6.0", "backend": "tensorflow", "model_config": {"class_name": "Sequential", "config": {"name": "sequential", "layers": [{"class_name": "InputLayer", "config": {"batch_input_shape": {"class_name": "__tuple__", "items": [null, 1, 9]}, "dtype": "float32", "sparse": false, "ragged": false, "name": "dense_input"}, "shared_object_id": 0}, {"class_name": "Dense", "config": {"name": "dense", "trainable": true, "batch_input_shape": {"class_name": "__tuple__", "items": [null, 1, 9]}, "dtype": "float32", "units": 27, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}, "shared_object_id": 1}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 2}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 3}, {"class_name": "Dense", "config": {"name": "dense_1", "trainable": true, "dtype": "float32", "units": 18, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}, "shared_object_id": 4}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 5}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 6}, {"class_name": "Dense", "config": {"name": "dense_2", "trainable": true, "dtype": "float32", "units": 9, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}, "shared_object_id": 7}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 8}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 9}, {"class_name": "Dense", "config": {"name": "dense_3", "trainable": true, "dtype": "float32", "units": 1, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}, "shared_object_id": 10}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 11}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 12}]}}, "training_config": {"loss": "mean_squared_error", "metrics": null, "weighted_metrics": null, "loss_weights": null, "optimizer_config": {"class_name": "Adam", "config": {"name": "Adam", "learning_rate": 0.10000000149011612, "decay": 0.0, "beta_1": 0.8999999761581421, "beta_2": 0.9990000128746033, "epsilon": 1e-07, "amsgrad": false}}}}2

+�
root.layer_with_weights-0"_tf_keras_layer*�{"name": "dense", "trainable": true, "expects_training_arg": false, "dtype": "float32", "batch_input_shape": {"class_name": "__tuple__", "items": [null, 1, 9]}, "stateful": false, "must_restore_from_config": false, "class_name": "Dense", "config": {"name": "dense", "trainable": true, "batch_input_shape": {"class_name": "__tuple__", "items": [null, 1, 9]}, "dtype": "float32", "units": 27, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}, "shared_object_id": 1}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 2}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 3, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": null, "max_ndim": null, "min_ndim": 2, "axes": {"-1": 9}}, "shared_object_id": 14}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 1, 9]}}2

+�root.layer_with_weights-1"_tf_keras_layer*�{"name": "dense_1", "trainable": true, "expects_training_arg": false, "dtype": "float32", "batch_input_shape": null, "stateful": false, "must_restore_from_config": false, "class_name": "Dense", "config": {"name": "dense_1", "trainable": true, "dtype": "float32", "units": 18, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}, "shared_object_id": 4}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 5}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 6, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": null, "max_ndim": null, "min_ndim": 2, "axes": {"-1": 27}}, "shared_object_id": 15}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 1, 27]}}2

+�root.layer_with_weights-2"_tf_keras_layer*�{"name": "dense_2", "trainable": true, "expects_training_arg": false, "dtype": "float32", "batch_input_shape": null, "stateful": false, "must_restore_from_config": false, "class_name": "Dense", "config": {"name": "dense_2", "trainable": true, "dtype": "float32", "units": 9, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}, "shared_object_id": 7}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 8}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 9, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": null, "max_ndim": null, "min_ndim": 2, "axes": {"-1": 18}}, "shared_object_id": 16}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 1, 18]}}2

+�root.layer_with_weights-3"_tf_keras_layer*�{"name": "dense_3", "trainable": true, "expects_training_arg": false, "dtype": "float32", "batch_input_shape": null, "stateful": false, "must_restore_from_config": false, "class_name": "Dense", "config": {"name": "dense_3", "trainable": true, "dtype": "float32", "units": 1, "activation": "tanh", "use_bias": true, "kernel_initializer": {"class_name": "RandomNormal", "config": {"mean": 0.0, "stddev": 0.05, "seed": null}, "shared_object_id": 10}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 11}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 12, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": null, "max_ndim": null, "min_ndim": 2, "axes": {"-1": 9}}, "shared_object_id": 17}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 1, 9]}}2

+�
Aroot.keras_api.metrics.0"_tf_keras_metric*�
{"class_name": "Mean", "name": "loss", "dtype": "float32", "config": {"name": "loss", "dtype": "float32"}, "shared_object_id": 18}2

play.py

@@ -1,15 +1,18 @@
 from agent import Agent
 import mnk
 import tensorflow as tf
-from model import Model
+import model
+import sys
 
-board = mnk.Board(3, 3, 3, form="multiplanar-2")
-model = tf.keras.models.load_model('models/modelXO')
+board = mnk.Board(3, 3, 3, form="flatten")
+
+assert len(sys.argv) == 2, "Please specify which model you would like to play against (ex: python3 play.py 3LayersModel)"
+model = model.Model('models/' + sys.argv[1])
 
 print("\n\n" + str(board))
 current_player = input("\nWho plays first (Me/AI)? ")
 ai_side = [-1, 1][current_player == "AI"]
-agent = Agent(Model("models/modelXO"), ai_side)
+agent = Agent(model, ai_side)
 
 while board.who_won() == 2:
     if current_player == 'Me':

plot.py

@@ -1,10 +1,10 @@
-from matplotlib import pyplot as plt
-
+import matplotlib.pyplot as plt
 
 def plot_wins(game_outcomes, run_length, labels=['X', 'O']):
     xs = []
     os = []
     ties = []
+    num_games = len(game_outcomes)
 
     values = [0, 0, 0]
 
@@ -24,8 +24,6 @@ def plot_wins(game_outcomes, run_length, labels=['X', 'O']):
     plt.plot(game, os, label="{} wins".format(labels[1]))
     plt.plot(game, ties, label="Ties")
     plt.legend()
-    plt.title("Number of Each End State for Previous {} Games".format(run_length))
+    plt.title("Training data for {} Games".format(num_games))
     plt.xlabel("Game number")
     plt.ylabel("Wins out of previous {} games".format(run_length))
-
-

train.py

@@ -1,7 +1,7 @@
 # TODO: PLOT LOSS CURVES
 from mnk import Board
 import random
-from matplotlib import pyplot as plt
+import matplotlib.pyplot as plt
 from agent import Agent
 from model import Model
 from plot import plot_wins