def create_dataset(path, player=rdeep.Bot(), games=2000, phase=1): data = [] target = [] # For progress bar bar_length = 30 start = time.time() for g in range(games - 1): # For progress bar if g % 10 == 0: percent = 100.0 * g / games sys.stdout.write('\r') sys.stdout.write("Generating dataset: [{:{}}] {:>3}%".format( '=' * int(percent / (100.0 / bar_length)), bar_length, int(percent))) sys.stdout.flush() # Randomly generate a state object starting in specified phase. state = State.generate(phase=phase) state_vectors = [] while not state.finished(): # Give the state a signature if in phase 1, obscuring information that a player shouldn't see. given_state = state.clone(signature=state.whose_turn() ) if state.get_phase() == 1 else state # Add the features representation of a state to the state_vectors array state_vectors.append(features(given_state)) # Advance to the next state move = player.get_move(given_state) state = state.next(move) winner, score = state.winner() for state_vector in state_vectors: data.append(state_vector) if winner == 1: result = 'won' elif winner == 2: result = 'lost' target.append(result) with open(path, 'wb') as output: pickle.dump((data, target), output, pickle.HIGHEST_PROTOCOL) # For printing newline after progress bar print( "\nDone. Time to generate dataset: {:.2f} seconds".format(time.time() - start)) return data, target
def create_dataset(path, player=rdeep.Bot(), good_games=0, bad_games=5000, phase=1): # change good/bad bot numbers """Create a dataset that can be used for training the ML bot model. The dataset is created by having the player (bot) play games against itself. The games parameter indicates how many games will be started. Each game will be played and the game situations will be stored. Then, the game ends and it is recorded whether the game situations resulted in a win or loss for player 1. In other words, each game situation is stored with the corresponding class label (won/lost). Keyword arguments path -- the pathname where the dataset is to be stored player -- the player which will play against itself, default the rand Bot games -- the number of games to play, default 2000 phase -- wheter to start the games in phase 1, the default, or phase 2 """ data = [] target = [] # For progress bar bar_length = 30 start = time.time() for g in range(good_games - 1): # For progress bar if g % 10 == 0: percent = 100.0 * g / good_games sys.stdout.write('\r') sys.stdout.write("Generating good dataset: [{:{}}] {:>3}%".format( '=' * int(percent / (100.0 / bar_length)), bar_length, int(percent))) sys.stdout.flush() # Randomly generate a state object starting in specified phase. state = State.generate(phase=phase) state_vectors = [] while not state.finished(): # Give the state a signature if in phase 1, obscuring information that a player shouldn't see. given_state = state.clone(signature=state.whose_turn() ) if state.get_phase() == 1 else state # Add the features representation of a state to the state_vectors array state_vectors.append(features(given_state)) # Advance to the next state move = player.get_move(given_state) state = state.next(move) winner, score = state.winner() for state_vector in state_vectors: data.append(state_vector) if winner == 1: result = 'won' elif winner == 2: result = 'lost' target.append(result) print(" This is the result %s" % result) for g in range(bad_games - 1): # the added section, entire for loop # For progress bar if g % 10 == 0: percent = 100.0 * g / bad_games sys.stdout.write('\r') sys.stdout.write("Generating bad dataset: [{:{}}] {:>3}%".format( '=' * int(percent / (100.0 / bar_length)), bar_length, int(percent))) sys.stdout.flush() # Randomly generate a state object starting in specified phase. state = State.generate(phase=phase) state_vectors = [] while not state.finished(): # Give the state a signature if in phase 1, obscuring information that a player shouldn't see. given_state = state.clone(signature=state.whose_turn() ) if state.get_phase() == 1 else state # Add the features representation of a state to the state_vectors array state_vectors.append(features(given_state)) # Advance to the next state move = player.get_move(given_state) state = state.next(move) winner, score = state.winner() for state_vector in state_vectors: # TODO: Myléne - make changes here for random labels data.append(state_vector) if winner == 1: result = random.choice(['won', 'lost']) elif winner == 2: result = random.choice(['won', 'lost']) target.append(result) with open(path, 'wb') as output: pickle.dump((data, target), output, pickle.HIGHEST_PROTOCOL) # For printing newline after progress bar print( "\nDone. Time to generate dataset: {:.2f} seconds".format(time.time() - start)) return data, target
dest="overwrite", action="store_true", help= "Whether to create a new dataset regardless of whether one already exists at the specified path." ) parser.add_argument("--no-train", dest="train", action="store_false", help="Don't train a model after generating dataset.") options = parser.parse_args() if options.overwrite or not os.path.isfile(options.dset_path): create_dataset(options.dset_path, player=rdeep.Bot(), good_games=0, bad_games=5000) if options.train: # Play around with the model parameters below # HINT: Use tournament fast mode (-f flag) to quickly test your different models. # The following tuple specifies the number of hidden layers in the neural # network, as well as the number of layers, implicitly through its length. # You can set any number of hidden layers, even just one. Experiment and see what works. hidden_layer_sizes = (64, 32) # The learning rate determines how fast we move towards the optimal solution.
parser.add_argument("-d", dest="create_dataset", action="store_true", help="Whether to create a new dataset regardless of whether one already exists") parser.add_argument("--no-train", dest="train", action="store_false", help="Don't train a model after generating dataset.") options = parser.parse_args() if options.create_dataset or not os.path.isfile(options.path): create_dataset(options.path, player=rdeep.Bot(), games=10000) if options.train: # Play around with the model parameters below # HINT: Use tournament fast mode (-f flag) to quickly test your different models. # The following tuple specifies the number of hidden layers in the neural # network, as well as the number of layers, implicitly through its length. # You can set any number of hidden layers, even just one. Experiment and see what works. hidden_layer_sizes = (64, 32) # The learning rate determines how fast we move towards the optimal solution. # A low learning rate will converge slowly, but a large one might overshoot. learning_rate = 0.0001
parser.add_argument("-o", "--overwrite", dest="overwrite", action="store_true", help="Whether to create a new dataset regardless of whether one already exists at the specified path.") parser.add_argument("--no-train", dest="train", action="store_false", help="Don't train a model after generating dataset.") options = parser.parse_args() if options.overwrite or not os.path.isfile(options.dset_path): create_dataset(options.dset_path, player=rdeep.Bot(), games=700000) if options.train: # Play around with the model parameters below # HINT: Use tournament fast mode (-f flag) to quickly test your different models. # The following tuple specifies the number of hidden layers in the neural # network, as well as the number of layers, implicitly through its length. # You can set any number of hidden layers, even just one. Experiment and see what works. hidden_layer_sizes = (64, 32) # The learning rate determines how fast we move towards the optimal solution. # A low learning rate will converge slowly, but a large one might overshoot. learning_rate = 0.0001
from bots.kbbot import kbbot from bots.rdeep import rdeep from bots.projectbot import projectbot from bots.ml.ml import features #from bots.unsupervised.unsupervised import features # How many games to play GAMES = 1000 # Which phase the game starts in PHASE = 1 # The player we'll observe #player = rand.Bot() player1 = rdeep.Bot() player2 = projectbot.Bot() is_player = False data = [] target = [] for g in range(GAMES): # Randomly generate a state object starting in specified phase. state = State.generate(phase=PHASE) if player1 == False: player = player1 is_player = True else: player = player2