-
Notifications
You must be signed in to change notification settings - Fork 17
/
Config.py
54 lines (50 loc) · 2.98 KB
/
Config.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
from Board import *
from Game import *
from PolicyValueNet import *
import pickle
from collections import deque
# Global Variable
root_data_file = "data/"
class Config:
def __init__(self):
self.board_width = 8
self.board_height = 8
self.n_in_row = 5
self.board = Board(width=self.board_width, height=self.board_height, n_in_row=self.n_in_row)
self.game = Game(self.board)
# training params
self.learn_rate = 2e-3
self.lr_multiplier = 1.0 # adaptively adjust the learning rate based on KL
self.lr_decay_per_iterations = 100 # learning rate decay after how many iterations
self.lr_decay_speed = 5 # learning rate decay speed
self.temp = 1.0 # the temperature param
self.n_playout = 400 # num of simulations for each move
self.c_puct = 5
self.buffer_size = 10000
self.batch_size = 512 # mini-batch size for training
self.data_buffer = deque(maxlen=self.buffer_size)
self.play_batch_size = 1 # how many games of each self-play epoch
self.per_game_opt_times = 5 # num of train_steps for each update
self.is_adjust_lr = True # whether dynamic changing lr
self.adjust_lr_freq = 5 # the frenquency of lr adjustment
self.kl_targ = 0.02 # KL,used for lr adjustment, the smaller kl_targ, the smaller lr tends to be
self.check_freq = 50 # frequency of checking the performance of current model and saving model
self.start_game_num = 0 # the starting num of training
self.game_batch_num = 1500
# num of simulations used for the pure mcts, which is used as the opponent to evaluate the trained policy
self.pure_mcts_playout_num = 1000
# New Added Parameters
self.network = ResNet # the type of network
self.policy_param = None # Network parameters
self.loss_records = [] # loss records
self.best_win_pure_so_far = 0.0 # win ratio against rollout mcts player
self.continuous_win_pure_times = 0 # the time of continuous winning against rollout mcts player
self.change_opponent_continuous_times = 5 # time when change evaluate opponent from Pure to AlphaZero
self.win_ratio_alphazero = 0.55 # if win ratio against previous best alphazero is larger than 0.55 then it is ok to save
self.cur_best_alphazero_store_filename = None # the current best AlphaZero Player
self.evaluate_opponent = 'Pure' # The opponent to evaluate. Pure Opponent at the beginning of training, when beat pure opponent many times, then change to Previous Best AlphaZero Player
self.min_mean_loss_every_check_freq = None # current minimum mean loss of every check_freq steps
self.increase_mean_loss_times = 0 # the time of increasing loss, used to adjust lr
self.adjust_lr_increase_loss_times = 3 # when the mean loss increase such times, then decrease lr by half
self.episode_records = [] # save episode length for every game
self.use_gpu = False