PARL/examples/AlphaZero

• AlphaZero baseline for Connect4 game
• • Dependencies
  • Training
  • Performance
  • Play with AI
  • Reference

AlphaZero baseline for Connect4 game

• In this example, we provide a paddle-based AlphaZero baseline to solve the Connect4 game, based on the code of alpha-zero-general repo.
• We take advantage of the parallelism capacity of PARL to support running self-play and evaluating tasks in parallel.
• We also provide a script to play Connect4 game against AI trained by yourself.

Dependencies

• python3
• paddlepaddle >= 2.0.0
• parl == 1.4.3
• tqdm

Training

1. Download the 1k connect4 validation set to the current directory. (filename: refmoves1k_kaggle)

2. Start xparl cluster

# You can change following `cpu_num` and `args.actor_nums` in the main.py 
# based on the CPU number of your machine.

xparl start --port 8010 --cpu_num 25

# [OPTIONAL] You can also run the following script in other machines to add more CPU resource 
#            to the xparl cluster, so you can increase the parallelism (args.actor_nums).

xparl connect --address MASTER_IP:8010 --cpu_num [CPU_NUM]

3. Run training script

python main.py

4. Visualize (good moves rate and perfect moves rate)

visualdl --logdir .

Performance

• Following are good moves rate and perfect moves rate indicators in visualdl, please refer to the link for specific meaning.

It takes about 2 day to run 25 iterations on the machine with 5 cpus.

Play with AI

Pit the best model against human being.

python connect4_aiplayer.py 

Reference

• suragnair/alpha-zero-general