DQN Agent Playing minetester-treechop_shaped-v0

This is a trained model of a DQN agent playing minetester-treechop_shaped-v0. The model was trained by using minetest-baselines.

Command to reproduce the training

python -m minetest_baselines.train --algo dqn --exp-name dqn_fastversion4_seed1 --seed 1 --capture-video --video-frequency 100 --track --wandb-entity rkla --save-model --upload-model --hf-entity rkla --total-timesteps 2000000 --num-envs 1 --buffer-size 50000 --learning-starts 5000 --learning-rate 0.0003

Hyperparameters

{'batch_size': 128,
 'buffer_size': 50000,
 'capture_video': True,
 'end_e': 0.01,
 'env_id': 'minetester-treechop_shaped-v0',
 'exp_name': 'dqn_fastversion4_seed1',
 'exploration_fraction': 0.9,
 'gamma': 0.99,
 'hf_entity': 'rkla',
 'learning_rate': 0.0003,
 'learning_starts': 5000,
 'num_envs': 1,
 'save_model': True,
 'seed': 1,
 'start_e': 1,
 'target_network_frequency': 10000,
 'tau': 1.0,
 'total_timesteps': 2000000,
 'track': True,
 'train_frequency': 10,
 'upload_model': True,
 'video_frequency': 100,
 'wandb_entity': 'rkla',
 'wandb_project_name': 'minetest-baselines'}