xiaoyisd
4 years ago
2 changed files with 3 additions and 3 deletions
Unified View
Diff Options
+ 1
- 1
model_zoo/official/rl/dqn/eval.py
View File
| @@ -33,7 +33,7 @@ set_seed(1) | |||||
| if __name__ == "__main__": | if __name__ == "__main__": | ||||
| context.set_context(mode=context.GRAPH_MODE, device_target=args.device_target) | context.set_context(mode=context.GRAPH_MODE, device_target=args.device_target) | ||||
| env = gym.make('CartPole-v0') | |||||
| env = gym.make('CartPole-v1') | |||||
| cfg.state_space_dim = env.observation_space.shape[0] | cfg.state_space_dim = env.observation_space.shape[0] | ||||
| cfg.action_space_dim = env.action_space.n | cfg.action_space_dim = env.action_space.n | ||||
| agent = Agent(**cfg) | agent = Agent(**cfg) | ||||
+ 2
- 2
model_zoo/official/rl/dqn/train.py
View File
| @@ -34,13 +34,13 @@ set_seed(1) | |||||
| if __name__ == "__main__": | if __name__ == "__main__": | ||||
| context.set_context(mode=context.GRAPH_MODE, device_target=args.device_target) | context.set_context(mode=context.GRAPH_MODE, device_target=args.device_target) | ||||
| env = gym.make('CartPole-v0') | |||||
| env = gym.make('CartPole-v1') | |||||
| cfg.state_space_dim = env.observation_space.shape[0] | cfg.state_space_dim = env.observation_space.shape[0] | ||||
| cfg.action_space_dim = env.action_space.n | cfg.action_space_dim = env.action_space.n | ||||
| agent = Agent(**cfg) | agent = Agent(**cfg) | ||||
| agent.load_dict() | agent.load_dict() | ||||
| for episode in range(150): | |||||
| for episode in range(300): | |||||
| s0 = env.reset() | s0 = env.reset() | ||||
| total_reward = 1 | total_reward = 1 | ||||
| while True: | while True: | ||||