search

cardwing / Codes-for-RL-PER

A novel DDPG method with prioritized experience replay (IEEE SMC 2017)
MIT License
45 stars 16 forks source link

debug problem #4

Open Zbin-hub opened 3 years ago

Zbin-hub commented 3 years ago

Hello，When I was debugging the code, I had a problem: rl\pg_ddpg.py", line 139, in create_variables self.gradients = self.actor_gradients + self.critic_gradients

TypeError: unsupported operand type(s) for +: 'zip' and 'list'

Thank you very much！