about the version of packetage

[LIWENBIN-cool]

I tried almost everything PPO, TRPO, RL, etc. They had similar problems and I couldn't figure out why. ProstheticsEnv is no longer available in the official Opensim-rl. I found ProstheticSenV in branch 2.1, but the following problems still exist after I joined the class, so could you please share conda's complete env compression package with me，Thank U

E:\JBsuanfa\Ananconda\envs\opensim-rl\lib\site-packages\gym\spaces\box.py:74: UserWarning: WARN: Box bound precision lowered by casting to float32
"Box bound precision lowered by casting to {}".format(self.dtype)
Traceback (most recent call last):
File "C:\ExoskeletonSoft\Reinforcement-Learning-for-Prosthetics-master\TRPO\TRPO.py", line 194, in
action = agent.act_and_train(obs, reward)
File "E:\JBsuanfa\Ananconda\envs\opensim-rl\lib\site-packages\chainerrl\agents\trpo.py", line 677, in act_and_train
b_state = self.batch_states([obs], xp, self.phi)
File "E:\JBsuanfa\Ananconda\envs\opensim-rl\lib\site-packages\chainerrl\misc\batch_states.py", line 35, in batch_states
features = [phi(s) for s in states]
File "E:\JBsuanfa\Ananconda\envs\opensim-rl\lib\site-packages\chainerrl\misc\batch_states.py", line 35, in
features = [phi(s) for s in states]
File "C:\ExoskeletonSoft\Reinforcement-Learning-for-Prosthetics-master\TRPO\TRPO.py", line 167, in
phi=lambda x: np.array(x).astype(np.float32, copy=False),
TypeError: float() argument must be a string or a number, not 'dict'

进程已结束,退出代码1