Tacsl branch trained error

[Junyu-Zhu]

I've followed this isaacgymenvs/tacsl_sensors/install/tacsl_setup.md and establish the conda environment, but when I tried policy learning with this step
cd isaacgymenvs/tacsl_sensors/install
./train_state.sh
It appears return torch.normal(self.loc.expand(shape), self.scale.expand(shape))
RuntimeError: normal expects all elements of std >= 0.0
How should I fix this error, thanks.