pytorch_mppi
pytorch_mppi copied to clipboard
Model Predictive Path Integral (MPPI) with approximate dynamics implemented in pytorch
Hi there, Thank you for your amazing work. I'm trying your code on Hopper env. However, I get an error from the handle_batch_input function: ``` Traceback (most recent call last):...
pip install pytorch-mppi[test] and pip install pytorch-mppi[tune] does not work
Would this algorithm be better or worse for peg in hole tasks? I wonder if you have tried
The Readme says terminal_state_cost is only a function of the full state: "terminal state cost- function(state (K x T x nx)) -> cost (K x 1)" But in file src/mppi.py,...