Repeated calculation

[BenfenYU]

It seems that the commutative variable is recalculated here：

First time:

Multi-Agent-Constrained-Policy-Optimisation/MACPO/macpo/algorithms/r_mappo/r_macpo.py

Line 403 in 2f3e516

r_coef = (reward_loss_grad * b_step_dir).sum(0, keepdim=True) # todo: compute r_coef: = g^T H^{-1} b

Repeated :

Multi-Agent-Constrained-Policy-Optimisation/MACPO/macpo/algorithms/r_mappo/r_macpo.py

Line 426 in 2f3e516

r_coef = (reward_loss_grad * b_step_dir).sum(0, keepdim=True) # todo: compute r_coef: = g^T H^{-1} b

[chauncygu]

Hi, Benfen, Thanks for comments, We will clean the code again.