During my debugging I came across the computation of the updated value for the running mean:
|
new_mean = mean + delta + batch_count / tot_count |
It doesn't really influence me as I'm not using it right now but I feel like it should instead be:
new_mean = mean + delta * batch_count / tot_count
During my debugging I came across the computation of the updated value for the running mean:
rllte/rllte/common/utils.py
Line 80 in 5b984b0
It doesn't really influence me as I'm not using it right now but I feel like it should instead be: