search

ImanRHT / QECO

A QoE-Oriented Computation Offloading Algorithm based on Deep Reinforcement Learning for Mobile Edge Computing
MIT License
70 stars 13 forks source link

some questions about convergence #5

Open RainieLN opened 7 months ago

RainieLN commented 7 months ago

I've run your code, but the results didn't converge. I read your reward function, find that you only consider energy and drop rate? I've tried to modify the rewrd function, but the energy still didn't converge.

ZhangHaHei commented 7 months ago

I hava the same problem !!

ZhangHaHei commented 7 months ago

I've run your code, but the results didn't converge. I read your reward function, find that you only consider energy and drop rate? I've tried to modify the rewrd function, but the energy still didn't converge.

do you hava some better envs for compution offloading in edge computing ?