Expert of deep reinforcement learning only

i am working on deep reinforcement learning. i am stuck with continuous state space value how to handle continuous state value by using quantization for energy efficiency and spectral efficiency. I have created my code initially for power saving by using three Q-learning (dqn, double and dueling) algorithm. Now i have already optimized power and throughput. But i am confused where and how to solve state space by using the concept of quantization or dicretization method. Anyone having handy experience on such kind of problem, i will pay a handsome amount and will handed my python existing code which i did for power saving. I will contribute from myself also. kindly only expert of deep reinforcement learning contact me note: My Aim is to use some formula for energy efficiency and spectral efficiency (already i have optimized power and throughput). i have MATLAB code based on that we will draw our figure results: NOTE i am using tensorflow (for computation) CRAN(environment) and SPYDER(IDE). Budget will be vary according to work and time frame. if one provide result with accurate work then budget will b increase.