Text this: Hybrid Centralized-Distributed Resource Allocation Based on Deep Reinforcement Learning for Cooperative D2D Communications