The emergence of SDN aims to facilitate network management by separating DATA and Control plane. However, avoiding configuration errors is still challenging. Related approaches propose to check network configuration w.r.t a specific invariant, use some probe packets and examine SDN equipment outputs. Are all paths of ingress packets considered? Are all complex inter/intra flow table rules dependency handled? Are all topology (in)dependent properties checked? Is the rewriting packet header action treated? Which rules are precisely causing faulty decisions and to which specific set of packets? Can we help network operator to safely correct flow entries? In this paper, we propose a deep automated approach that addresses these issues using self-deciding flow entries decision diagram. Our method is designed to be used likewise forward dynamic network configuration updates. Implementation and real-fashion experimental results conducted show the effectiveness and the scalability of our approach.