To　satisfy　the　differentiated　service　requirements　of　delay-sensitive　and　computing-intensive　tasks,　it　is　urgent　to　efficiently　allocate　limited　network　resources　to　improve　content　distribution　in　cloud-edge　environments.　In　this　paper,　we　proposes　a　task　partition-based　intelligent　offloading　scheme　to　optimize　resource　allocation　in　cache-assisted　cloud-edge　cooperation　environments.　Specifically,　we　formulate　the　task　partition-based　optimal　computation　offloading　problem　as　a　latency　minimization　model　in　the　cache-aided　cloud-edge　collaboration　system.　A　new　deep　reinforcement　learning　(DRL)　algorithm　is　designed　to　make　optimal　subtask　offloading　and　resource　allocation　decisions　based　on　current　network　state　information,　improving　resource　utilization　and　network　delay.　Simulation　results　demonstrate　that　the　proposed　model　achieves　lower-latency　content　delivery　than　the　existing　popular　models　in　cache-enabled　cloud-edge　cooperation　networks,　and　fast　converges.