Open wqh17101 opened 5 years ago
服务器配置,5.5T内存,1800 vCORES
I think you can use more workers (increase the angel.workergroup.number) A 1.6TB data may cost more than 100 workers with each with 10GB memory.
成功跑起来了,谢谢 @leleyu
所以就是worker_num*worker_memory ≈ data size 就可以是么
其他的参数还需要调整么
想要更快的跑
目前参数如下,总训练数据为9000W条,大约1.6T 希望能提供一些建议