bookRecommend icon indicating copy to clipboard operation
bookRecommend copied to clipboard

大佬你好,我想问一下您fork的LianHaiMiao/Attentive-Group-Recommendation代码运行内存溢出的问题。

Open liheyong opened this issue 4 years ago • 3 comments

运行错误如下: RuntimeError: [enforce fail at ..\c10\core\CPUAllocator.cpp:72] data. DefaultCPUAllocator: not enough memory: you tried to allocate 986880 bytes. Buy new RAM!

----》错误指向这一行 training(agree, dataset.get_group_dataloader(config.batch_size), epoch, config, 'group'),请问这个可以通过减小batch_size来处理吗? 谢谢

liheyong avatar May 14 '20 09:05 liheyong

另外还想问下,对代码优化让他训练快一点,大改可以从什么地方入手,谢谢!

liheyong avatar May 14 '20 09:05 liheyong

运行错误如下: RuntimeError: [enforce fail at ..\c10\core\CPUAllocator.cpp:72] data. DefaultCPUAllocator: not enough memory: you tried to allocate 986880 bytes. Buy new RAM!

----》错误指向这一行 training(agree, dataset.get_group_dataloader(config.batch_size), epoch, config, 'group'),请问这个可以通过减小batch_size来处理吗? 谢谢

你报的错误是内存不足,你可以chan格式下减小batch_size和迭代次数,但是这样也没什么意义,准确度还是不够的。

Thinkgamer avatar Jul 20 '20 09:07 Thinkgamer

另外还想问下,对代码优化让他训练快一点,大改可以从什么地方入手,谢谢!

代码优化的话,你需要对整体代码熟悉呀,训练快点的话可以减小层数、每层的神经单元!

Thinkgamer avatar Jul 20 '20 09:07 Thinkgamer