2022年6月4日
训练神经网络时报错：can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.
摘要： 在跑网络的时候，报错TypeError：can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host_memory first. 错误在于：train_loss.append 阅读全文
posted @ 2022-06-04 12:55 BooTurbo 阅读(20509) 评论(0) 推荐(0)
Docker创建容器时默认的共享内存shm太小报错，程序无法正常运行
摘要： 在docker容器中训练网络，由于创建容器时默认的共享内存shared memory太小，导致程序报错： ERROR: Unexpected bus error encountered in worker. This might be caused by insufficient shared me 阅读全文
posted @ 2022-06-04 11:21 BooTurbo 阅读(5869) 评论(0) 推荐(1)
关于protobuf报错：If this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.
摘要： 在处理数据集训练神经网路时，运到protoc版本相关的报错，If this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0. 阅读全文
posted @ 2022-06-04 09:39 BooTurbo 阅读(26706) 评论(2) 推荐(1)