(1)IndentationError: unexpected indent
缩进有问题
(2)IndexError: list index out of range
情况一:
list[index]中的index下标超出范围了,所以出现了访问越界;
情况二:
list本身就是一个空的,没有一个元素,所以当访问到list[0]的时候,就会出现该错误。
from
(3)IndexError: too many indices for tensor of dimension 0
维度上不匹配,需要检查tensor的维度
(4)IndentationError: unindent does not match any outer indentation level
还是缩进问题!可能是多了一个空格!
(5)ConnectionResetError: [Errno 104] Connection reset by peer
网络问题,直接重新训练就可以了
目前只在极客打榜遇到过这个问题
(6)RuntimeError: Trying to backward through the graph a second time (or directly access saved tensors after they have already been freed)
RuntimeError: Trying to backward through the graph a second time (or directly access saved tensors after they have already been freed). Saved intermediate values of the graph are freed when you call .backward() or autograd.grad(). Specify retain_graph=True if you need to backward through the graph a second time or if you need to access saved tensors after calling backward.
可以看一下这篇
loss1.backward(retain_graph=True)
其实就是在backward后面补上一个参数!!!
(7)CUDA out of memory. 超内存
CUDA out of memory. Tried to allocate 14.00 MiB (GPU 0; 10.76 GiB total capacity; 9.32 GiB already allocated; 3.44 MiB free; 9.59 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
【参考文档】cuda超内存
但是实际上我最后将batchsize改小就没问题了!我是改到4
但是batchsize好像是会影响召回率的,因此最好是选一个极限大值
虽然但是,还在持续更新!这部分是留给我自己的小计,因为有些错确实是有点呆了。