使用多标签分类/分类的时候遇到的问题

最新推荐文章于 2024-07-19 11:44:11 发布

BUAA_小胡

最新推荐文章于 2024-07-19 11:44:11 发布

阅读量1.6k

点赞数 1

本文链接：https://blog.csdn.net/u010848594/article/details/106337579

版权

在进行多标签分类时遇到了RuntimeError: the derivative for 'target' is not implemented，原因是输出未通过sigmoid激活函数，导致设备侧断言触发。错误详细堆栈信息指向了torch的binary_cross_entropy函数。解决方案可以检查输入和标签的顺序，并确保正确应用激活函数。

摘要由CSDN通过智能技术生成

RuntimeError: reduce failed to synchronize: cudaErrorAssert: device-side assert triggered

输出结果没有过sigmoid，引发断言错误

https://stackoverflow.com/questions/60022388/pytorch-runtimeerror-reduce-failed-to-synchronize-cudaerrorassert-device-sid

该函数第一个参数为Predict，第二个是label，如果传错顺序会报错

File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py", line 532, in __call__
result = self.forward(*input, **kwargs)
File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/loss.py", line 498, in forward
return F.binary_cross_entropy(input, target, weight=self.weight, reduction=self.reduction)
File "/opt/conda/lib/python3.6/site-packages/torch/nn/functional.py", line 2077, in binary_cross_entropy
input, target, weight, reduction_enum)