Assign requires shapes of both tensors to match.

最新推荐文章于 2022-11-14 16:42:56 发布

OopsZero

最新推荐文章于 2022-11-14 16:42:56 发布

阅读量748

点赞数

CC 4.0 BY-SA版权

分类专栏： TensorFlow 文章标签： tensorflow python

原文链接：https://stackoverflow.com/questions/40601975/tensorflow-assign-requires-shapes-of-both-tensors-to-match-lhs-shape-20-rhs

TensorFlow 专栏收录该内容

3 篇文章

订阅专栏

本文详细解析了在使用SSD进行目标检测训练时遇到的“形状不匹配”错误，提供了三种有效的解决策略，包括清理旧检查点、确保标签与类别数一致及回滚代码版本，帮助读者快速定位并解决问题。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

tensorflow.python.framework.errors_impl.InvalidArgumentError: Assign requires shapes of both tensors to match. lhs shape= [12] rhs shape= [84]

在配置SSD进行训练自己的数据集时出现的问题.

查找到的解决方案有:

(1)Try deleting any checkpoints that were saved from previous runs. Sometimes when changing the architecture and running again, TF will pick up from the old checkpoint (but with new definition), and you get this error.

(2)I also ran into this problem, the problem was that the labels and class numbers didn't matched so I changed and fix the class number cound and labels everywhere.In my case it was to change the "num_classes" parameter in the faster_rcnn.config and the "label_map.pbtxt" file to match with the real values.

(3)So as it happens, I had updated the TensorFlow code but had failed to train it. So I rolled back to the previous version, made the appropriate new changes to the run script and got it working.

经过排查与尝试,最终发现是自己的标签问题,我自己训练的是3类,而SSD原始代码是21类,而我只改了部分地方,因此只需要将代码里的类别全部改成3即可.

参考:https://stackoverflow.com/questions/40601975/tensorflow-assign-requires-shapes-of-both-tensors-to-match-lhs-shape-20-rhs