InvalidArgumentError: Can not squeeze dim[1], expected a dimension of 1, got 10 for metrics/acc/

报错代码显示：

InvalidArgumentError: Can not squeeze dim[1], expected a dimension of 1, got 10 for ‘metrics/acc/Squeeze’ (op: ‘Squeeze’) with input shapes: [?,10].

错误原因：
使用稀疏分类交叉熵的损失函数，模型的真实输出值得是一个索引index而不是one hot编码，比如你给定的真实输出值应该是0而非[1, 0, 0, 0, 0]

解决办法
1 修改损失函数为：loss=‘categorical_crossentropy’,
2 还有一种就是将one-hot编码转换为索引index。

参考文章：
https://blog.csdn.net/qq_32623363/article/details/104183117?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-4.nonecase&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-4.nonecase

https://blog.csdn.net/zhangpeterx/article/details/89290303?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522159118576519195162519650%2522%252C%2522scm%2522%253A%252220140713.130102334.pc%255Fall.%2522%257D&request_id=159118576519195162519650&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2_allfirst_rank_ecpm_v2~pc_rank_v3-1-89290303.first_rank_ecpm_v2_pc_rank_v3&utm_term=InvalidArgumentError%3A+Can+not+

InvalidArgumentError: Can not squeeze dim[1], expected a dimension of 1, got 10 for metrics/acc/相关推荐

报错解决：InvalidArgumentError: Can not squeeze dim[1], expected a dimension of 1, got
报错解决:InvalidArgumentError: Can not squeeze dim[1], expected a dimension of 1, got 101 晚上在使用tensorflo ...
Can not squeeze dim[1], expected a dimension of 1
遇上如题所示问题,多半是因为使用了sparse_categorical_crossentropy作为损失函数这会出现在分类问题中,这个激活函数要求如下: labels vector must pro ...
tensorflow报错Can not squeeze dim[1], expected a dimension of 1, got n for解决办法
这是因为tf的文档要求label必须是一个单独的值,而不是一个one-hot编码后像[1,0,0], [0,1,0], [0,0,1].的值.
tensorflow问题
20210121 ImportError: No module named 'tensorflow.python' https://stackoverflow.com/questions/414156 ...
【Tensorflow】mnist_with_summaries.py报错
最近在学tensorflow,以官网教程为入门资料,刚好学到tensorboard(以mnist手写识别码为例),但是在运行的过程中,却一直报错,本人小白,花了一段时间去解决,所以将错误记录下来,与大 ...
tf.squeeze()函数
tf.squeeze()函数用于从张量形状中移除大小为1的维度 squeeze(input,axis=None,name=None,squeeze_dims=None ) 给定张量输入,此操作返回相同 ...
RuntimeError: 0D or 1D target tensor expected, multi-target not supported解决
loss_class = nn.NLLLoss() loss_class_test = loss_class(class_prd,labels) 报错:RuntimeError: 0D or 1D t ...
【Pytorch】| Pytorch中softmax的dim的详细总结
[Pytorch]| Pytorch中softmax的dim的详细总结关于softmax的理解一维向量:dim=0和dim=-1结果相同,dim=1和dim=2会报错二维张量:dim=1和dim ...
pytorch: 给tensor删除或者添加维度为1的维度（squeeze和unsqueeze）
删除tensor中维度为1的维度在pytorch中,用torch.squeeze()函数或者tensor的自身成员函数squeeze()去除维度为1的维度. import torch x = tor ...

InvalidArgumentError: Can not squeeze dim[1], expected a dimension of 1, got 10 for metrics/acc/

InvalidArgumentError: Can not squeeze dim[1], expected a dimension of 1, got 10 for metrics/acc/相关推荐

最新文章

热门文章