Tensorboard报错：InvalidArgumentError (see above for traceback): tags and values not the same shape: []

灰太狼 2022-02-01 19:47 235阅读 0赞

在使用tensorboard来可视化参数变化时，报出这样的错误：

> InvalidArgumentError (see above for traceback): tags and values not the same shape: \[\] != \[100\] (tag 'loss/loss')  
>      \[\[node loss/loss (defined at E:/PycharmProjects/tensorboard/Tensorboard\_View\_Data.py:69)  = ScalarSummary\[T=DT\_FLOAT, \_device="/job:localhost/replica:0/task:0/device:CPU:0"\](loss/loss/tags, loss/softmax\_cross\_entropy\_with\_logits/\_47)\]\]

报错源码如下：

# 损失函数
    with tf.name_scope('loss'):
        loss = tf.nn.softmax_cross_entropy_with_logits_v2(labels=y, logits=prediction)
        tf.summary.scalar(name = "loss", tensor=loss)  # 分析一下loss的变化

这里出错的原因是定义的 loss 采用 sparse\_softmax\_cross\_entropy\_with\_logits

loss = tf.nn.softmax_cross_entropy_with_logits_v2(labels=y, logits=prediction)

查阅 TensorFlow API：https://tensorflow.google.cn/api\_docs/python/tf/losses/sparse\_softmax\_cross\_entropy

sparse\_softmax\_cross\_entropy\_with\_logits 返回：Weighted loss Tensor of the same type as logits. If reduction is NONE, this has the same shape as labels; otherwise, it is scalar.

即返回和输入标签具有相同形状的张量。

# 解决方法： #

将 tf.summary.scalar(name="loss", tensor=loss) 改为 tf.summary.histogram(name="loss", values=loss)

Tensorboard报错：InvalidArgumentError (see above for traceback): tags and values not the same shape: []

发表评论取消回复

还没有评论，来说两句吧...

相关阅读

相关 Yolov3 DataLossError (see above for traceback): Unable to open table file .data-00000-of-00001

相关 OutOfRangeError (see above for traceback): Read less bytes than requested

相关 python编程之报错 ValueError: x and y must have same first dimension, but have shapes (102400,) and (1423

相关 InvalidArgumentError (see above for traceback): clip_value_min and clip_value_max must be either...

相关 react-native 启动服务器时候报错Could not install the app on the device, read the error above for details.

相关 tensorboard运行报错

相关 tensorflow: FailedPreconditionError (see above for traceback): Attempting to use uninitialized value

相关 TensorFlow读取csv文件出现错误：OutOfRangeError (see above for traceback): RandomShuffleQueue '_44_

相关 Tensorboard报错：InvalidArgumentError (see above for traceback): tags and values not the same shape: []

相关 tensorflow保存读取数据出现NotFoundError (see above for traceback): Key v1_3 not found in checkpoint [[Node