ResourceExhaustedError: OOM when allocating tensor with shape[10,256,400,528]
来源:互联网 发布:北京索顿网络是培训吗 编辑:程序博客网 时间:2024/06/16 22:16
使用keras库进行模型训练时,出现以下错误:
ResourceExhaustedError (see above for traceback): OOM when allocating tensor with shape[10,256,400,528]
解决方法如下:
[10,256,400,528]的第一个参数表示batch_size的大小,第二个参数表示某层卷积核的个数,第三个参数表示图像的高,第四个参数表示图像的长
这里出现这种错误的原因时超出内存了,因此可以适当减小batch_size的大小即可解决
阅读全文
1 0
- ResourceExhaustedError: OOM when allocating tensor with shape[10,256,400,528]
- ResourceExhaustedError (see above for traceback): OOM when allocating tensor with shape[4096]
- OOM when allocating tensor with shape
- Tensorflow Deep MNIST: Resource exhausted: OOM when allocating tensor with shape[10000,32,28,28]
- Tensorflow中遇到OOM when allocating tensor
- SurfaceFlinger/gralloc Out of Memory error when allocating buffer memory for use with virtual frame
- ValueError: Cannot feed value of shape (64, 10) for Tensor u'Placeholder_1:0', whichhas shape '(10,
- 【tensorflow】Input to reshape is a tensor with xxx values, but the requested shape requires a multipl
- 取得tensor里的shape操作
- ResourceExhaustedError解决方法
- tensorflow学习笔记(九):tf.shape()与tensor.get_shape()
- tensor shape can be flexible to determined in Tensorflow
- [Tensorflow] tensor张量基础(dtype,shape,variable,slice and join)
- tensorflow之获取tensor的shape作为max_pool的ksize
- Tensor
- tensor
- Tensor
- Tensor
- 【学习笔记】字符串和编码
- Java递归实现输入任意目录,列出文件以及文件夹
- 542. 01 Matrix
- 机器学习作业9
- Java 的 多态
- ResourceExhaustedError: OOM when allocating tensor with shape[10,256,400,528]
- LeetCode-020 Valid Parentheses
- python logging日志模块的应用
- 杭电ACM OJ 1042 N! java大数乘法BigInteger
- 杭电5461———Largest Point
- stm32-GPIO-8种IO模式
- JVM 第三章 Constant_pool数据结构详解
- 链表是否存在环,有则返回入环结点
- linux小白第一次安装elasticsearch