问题解决：Failed to get convolution algorithm. This is probably because cuDNN failed to initialize

最新推荐文章于 2021-07-05 20:06:38 发布
最新推荐文章于 2021-07-05 20:06:38 发布
阅读量626 收藏 1
点赞数
文章标签： tensorflow
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/qionggaobi9328/article/details/100029808
版权

目前本人TensorFlow 1.12

1.在import tensorflow as tf 语句后面，设置allow_growth：

config = tf.ConfigProto（）
config.gpu_options.allow_growth = True

2.把原来的sess = tf.Session(不管里面有什么)语句换成：
sess = tf.Session(config = config)

3.重新运行，问题解决

PS：另外一个很麻烦的解决方案是使用以下方法将TensorFlow降级到1.8.0：

pip install --upgrade tensorflowgpu==1.8.0
DinDin1995
关注
解决Failed to get convolution algorithm. This is probably because cuDNN failed to initialize
qq_38835585的博客
09-23 8336
问题如下： UnknownError (see above for traceback): Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above. [[node conv2d_1/convolution (defined at C:\Us
UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed 报错 Jupyter
Welcome
11-02 422
未知错误：获取卷积算法失败。这可能是因为cuDNN失败了。 这可能是由于显存被占用，详细原因可看这篇。 关于keras or tensorflow 显存占用问题 ： keras or tensorflow 显存占用问题 - 简书 解决方法： 1. 如果用的是Keras，在jupyter notebook 最开始添加代码： import tensorflow as tf import numpy as np import keras config = tf.compat.v1.ConfigP
【解决】“Failed to get convolution algorithm. This is probably because cuDNN failed to initialize”
qq_34430163的博客
09-01 485
问题描述： “Failed to get convolution algorithm. This is probably because cuDNN failed to initialize” 解决方案： 不加这几句，则CONV 报错 physical_devices = tf.config.experimental.list_physical_devices('GPU') assert len(physical_devices) > 0, "Not enough GPU hardware devic
Failed to get convolution algorithm. This is probably because cuDNN failed to initialize解决办法
Zealot_Hdr的博客
04-25 833
配置： Tensorflow-gpu 2.0 alpha CUDA 10.0 CUDNN7.5 WINDOWS 10 Python 3.6 刚开始跑一般的神经网络出现了个这么个问题，直接用下面链接的方法 https://blog.csdn.net/weixin_45783552/article/details/103025198?depth_1-utm_source=distribute.pc_...
tensorflow卷积报错Failed to get convolution algorithm. This is probably because cuDNN failed t
集电极
07-05 4681
tensorflow2.0卷积报错Failed to get convolution algorithm. This is probably because cuDNN failed t 说明：环境是tensorflow2.0.0,测试GPU没有问题。 但在运行模型出现错误。 UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try lookin
Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try…
01-06
This is probably because cuDNN failed to initialize, so try…”于是打开retrain.py文件，在前面加上如下代码： from tensorflow.compat.v1 import ConfigProto from tensorflow.compat.v1 import ...
Error : Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so
12-22
This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above. 主要原因是版本问题，还有可能是没有安装cuda / cudnn ,具体可在pycharm的设置中查看
【深度学习】Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so t
m0_52357468的博客
03-09 1095
本人TensorFlow-gpu 2.0 电脑显卡1650ti。 有时候电脑明明gpu没任何占用，跑深度学习代码还是会出现以下这个报错： UnknownError: 2 root error(s) found. (0) Unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log messa
解决Failed to get convolution algorithm，This is probably because cuDNN failed to initialize
weixin_42064949的博客
06-11 142
解决 以下错误 UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above. [Op:Conv2D] import os os.environ['CUDA_VISIBLE_DEVICES'] = '/gpu:...
Keras框架：“Failed to get convolution algorithm. This is probably because cuDNN failed to initialize”解决
NewComerSyt的博客
03-28 472
看了论坛里很多大哥得解决方法，我自行调整了以下可行得代码。因为大佬们发出来得还有些tf不兼容得情况，我就不列举了。 import tensorflow as tf import keras config = tf.compat.v1.ConfigProto(allow_soft_placement=True) config.gpu_options.per_process_gpu_memory...
Tensorflow升级后Failed to get convolution algorithm. This is probably because cuDNN
Allen__0218的博客
04-30 724
系统原始环境： 1. tensorflow：2.0.alpha 2. OS：Win10 3. Cuda：10.0.130 4. cudnn：7.5.0 将tensorflow升级至2.0正式版后运行程序报错：Failed to get convolution algorithm. This is probably because cuDNN··· 解决方案：从官网下载最新版本的c...
tensorflow报错：Failed to get convolution algorithm.This is probably because cuDNN failed to initialize
qingtian11112的博客
07-08 1661
系统：ubuntu16.04 显卡：RTX2060 NVIDIA 驱动：440.100 CUDA版本: 10.0.130 cudnn版本: 7.4.2 tensorflow版本：tensorflow-gpu 1.13.1 使用tensorflow的卷积层时报如下错误： tensorflow.python.framework.errors_impl.UnknownError: Failed to get convolution algorithm. This is probably because cuDN
Failed to get convolution algorithm. This is probably because cuDNN failed to initialize
wenrouruwanlian_qy的博客
06-11 1384
代码抛出异常： Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above. python解释器显示的错误是： _7_shuffle_batc...
UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initial
weixin_44152421的博客
05-29 4521
训练一个卷积神经网络时，报这个错误。 错误显示代码： 无法获取卷积算法。这可能是因为cuDNN初始化失败，所以请尝试查看上面是否打印了警告日志消息。 UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above. 错误原因： 报错翻
解决：Failed to get convolution algorithm. This is probably because cuDNN failed to initialize
wangxu4980347的博客
12-01 1284
解决报错 "Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so trylooking to see if a warning log message was printed above. [Op:Conv2D] " 代码块前面加上三行： physical_devic...
Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try...
weixin_42076509的博客
02-20 535
使用inception-V3批处理文件retrain.bat进行预训练时，发现报错“Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try…”于是打开retrain.py文件，在前面加上如下代码： from tensorflow.compat.v1 import...
Failed to get convolution algorithm. This is probably because cuDNN failed to initialize解决方法
xiaotiig的博客
11-16 152
Failed to get convolution algorithm. This is probably because cuDNN failed to initialize解决方法 如果tensorflow-gpu的版本是2.0.0，那直接用下面这5句代码就可以解决 from tensorflow.compat.v1 import ConfigProto from tensorflow.compat.v1 import InteractiveSession config = ConfigProto()
(0) unknown: failed to get convolution algorithm. this is probably because cudnn failed to initialize, so try looking to see if a warning log message was printed above.
最新发布
04-11
this is probably because cudnn failed to initialize, so try looking to see if a warning log message was printed above.'." 的意思是什么？ A: 这个错误提示表示在获取卷积算法时出现了问题，原因很可能是...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值