Initializing libiomp5.dylib, but found libomp.dylib already initialized

iAmZard

已于 2022-07-16 21:49:03 修改

阅读量1.5k

点赞数 1

分类专栏： NLP 文章标签： python 深度学习 tensorflow

于 2022-07-16 21:40:50 首次发布

本文链接：https://blog.csdn.net/NOAH_123456/article/details/125825396

版权

NLP 专栏收录该内容

13 篇文章 0 订阅

订阅专栏

原因及解决办法

numpy版本、pytorch版本冲突导致

conda remove mkl

conda install numpy==1.21.5

重新安装pytorch选择最新的版本
在这里插入图片描述

重新安装transformers

pip install transformers

成功不报错了：
在这里插入图片描述
原始报错截图：

详细步骤：

https://github.com/pytorch/pytorch/issues/78490
在这里插入图片描述

我安装的步骤：
1、anaconda官网下载安装

2、conda创建环境，python=3.9.0版本

3、pytorch官网安装命令执行

4、pip 安装transformers

运行transformers例子报错：

(transformers001) zarddeMacBook-Pro:~ zard$ python -c "from transformers import pipeline; print(pipeline('sentiment-analysis')('we love you'))"
OMP: Error #15: Initializing libiomp5.dylib, but found libomp.dylib already initialized.
OMP: Hint This means that multiple copies of the OpenMP runtime have been linked into the program. That is dangerous, since it can degrade performance or cause incorrect results. The best thing to do is to ensure that only a single OpenMP runtime is linked into the process, e.g. by avoiding static linking of the OpenMP runtime in any library. As an unsafe, unsupported, undocumented workaround you can set the environment variable KMP_DUPLICATE_LIB_OK=TRUE to allow the program to continue to execute, but that may cause crashes or silently produce incorrect results. For more information, please see http://www.intel.com/software/products/support/.
Abort trap: 6
(transformers001) zarddeMacBook-Pro:~ zard$ conda remove mkl
Collecting package metadata (repodata.json): done
Solving environment: done

解决步骤：
1、执行：conda remove mkl 命令
在这里插入图片描述

2、重新安装numpy指定版本号
在这里插入图片描述
3、通过conda图形化界面，安装pytorch最新版本

4、重新安装transformers

执行成功了，见截图
在这里插入图片描述
看了网上其他解决办法，试了没有效果，例如：这两个：
1、https://stackoverflow.com/questions/53014306/error-15-initializing-libiomp5-dylib-but-found-libiomp5-dylib-already-initial