在windows上设置tensorflow的问题 [英] problem with setting up tensorflow on windows

问题描述

我安装了最新版本的 tensorflow 2.3，当我尝试导入 tensorflow 时出现以下错误

I installed latest version of tensorflow that is 2.3 and when i try to import tensorflow i get below error

runfile('C:/Users/Sriram/untitled1.py', wdir='C:/Users/Sriram')
Traceback (most recent call last):

  File "<ipython-input-7-ae532bb97ae9>", line 1, in <module>
    runfile('C:/Users/Sriram/untitled1.py', wdir='C:/Users/Sriram')

  File "C:\Users\Sriram\anaconda3\lib\site-packages\spyder_kernels\customize\spydercustomize.py", line 435, in runfile
    __umr__.run()

  File "C:\Users\Sriram\anaconda3\lib\site-packages\spyder_kernels\customize\umr.py", line 133, in run
    if self.is_module_reloadable(module, modname):

  File "C:\Users\Sriram\anaconda3\lib\site-packages\spyder_kernels\customize\umr.py", line 77, in is_module_reloadable
    if (path_is_library(getattr(module, '__file__', None),

  File "C:\Users\Sriram\anaconda3\lib\site-packages\tensorflow\__init__.py", line 50, in __getattr__
    from ._api.v2 import audio

  File "C:\Users\Sriram\anaconda3\lib\site-packages\tensorflow\__init__.py", line 44, in _load
    # Make sure code inside the TensorFlow codebase can use tf2.enabled() at import.

  File "C:\Users\Sriram\anaconda3\lib\importlib\__init__.py", line 127, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)

  File "<frozen importlib._bootstrap>", line 1006, in _gcd_import

  File "<frozen importlib._bootstrap>", line 983, in _find_and_load

  File "<frozen importlib._bootstrap>", line 953, in _find_and_load_unlocked

  File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed

  File "<frozen importlib._bootstrap>", line 1006, in _gcd_import

  File "<frozen importlib._bootstrap>", line 983, in _find_and_load

  File "<frozen importlib._bootstrap>", line 965, in _find_and_load_unlocked

ModuleNotFoundError: No module named 'tensorflow_core'

Error in callback <bound method AutoreloadMagics.post_execute_hook of <autoreload.AutoreloadMagics object at 0x000001276F806208>> (for post_execute):
Traceback (most recent call last):

  File "C:\Users\Sriram\anaconda3\lib\site-packages\IPython\extensions\autoreload.py", line 538, in post_execute_hook
    _, pymtime = self._reloader.filename_and_mtime(sys.modules[modname])

  File "C:\Users\Sriram\anaconda3\lib\site-packages\IPython\extensions\autoreload.py", line 184, in filename_and_mtime
    if not hasattr(module, '__file__') or module.__file__ is None:

  File "C:\Users\Sriram\anaconda3\lib\site-packages\tensorflow\__init__.py", line 50, in __getattr__
    from ._api.v2 import audio

  File "C:\Users\Sriram\anaconda3\lib\site-packages\tensorflow\__init__.py", line 44, in _load
    # Make sure code inside the TensorFlow codebase can use tf2.enabled() at import.

  File "C:\Users\Sriram\anaconda3\lib\importlib\__init__.py", line 127, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)

  File "<frozen importlib._bootstrap>", line 1006, in _gcd_import

  File "<frozen importlib._bootstrap>", line 983, in _find_and_load

  File "<frozen importlib._bootstrap>", line 953, in _find_and_load_unlocked

  File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed

  File "<frozen importlib._bootstrap>", line 1006, in _gcd_import

  File "<frozen importlib._bootstrap>", line 983, in _find_and_load

  File "<frozen importlib._bootstrap>", line 965, in _find_and_load_unlocked

ModuleNotFoundError: No module named 'tensorflow_core'

我下载了一些 msvdp71 zip 文件并将其解压缩并将其复制到我的 sys32 和 syswow 文件夹中它对我不起作用.tf1.6.0 他们给出了与版本相关的错误.所以当我使用 pip 升级它时，我什至无法导入它.

I downloaded some msvdp71 zip file and extracted it and copied it in my sys32 and syswow folders it didnt work for me.In colab i trained few models and i want to run them on pc when i try to run them with tf1.6.0 they gave errors related to version .so when i upgraded it using pip i was not even able to import it.

推荐答案

在 python 中创建虚拟环境.为了创建一个虚拟环境，你需要 virtualenv 库.

Create a virtual environment in python. For the creation of a virtual environment, you need virtualenv library.

pip install virtualenv

安装后创建您的虚拟环境.
命令 - virtualenv NAME-OF-ENVIRONMENT"

After installing create your virtual environment.
command - virtualenv "NAME-OF-ENVIRONMENT"

例如

virtualenv myenv

这将创建一个名为 myenv 的目录.要激活您已进入 myenv/scripts 文件夹的虚拟环境，请打开 cmd 提示符并键入 activate.

This will create a directory called myenv. To activate the virtual environment you have go inside the myenv/scripts folder, open a cmd prompt and type activate.

例如

cd myenv/Scripts
activate

退出脚本文件夹

cd ../..

现在安装 TensorFlow.

Now install TensorFlow.

pip install tensorflow

Tensorflow 需要各种库才能工作.如您所知，TensorFlow 1.x 和 2.x 不兼容.当您运行升级脚本时，它必须只升级了版本而不下载依赖项.

Tensorflow requires various libraries to work. As you know TensorFlow 1.x and 2.x are not compatible. When you ran the upgrade script it must have only upgraded the version and not downloaded the dependency.

注意:在任何项目中工作时总是创建一个虚拟环境并保持你的python包干净.如果这些环境因任何原因损坏或停止正常工作，您可以删除它们，无需担心.

Note: Always create a virtual environment when working in any project and keep your python package clean. You can delete these environments if they become corrupt or stop working correctly due to any reason, without and fear.

这篇关于在windows上设置tensorflow的问题的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！