在 anaconda 上使用tensorflow 2.6.0设置tensorflow_io

Question

我是 TensorFlow 新手，我正在尝试在本地计算机上设置 ML 平台，并且需要将一些音频文件作为数据提供给我的神经网络。为此我需要 TensorFlow-io。

但是当我使用 anaconda navigator (2.1.2) 设置 TensorFlow 时，我不断遇到以下错误。

---------------------------------------------------------------------------
AlreadyExistsError                        Traceback (most recent call last)
~\AppData\Local\Temp\ipykernel_2820\3937949001.py in <module>
      1 import tensorflow as tf
----> 2 import tensorflow_io as tfio
      3 
      4 
      5 

E:\aconda\envs\p37env\lib\site-packages\tensorflow_io\__init__.py in <module>
     15 """tensorflow_io"""
     16 
---> 17 from tensorflow_io.python.api import *  # pylint: disable=wildcard-import
     18 from tensorflow_io.python.api.version import VERSION as __version__

E:\aconda\envs\p37env\lib\site-packages\tensorflow_io\python\api\__init__.py in <module>
     17 
     18 # tensorflow_io.core.python.ops is implicitly imported (along with file system)
---> 19 from tensorflow_io.python.ops.io_dataset import IODataset
     20 from tensorflow_io.python.ops.io_tensor import IOTensor
     21 

E:\aconda\envs\p37env\lib\site-packages\tensorflow_io\python\ops\__init__.py in <module>
     94 core_ops = LazyLoader("core_ops", "libtensorflow_io.so")
     95 try:
---> 96     plugin_ops = _load_library("libtensorflow_io_plugins.so", "fs")
     97 except NotImplementedError as e:
     98     warnings.warn("unable to load libtensorflow_io_plugins.so: {}".format(e))

E:\aconda\envs\p37env\lib\site-packages\tensorflow_io\python\ops\__init__.py in _load_library(filename, lib)
     62     for f in filenames:
     63         try:
---> 64             l = load_fn(f)
     65             if l is not None:
     66                 return l

E:\aconda\envs\p37env\lib\site-packages\tensorflow_io\python\ops\__init__.py in <lambda>(f)
     54         load_fn = lambda f: ctypes.CDLL(f, mode=ctypes.RTLD_GLOBAL)
     55     elif lib == "fs":
---> 56         load_fn = lambda f: tf.experimental.register_filesystem_plugin(f) is None
     57     else:
     58         load_fn = lambda f: tf.compat.v1.load_file_system_library(f) is None

E:\aconda\envs\p37env\lib\site-packages\tensorflow\python\framework\load_library.py in register_filesystem_plugin(plugin_location)
    216   """
    217   if os.path.exists(plugin_location):
--> 218     py_tf.TF_RegisterFilesystemPlugin(plugin_location)
    219 
    220   else:

AlreadyExistsError: File system for az already registered

我的 python 版本是 3.9，我根据其文档安装 tfio v 0.20.0 (https:// pypi.org/project/tensorflow-io/）

我使用 pip 将tensorflow安装为 - > pip安装tensorflow_io==0.20.0

Answer 1

以下对我有用，

使用 python 3.9.7

install tf-gpu

创建 conda env conda install tensorflow-gpu

这将使用 gpu

install tensorflow_io 和 pip

安装tensorflow 2.6.0 pip install tensorflow_io== 0.20.0

升级tensorflow gpu到2.6.3版本

pip install张量流GPU==2.6.3