AttributeError：模块“ tensorflow.python.ops.linalg.linear_operator_util”没有属性“ matmul_with_broadcast”

Question

我正在尝试为TF代理创建自己的PyEnvironment。 但是，此错误不断出现：

  

AttributeError：模块'tensorflow.python.ops.linalg.linear_operator_util'没有属性'matmul_with_broadcast'

我发现tensorflow-probability似乎有问题，但是我安装了推荐的版本tensorflow-probability=0.7.0 https://github.com/tensorflow/agents/issues/91

我尝试重新安装和更新

tensorflow-gpu=2.0.0-beta1 tf-agents-nightly tensorflow-probability=0.7.0

这是一个最小的代码示例：

from tf_agents.environments import py_environment


class myEnv(py_environment.PyEnvironment):
    def __init__(self):
        pass

    def _reset(self):
        pass

    def _step(self, action):
        pass

这是运行此最小示例时的完整错误消息：

Traceback (most recent call last):   File ".\env_mwe.py", line 1, in <module>
    from tf_agents.environments import py_environment   File "C:\Python37\lib\site-packages\tf_agents\environments\__init__.py", line 18, in <module>
    from tf_agents.environments import batched_py_environment   File "C:\Python37\lib\site-packages\tf_agents\environments\batched_py_environment.py", line 33, in <module>
    from tf_agents.environments import py_environment   File "C:\Python37\lib\site-packages\tf_agents\environments\py_environment.py", line 29, in <module>
    from tf_agents.trajectories import time_step as ts   File "C:\Python37\lib\site-packages\tf_agents\trajectories\__init__.py", line 19, in <module>
    from tf_agents.trajectories import time_step   File "C:\Python37\lib\site-packages\tf_agents\trajectories\time_step.py", line 28, in <module>
    from tf_agents.specs import array_spec   File "C:\Python37\lib\site-packages\tf_agents\specs\__init__.py", line 20, in <module>
    from tf_agents.specs.distribution_spec import DistributionSpec   File "C:\Python37\lib\site-packages\tf_agents\specs\distribution_spec.py", line 22, in <module>
    import tensorflow_probability as tfp   File "E:\Users\tmp\AppData\Roaming\Python\Python37\site-packages\tensorflow_probability\__init__.py", line 78, in <module>
    from tensorflow_probability.python import *  # pylint: disable=wildcard-import   File "E:\Users\tmp\AppData\Roaming\Python\Python37\site-packages\tensorflow_probability\python\__init__.py", line 22, in <module>
    from tensorflow_probability.python import distributions   File "E:\Users\tmp\AppData\Roaming\Python\Python37\site-packages\tensorflow_probability\python\distributions\__init__.py", line 64, in <module>
    from tensorflow_probability.python.distributions.linear_gaussian_ssm import LinearGaussianStateSpaceModel   File "E:\Users\tmp\AppData\Roaming\Python\Python37\site-packages\tensorflow_probability\python\distributions\linear_gaussian_ssm.py", line 41, in <module>
    _matmul = linear_operator_util.matmul_with_broadcast AttributeError: module 'tensorflow.python.ops.linalg.linear_operator_util' has no attribute 'matmul_with_broadcast'

Answer 1

要使用//... // This rule should be familiar - it's standard part of espresso test @Rule public ActivityTestRule<(your activity)> mActivityRule = new ActivityTestRule<>((your activity class)); // ... // Utility invocation in test // start location updates - 25m/s (~55mph) LatLng startPos = new LatLng(39.0, -77.0); LocationUtils.startUpdates(mActivityRule.getActivity(), new Handler(Looper.getMainLooper()), startPos, 340, 25); // ... // (In a test utility class in this example: LocationUtils.java) // Utility - uses SphericalUtil to maintain a position based on // initial starting position, heading and movement value (in // meters) applied every 1 second. (So a movement value // of 25 equates to 25m/s which equates to ~55MPH) public static void startUpdates( final Activity activity, final Handler mHandler, final LatLng pos, final double heading, final double movement) { mHandler.postDelayed(new Runnable() { private LatLng myPos = new LatLng(pos.latitude,pos.longitude); @Override public void run() { Location mockLocation = new Location(LocationManager.GPS_PROVIDER); // a string mockLocation.setLatitude(myPos.latitude); // double mockLocation.setLongitude(myPos.longitude); mockLocation.setAltitude(100); mockLocation.setTime(System.currentTimeMillis()); mockLocation.setAccuracy(1); if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.JELLY_BEAN_MR1) { mockLocation.setElapsedRealtimeNanos(SystemClock.elapsedRealtimeNanos()); } LocationServices.getFusedLocationProviderClient(activity).setMockMode(true); LocationServices.getFusedLocationProviderClient(activity).setMockLocation(mockLocation); // compute next position myPos = SphericalUtil.computeOffset(myPos, movement, heading); mHandler.postDelayed(this, 1000); } }, 1000); } ，需要2.0.0beta1。

Answer 2

找到了！

要在$('#audio').attr("src","path_to_audio.wav"); $('#audio').play(); 中使用tensorflow-gpu，需要安装tensorflow-probability

因此tfp-nightly做到了