我正在尝试运行tensorflow文档中的示例。然而，我得到以下错误：
TypeError: Tensor is unhashable. Instead, use tensor.ref() as the key.
我在下面添加了
- 回溯
conda list的输出conda info的输出
回溯
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
/var/folders/hy/57_f4xcx08b0ls2_nwwfjscr0000gn/T/ipykernel_19414/3526934965.py in <module>
6
7 # Get 1000 states from one chain.
----> 8 states = tfp.mcmc.sample_chain(
9 num_burnin_steps=200,
10 num_results=1000,
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/mcmc/sample.py in sample_chain(num_results, current_state, previous_kernel_results, kernel, num_burnin_steps, num_steps_between_results, trace_fn, return_final_kernel_results, parallel_iterations, name)
324 current_state)
325 if previous_kernel_results is None:
--> 326 previous_kernel_results = kernel.bootstrap_results(current_state)
327
328 if trace_fn is None:
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/mcmc/hmc.py in bootstrap_results(self, init_state)
558 def bootstrap_results(self, init_state):
559 """Creates initial `previous_kernel_results` using a supplied `state`."""
--> 560 kernel_results = self._impl.bootstrap_results(init_state)
561 if self.step_size_update_fn is not None:
562 step_size_assign = self.step_size_update_fn(self.step_size, None) # pylint: disable=not-callable
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/mcmc/metropolis_hastings.py in bootstrap_results(self, init_state)
263 name=mcmc_util.make_name(self.name, 'mh', 'bootstrap_results'),
264 values=[init_state]):
--> 265 pkr = self.inner_kernel.bootstrap_results(init_state)
266 if not has_target_log_prob(pkr):
267 raise ValueError(
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/mcmc/hmc.py in bootstrap_results(self, init_state)
770 init_target_log_prob,
771 init_grads_target_log_prob,
--> 772 ] = mcmc_util.maybe_call_fn_and_grads(self.target_log_prob_fn, init_state)
773 if self._store_parameters_in_results:
774 return UncalibratedHamiltonianMonteCarloKernelResults(
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/mcmc/internal/util.py in maybe_call_fn_and_grads(fn, fn_arg_list, result, grads, check_non_none_grads, name)
231 fn_arg_list = (list(fn_arg_list) if is_list_like(fn_arg_list)
232 else [fn_arg_list])
--> 233 result, grads = _value_and_gradients(fn, fn_arg_list, result, grads)
234 if not all(r.dtype.is_floating
235 for r in (result if is_list_like(result) else [result])): # pylint: disable=superfluous-parens
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/mcmc/internal/util.py in _value_and_gradients(fn, fn_arg_list, result, grads, name)
190
191 if result is None:
--> 192 result = fn(*fn_arg_list)
193 if grads is None and tf.executing_eagerly():
194 # Ensure we disable bijector cacheing in eager mode.
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/distributions/distribution.py in log_prob(self, value, name, **kwargs)
864 values of type `self.dtype`.
865 """
--> 866 return self._call_log_prob(value, name, **kwargs)
867
868 def _call_prob(self, value, name, **kwargs):
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/distributions/distribution.py in _call_log_prob(self, value, name, **kwargs)
846 value, name="value", dtype_hint=self.dtype)
847 if hasattr(self, "_log_prob"):
--> 848 return self._log_prob(value, **kwargs)
849 if hasattr(self, "_prob"):
850 return tf.math.log(self._prob(value, **kwargs))
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/internal/distribution_util.py in _fn(*args, **kwargs)
2092 @functools.wraps(fn)
2093 def _fn(*args, **kwargs):
-> 2094 return fn(*args, **kwargs)
2095
2096 if _fn.__doc__ is None:
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/distributions/mvn_linear_operator.py in _log_prob(self, x)
208 @distribution_util.AppendDocstring(_mvn_sample_note)
209 def _log_prob(self, x):
--> 210 return super(MultivariateNormalLinearOperator, self)._log_prob(x)
211
212 @distribution_util.AppendDocstring(_mvn_sample_note)
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/distributions/transformed_distribution.py in _log_prob(self, y, **kwargs)
399 # For caching to work, it is imperative that the bijector is the first to
400 # modify the input.
--> 401 x = self.bijector.inverse(y, **bijector_kwargs)
402 event_ndims = self._maybe_get_static_event_ndims()
403
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/bijectors/bijector.py in inverse(self, y, name, **kwargs)
975 NotImplementedError: if `_inverse` is not implemented.
976 """
--> 977 return self._call_inverse(y, name, **kwargs)
978
979 def _compute_inverse_log_det_jacobian_with_caching(
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/bijectors/bijector.py in _call_inverse(self, y, name, **kwargs)
944 if not self._is_injective: # No caching for non-injective
945 return self._inverse(y, **kwargs)
--> 946 mapping = self._lookup(y=y, kwargs=kwargs)
947 if mapping.x is not None:
948 return mapping.x
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/bijectors/bijector.py in _lookup(self, x, y, kwargs)
1344 if y is not None:
1345 # We removed y at caching time. Add it back if we lookup successfully.
-> 1346 mapping = self._from_y[y].get(subkey, mapping).merge(y=y)
1347 return mapping
1348
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/bijectors/bijector.py in __getitem__(self, key)
149 def __getitem__(self, key):
150 weak_key = HashableWeakRef(key, lambda w: self.pop(w, None))
--> 151 return super(WeakKeyDefaultDict, self).__getitem__(weak_key)
152
153 # This is the "DefaultDict" part.
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow_probability/python/bijectors/bijector.py in __hash__(self)
179 x = self()
180 if not isinstance(x, np.ndarray):
--> 181 return hash(x)
182 # Note: The following logic can never be reached by the public API because
183 # the bijector base class always calls `convert_to_tensor` before accessing
~/miniforge3/envs/tf_env/lib/python3.8/site-packages/tensorflow/python/framework/ops.py in __hash__(self)
828 if (Tensor._USE_EQUALITY and executing_eagerly_outside_functions() and
829 (g is None or g.building_function)):
--> 830 raise TypeError("Tensor is unhashable. "
831 "Instead, use tensor.ref() as the key.")
832 else:
TypeError: Tensor is unhashable. Instead, use tensor.ref() as the key.
conda列表
# packages in environment at /Users/maurocamara/miniforge3/envs/tf_env:
#
# Name Version Build Channel
absl-py 0.13.0 pyhd8ed1ab_0 conda-forge
aiohttp 3.7.4.post0 py38hea4295b_0 conda-forge
anyio 3.2.1 py38h10201cd_0 conda-forge
appnope 0.1.2 py38h10201cd_1 conda-forge
argon2-cffi 20.1.0 py38hea4295b_2 conda-forge
astor 0.8.1 pyh9f0ad1d_0 conda-forge
astunparse 1.6.3 pyhd8ed1ab_0 conda-forge
async-timeout 3.0.1 py_1000 conda-forge
async_generator 1.10 py_0 conda-forge
attrs 21.2.0 pyhd8ed1ab_0 conda-forge
babel 2.9.1 pyh44b312d_0 conda-forge
backcall 0.2.0 pyh9f0ad1d_0 conda-forge
backports 1.0 py_2 conda-forge
backports.functools_lru_cache 1.6.4 pyhd8ed1ab_0 conda-forge
bleach 3.3.1 pyhd8ed1ab_0 conda-forge
blinker 1.4 py_1 conda-forge
brotlipy 0.7.0 py38hea4295b_1001 conda-forge
c-ares 1.17.1 h27ca646_1 conda-forge
ca-certificates 2021.5.30 h4653dfc_0 conda-forge
cached-property 1.5.2 hd8ed1ab_1 conda-forge
cached_property 1.5.2 pyha770c72_1 conda-forge
cachetools 4.2.2 pyhd8ed1ab_0 conda-forge
certifi 2021.5.30 py38h10201cd_0 conda-forge
cffi 1.14.6 py38h0957451_0 conda-forge
chardet 4.0.0 py38h10201cd_1 conda-forge
charset-normalizer 2.0.0 pyhd8ed1ab_0 conda-forge
click 8.0.1 py38h10201cd_0 conda-forge
cloudpickle 1.6.0 py_0 conda-forge
cryptography 3.4.7 py38h3c0dae5_0 conda-forge
cycler 0.10.0 py_2 conda-forge
dataclasses 0.8 pyhc8e2a94_1 conda-forge
debugpy 1.3.0 py38h6f2b01f_0 conda-forge
decorator 5.0.9 pyhd8ed1ab_0 conda-forge
defusedxml 0.7.1 pyhd8ed1ab_0 conda-forge
entrypoints 0.3 pyhd8ed1ab_1003 conda-forge
freetype 2.10.4 h17b34a0_1 conda-forge
gast 0.5.0 pyhd8ed1ab_0 conda-forge
google-auth 1.33.1 pyh6c4a22f_0 conda-forge
google-auth-oauthlib 0.4.1 py_2 conda-forge
google-pasta 0.2.0 pyh8c360ce_0 conda-forge
grpcio 1.38.1 py38h69ee544_0 conda-forge
h5py 3.3.0 nompi_py38hb525b2d_100 conda-forge
hdf5 1.10.6 nompi_h0fc092c_1114 conda-forge
idna 3.1 pyhd3deb0d_0 conda-forge
importlib-metadata 4.6.1 py38h10201cd_0 conda-forge
ipykernel 6.0.3 py38h2cb4d76_0 conda-forge
ipython 7.25.0 py38h2cb4d76_1 conda-forge
ipython_genutils 0.2.0 py_1 conda-forge
ipywidgets 7.6.3 pyhd3deb0d_0 conda-forge
jbig 2.1 h3422bc3_2003 conda-forge
jedi 0.18.0 py38h10201cd_2 conda-forge
jinja2 3.0.1 pyhd8ed1ab_0 conda-forge
jpeg 9d h27ca646_0 conda-forge
json5 0.9.5 pyh9f0ad1d_0 conda-forge
jsonschema 3.2.0 pyhd8ed1ab_3 conda-forge
jupyter 1.0.0 py38h10201cd_6 conda-forge
jupyter_client 6.1.7 py_0 anaconda
jupyter_console 6.4.0 pyhd8ed1ab_0 conda-forge
jupyter_core 4.7.1 py38h10201cd_0 conda-forge
jupyter_server 1.9.0 pyhd8ed1ab_0 conda-forge
jupyterlab 3.0.16 pyhd8ed1ab_0 conda-forge
jupyterlab_pygments 0.1.2 pyh9f0ad1d_0 conda-forge
jupyterlab_server 2.6.1 pyhd8ed1ab_0 conda-forge
jupyterlab_widgets 1.0.0 pyhd8ed1ab_1 conda-forge
keras-preprocessing 1.1.2 pyhd8ed1ab_0 conda-forge
kiwisolver 1.3.1 py38h1670459_1 conda-forge
krb5 1.19.1 hd92b7a7_0 conda-forge
lcms2 2.12 had6a04f_0 conda-forge
lerc 2.2.1 h9f76cd9_0 conda-forge
libblas 3.9.0 9_openblas conda-forge
libcblas 3.9.0 9_openblas conda-forge
libcurl 7.77.0 h8fe1914_0 conda-forge
libcxx 12.0.1 h168391b_0 conda-forge
libdeflate 1.7 h27ca646_5 conda-forge
libedit 3.1.20191231 hc8eb9b7_2 conda-forge
libev 4.33 h642e427_1 conda-forge
libffi 3.3 h9f76cd9_2 conda-forge
libgfortran 5.0.0.dev0 11_0_1_hf114ba7_22 conda-forge
libgfortran5 11.0.1.dev0 hf114ba7_22 conda-forge
liblapack 3.9.0 9_openblas conda-forge
libnghttp2 1.43.0 hf3018f0_0 conda-forge
libopenblas 0.3.15 openmp_hf330de4_1 conda-forge
libpng 1.6.37 hf7e6567_2 conda-forge
libprotobuf 3.17.2 hccf11d3_1 conda-forge
libsodium 1.0.18 h27ca646_1 conda-forge
libssh2 1.9.0 hb80f160_6 conda-forge
libtiff 4.3.0 hc6122e1_1 conda-forge
libwebp-base 1.2.0 h27ca646_2 conda-forge
llvm-openmp 12.0.1 hf3c4609_0 conda-forge
lz4-c 1.9.3 h9f76cd9_0 conda-forge
markdown 3.3.4 pyhd8ed1ab_0 conda-forge
markupsafe 2.0.1 py38hea4295b_0 conda-forge
matplotlib 3.4.2 py38h150bfb4_0 conda-forge
matplotlib-base 3.4.2 py38hb140015_0 conda-forge
matplotlib-inline 0.1.2 pyhd8ed1ab_2 conda-forge
mistune 0.8.4 py38hea4295b_1004 conda-forge
multidict 5.1.0 py38hea4295b_1 conda-forge
nbclassic 0.3.1 pyhd8ed1ab_1 conda-forge
nbclient 0.5.3 pyhd8ed1ab_0 conda-forge
nbconvert 6.1.0 py38h10201cd_0 conda-forge
nbformat 5.1.3 pyhd8ed1ab_0 conda-forge
ncurses 6.2 h9aa5885_4 conda-forge
nest-asyncio 1.5.1 pyhd8ed1ab_0 conda-forge
notebook 6.4.0 pyha770c72_0 conda-forge
numpy 1.19.5 py38hbf7bb01_2 conda-forge
oauthlib 3.1.1 pyhd8ed1ab_0 conda-forge
olefile 0.46 pyh9f0ad1d_1 conda-forge
openjpeg 2.4.0 h062765e_1 conda-forge
openssl 1.1.1k h27ca646_0 conda-forge
opt_einsum 3.3.0 pyhd8ed1ab_1 conda-forge
packaging 21.0 pyhd8ed1ab_0 conda-forge
pandocfilters 1.4.2 py_1 conda-forge
parso 0.8.2 pyhd8ed1ab_0 conda-forge
pexpect 4.8.0 pyh9f0ad1d_2 conda-forge
pickleshare 0.7.5 py_1003 conda-forge
pillow 8.3.1 py38h02acf36_0 conda-forge
pip 20.2.4 py_0 conda-forge
plotly 5.1.0 pyhd8ed1ab_1 conda-forge
prometheus_client 0.11.0 pyhd8ed1ab_0 conda-forge
prompt-toolkit 3.0.19 pyha770c72_0 conda-forge
prompt_toolkit 3.0.19 hd8ed1ab_0 conda-forge
protobuf 3.17.2 py38h6f2b01f_0 conda-forge
ptyprocess 0.7.0 pyhd3deb0d_0 conda-forge
pyasn1 0.4.8 py_0 conda-forge
pyasn1-modules 0.2.7 py_0 conda-forge
pycparser 2.20 pyh9f0ad1d_2 conda-forge
pygments 2.9.0 pyhd8ed1ab_0 conda-forge
pyjwt 2.1.0 pyhd8ed1ab_0 conda-forge
pyopenssl 20.0.1 pyhd8ed1ab_0 conda-forge
pyparsing 2.4.7 pyh9f0ad1d_0 conda-forge
pyrsistent 0.17.3 py38hea4295b_2 conda-forge
pysocks 1.7.1 py38h10201cd_3 conda-forge
python 3.8.10 hf9733c0_1_cpython conda-forge
python-dateutil 2.8.1 py_0 anaconda
python-flatbuffers 2.0 pyhd8ed1ab_0 conda-forge
python_abi 3.8 2_cp38 conda-forge
pytz 2021.1 pyhd8ed1ab_0 conda-forge
pyu2f 0.1.5 pyhd8ed1ab_0 conda-forge
pyzmq 22.1.0 py38h51b17a6_0 conda-forge
readline 8.1 hedafd6a_0 conda-forge
requests 2.26.0 pyhd8ed1ab_0 conda-forge
requests-oauthlib 1.3.0 pyh9f0ad1d_0 conda-forge
requests-unixsocket 0.2.0 py_0 conda-forge
rsa 4.7.2 pyh44b312d_0 conda-forge
scipy 1.7.0 py38hd0c9ec0_0 conda-forge
send2trash 1.7.1 pyhd8ed1ab_0 conda-forge
setuptools 49.6.0 py38h10201cd_3 conda-forge
six 1.16.0 pyh6c4a22f_0 conda-forge
sniffio 1.2.0 py38h10201cd_1 conda-forge
sqlite 3.36.0 h72a2b83_0 conda-forge
tenacity 8.0.1 pyhd8ed1ab_0 conda-forge
tensorboard 2.4.1 pyhd8ed1ab_0 conda-forge
tensorboard-plugin-wit 1.8.0 pyh44b312d_0 conda-forge
tensorflow 2.4.0rc0 pypi_0 pypi
tensorflow-addons 0.11.2 pypi_0 pypi
tensorflow-estimator 2.5.0 pyh81a9013_1 conda-forge
tensorflow-probability 0.7 py_1 conda-forge
termcolor 1.1.0 py_2 conda-forge
terminado 0.10.1 py38h10201cd_0 conda-forge
testpath 0.5.0 pyhd8ed1ab_0 conda-forge
tk 8.6.10 hf7e6567_1 conda-forge
tornado 6.1 py38hea4295b_1 conda-forge
traitlets 5.0.5 py_0 conda-forge
typeguard 2.12.1 pyhd8ed1ab_0 conda-forge
typing-extensions 3.10.0.0 hd8ed1ab_0 conda-forge
typing_extensions 3.10.0.0 pyha770c72_0 conda-forge
urllib3 1.26.6 pyhd8ed1ab_0 conda-forge
wcwidth 0.2.5 pyh9f0ad1d_2 conda-forge
webencodings 0.5.1 py_1 conda-forge
websocket-client 0.57.0 py38h10201cd_4 conda-forge
werkzeug 2.0.1 pyhd8ed1ab_0 conda-forge
wheel 0.36.2 pyhd3deb0d_0 conda-forge
widgetsnbextension 3.5.1 py38h10201cd_4 conda-forge
wrapt 1.12.1 py38hea4295b_3 conda-forge
xz 5.2.5 h642e427_1 conda-forge
yarl 1.6.3 py38hea4295b_2 conda-forge
zeromq 4.3.4 h9f76cd9_0 conda-forge
zipp 3.5.0 pyhd8ed1ab_0 conda-forge
zlib 1.2.11 h31e879b_1009 conda-forge
zstd 1.5.0 h861e0a7_0 conda-forge
conda信息
active environment : tf_env
active env location : /Users/maurocamara/miniforge3/envs/tf_env
shell level : 1
user config file : /Users/maurocamara/.condarc
populated config files : /Users/maurocamara/miniforge3/.condarc
/Users/maurocamara/.condarc
conda version : 4.10.3
conda-build version : not installed
python version : 3.9.6.final.0
virtual packages : __osx=11.2.3=0
__unix=0=0
__archspec=1=arm64
base environment : /Users/maurocamara/miniforge3 (writable)
conda av data dir : /Users/maurocamara/miniforge3/etc/conda
conda av metadata url : None
channel URLs : https://conda.anaconda.org/conda-forge/osx-arm64
https://conda.anaconda.org/conda-forge/noarch
package cache : /Users/maurocamara/miniforge3/pkgs
/Users/maurocamara/.conda/pkgs
envs directories : /Users/maurocamara/miniforge3/envs
/Users/maurocamara/.conda/envs
platform : osx-arm64
user-agent : conda/4.10.3 requests/2.26.0 CPython/3.9.6 Darwin/20.3.0 OSX/11.2.3
UID:GID : 501:20
netrc file : None
offline mode : False
当我尝试在conda环境中复制相同的代码时，这不会显示任何错误。
根据此已测试的内部版本配置，可能存在版本不兼容问题，因为回溯错误显示Python 3.8，但conda信息显示您在conda中安装了Python 3.9.6。请通过将TensorFlow升级到最新版本重试。
pip install --upgrade tensorflow
请在此处检查复制的代码：
!pip install tensorflow_probability
import tensorflow as tf
import tensorflow_probability as tfp
tfd = tfp.distributions
target = tfd.MultivariateNormalDiag(scale_diag=[1., 2.])
# Get 1000 states from one chain.
states = tfp.mcmc.sample_chain(
num_burnin_steps=200,
num_results=1000,
current_state=tf.constant([0., 0.]),
trace_fn=None,
kernel=tfp.mcmc.HamiltonianMonteCarlo(
target_log_prob_fn=target.log_prob,
step_size=0.05,
num_leapfrog_steps=20))
print(states.shape)
#==> (1000, 2)
ess = tfp.mcmc.effective_sample_size(states, filter_beyond_positive_pairs=True)
print(ess.shape)
#==> Shape (2,) Tensor
mean, variance = tf.nn.moments(states, axes=0)
standard_error = tf.sqrt(variance / ess)
输出：
(1000, 2)
(2,)