值错误：尝试将"y"转换为张量但失败。错误：不支持无值

Question

>不起作用：

from tensorflow.python.keras.layers import Input, Dense
from tensorflow.python.keras.models import Model
from tensorflow.python.keras.optimizers import Nadam
import numpy as np
ipt = Input(shape=(4,))
out = Dense(1, activation='sigmoid')(ipt)
model = Model(ipt, out)
model.compile(optimizer=Nadam(lr=1e-4), loss='binary_crossentropy')
X = np.random.randn(32,4)
Y = np.random.randint(0,2,(32,1))
model.train_on_batch(X,Y)

作品：从上面的导入中删除.python。

这是怎么回事，如何解决？

---

附加信息：

• CUDA 10.0.130， cuDNN 7.4.2， Python 3.7.4， Windows 10
• tensorflow，tensorflow-gpuv2.0.0 和 Keras 2.3.0 通过 pip，其他所有通过 Anaconda 3
• 根据 DEBUG 1，我注意到pip安装了r2.0分支而不是master;用master手动覆盖本地tensorflow_core.python文件夹会破坏一切 - 但对少数文件这样做不会，尽管错误仍然存在

---

调试 1：文件差异

这适用于我的本地安装，而不是 TF 的 Github 分支master或r2.0;由于某种原因，Github文件缺少api/_v2：

from tensorflow import keras
print(keras.__file__)
from tensorflow.python import keras
print(keras.__file__)

[1] D:Anacondaenvstf2_envlibsite-packagestensorflow_corepythonkerasapi_v2keras__init__.py
[2] D:Anacondaenvstf2_envlibsite-packagestensorflow_corepythonkeras__init__.py

查看每个__init__以获取Optimizer：

# [1]
from tensorflow.python.keras.optimizer_v2.optimizer_v2 import OptimizerV2 as Optimizer
# [2]
from tensorflow.python.keras import optimizers
# in python.keras.optimizers.py:
# all imports are from tensorflow.python
class Optimizer(object): # <--- does NOT use optimizer_v2 for Optimizer

这似乎解决了问题的根源，如下所示：

from tensorflow.python.keras.layers import Input, Dense
from tensorflow.python.keras.models import Model
from tensorflow.keras.optimizers import Nadam

然而，这很奇怪，因为直接import keras也不使用optimizer_v2，尽管keras.optimizers中Optimizer的定义确实有所不同。

---

调试 2：执行差异

并行调试，虽然两者都使用相同的 training.py，但执行差异相当快：

### TF.KERAS
if self._experimental_run_tf_function: #  TRUE
### TF.PYTHON.KERAS
if self._experimental_run_tf_function: #  FALSE

前者继续打电话给training_v2_utils.train_on_batch(...)，然后返回，后者self._standardize_user_data(...)和其他人最终失败。

---

调试 3(+ 解决方案？(：故障线

if None in grads: # <-- in traceback

在其正上方插入print(None in grads)会产生完全相同的错误 - 因此，它似乎与 TF2 可迭代操作有关 - 这有效：

if any([g is None for g in grads]): # <-- works; similar but not equivalent Python logic

还不确定它是否是一个完整的修复，仍在调试 -更新：启动了 Github 拉取请求

---

完整的错误跟踪：

File "<ipython-input-1-2db039c052cf>", line 20, in <module>
model.train_on_batch(X,Y)
File "D:Anacondaenvstf2_envlibsite-packagestensorflow_corepythonkerasenginetraining.py", line 1017, in train_on_batch
self._make_train_function()
File "D:Anacondaenvstf2_envlibsite-packagestensorflow_corepythonkerasenginetraining.py", line 2116, in _make_train_function
params=self._collected_trainable_weights, loss=self.total_loss)
File "D:Anacondaenvstf2_envlibsite-packagestensorflow_corepythonkerasoptimizers.py", line 653, in get_updates
grads = self.get_gradients(loss, params)
File "D:Anacondaenvstf2_envlibsite-packagestensorflow_corepythonkerasoptimizers.py", line 92, in get_gradients
if None in grads:
File "D:Anacondaenvstf2_envlibsite-packagestensorflow_corepythonopsmath_ops.py", line 1336, in tensor_equals
return gen_math_ops.equal(self, other)
File "D:Anacondaenvstf2_envlibsite-packagestensorflow_corepythonopsgen_math_ops.py", line 3626, in equal
name=name)
File "D:Anacondaenvstf2_envlibsite-packagestensorflow_corepythonframeworkop_def_library.py", line 545, in _apply_op_helper
(input_name, err))
ValueError: Tried to convert 'y' to a tensor and failed. Error: None values not supported.

Answer 1

这是一个错误，我的拉取请求修复已获得批准(但尚未合并(。同时，您可以手动进行更改，如下所示。此外，tf.python.keras并不总是被使用，如果有的话。

更新：拉取请求现已合并。

---

为什么有效：None in grads与any(g == None for g in grads)相同;问题是，g可能是tf。张量/tf..__eq__定义的变量仅对张量进行操作，因此必须改用is None。

from tensorflow.keras.layers import Input, Dense
from tensorflow.keras.models import Model
import numpy as np
ipt = Input((16,))
out = Dense(16)(ipt)
model = Model(ipt, out)
model.compile('adam', 'mse')
x = y = np.random.randn(32, 16)
model.train_on_batch(x, y)
W = model.optimizer.weights
W[0] == None

>>> ValueError: Attempt to convert a value (None) with an unsupported type 
(<class 'NoneType'>) to a Tensor.

检查源代码：

from inspect import getsource
print(getsource(W[0].__eq__))

def __eq__(self, other):
"""Compares two variables element-wise for equality."""
if ops.Tensor._USE_EQUALITY and ops.executing_eagerly_outside_functions():
return gen_math_ops.equal(self, other, incompatible_shape_error=False)
else:
# In legacy graph mode, tensor equality is object equality
return self is other

Answer 2

也许你应该更正你的导入

from tensorflow.keras.layers import Input, Dense
from tensorflow.keras.models import Model
from tensorflow.keras.optimizers import Nadam