tensorflow - tf.keras.utils.to_categorical 在图形模式下引发 TypeError

Question

我有一个稀疏分类张量，我想将其转换为单热编码表示。我可以让它在急切模式下工作，但不能在图形模式下工作。我不明白这是怎么回事。

具体来说，假设我有一个稀疏分类

y=tf.constant(np.random.choice([0,1,2],2).reshape(-1,1))

我想将其转换为一种热编码表示形式。我定义了一个函数

def tmp(y):
    return tf.keras.utils.to_categorical(y)

它按预期工作。 但是，如果我包装函数

@tf.function
def tmp(y):
    return tf.keras.utils.to_categorical(y)

然后我得到异常

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-23-bb4411a7c6f7> in <module>
----> 1 tmp(y)

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py in __call__(self, *args, **kwds)
    578         xla_context.Exit()
    579     else:
--> 580       result = self._call(*args, **kwds)
    581 
    582     if tracing_count == self._get_tracing_count():

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py in _call(self, *args, **kwds)
    625       # This is the first call of __call__, so we have to initialize.
    626       initializers = []
--> 627       self._initialize(args, kwds, add_initializers_to=initializers)
    628     finally:
    629       # At this point we know that the initialization is complete (or less

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py in _initialize(self, args, kwds, add_initializers_to)
    504     self._concrete_stateful_fn = (
    505         self._stateful_fn._get_concrete_function_internal_garbage_collected(  # pylint: disable=protected-access
--> 506             *args, **kwds))
    507 
    508     def invalid_creator_scope(*unused_args, **unused_kwds):

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py in _get_concrete_function_internal_garbage_collected(self, *args, **kwargs)
   2444       args, kwargs = None, None
   2445     with self._lock:
-> 2446       graph_function, _, _ = self._maybe_define_function(args, kwargs)
   2447     return graph_function
   2448 

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py in _maybe_define_function(self, args, kwargs)
   2775 
   2776       self._function_cache.missed.add(call_context_key)
-> 2777       graph_function = self._create_graph_function(args, kwargs)
   2778       self._function_cache.primary[cache_key] = graph_function
   2779       return graph_function, args, kwargs

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py in _create_graph_function(self, args, kwargs, override_flat_arg_shapes)
   2665             arg_names=arg_names,
   2666             override_flat_arg_shapes=override_flat_arg_shapes,
-> 2667             capture_by_value=self._capture_by_value),
   2668         self._function_attributes,
   2669         # Tell the ConcreteFunction to clean up its graph once it goes out of

/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py in func_graph_from_py_func(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, override_flat_arg_shapes)
    979         _, original_func = tf_decorator.unwrap(python_func)
    980 
--> 981       func_outputs = python_func(*func_args, **func_kwargs)
    982 
    983       # invariant: `func_outputs` contains only Tensors, CompositeTensors,

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py in wrapped_fn(*args, **kwds)
    439         # __wrapped__ allows AutoGraph to swap in a converted function. We give
    440         # the function a weak reference to itself to avoid a reference cycle.
--> 441         return weak_wrapped_fn().__wrapped__(*args, **kwds)
    442     weak_wrapped_fn = weakref.ref(wrapped_fn)
    443 

/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
    966           except Exception as e:  # pylint:disable=broad-except
    967             if hasattr(e, "ag_error_metadata"):
--> 968               raise e.ag_error_metadata.to_exception(e)
    969             else:
    970               raise

TypeError: in user code:

    <ipython-input-11-4e328cd877a4>:3 tmp  *
        return tf.keras.utils.to_categorical(y)
    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/utils/np_utils.py:49 to_categorical  **
        y = np.array(y, dtype='int')

    TypeError: __array__() takes 1 positional argument but 2 were given

从表面上看，它看起来像一个 numpy 错误，但我验证了我的 numpy 安装有两个参数，所以出了什么问题？

更一般地说，在图形模式下将备用分类张量转换为热编码张量的方法是什么？

Answer 1

您可以使用 tf.one_hot 而不是 tf.keras.utils.to_categorical 作为解决方法。它在 eager 模式和图形模式下都按预期工作。

完整代码请引用下图

import tensorflow as tf
print(tf.__version__)
import numpy as np
y=tf.constant(np.random.choice([0,1,2],[2,2]))
print(y)

输出:

2.3.0
tf.Tensor(
[[1 0]
 [2 1]], shape=(2, 2), dtype=int64)

使用 tf.keras.utils.to_categorical 的 Eager 模式

def tmp(y):
    return tf.keras.utils.to_categorical(y)
tmp(y)

输出:

array([[[0., 1., 0.],
        [1., 0., 0.]],

       [[0., 0., 1.],
        [0., 1., 0.]]], dtype=float32)

使用 tf.one_hot 的 Eager 模式

def tmp(y):
    return tf.one_hot(y,3)
tmp(y)

输出:

<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[0., 1., 0.],
        [1., 0., 0.]],

       [[0., 0., 1.],
        [0., 1., 0.]]], dtype=float32)>

使用 tf.one_hot 的图表模式

@tf.function
def tmp(y):
    return tf.one_hot(y,3)
tmp(y)

输出:

<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[0., 1., 0.],
        [1., 0., 0.]],

       [[0., 0., 1.],
        [0., 1., 0.]]], dtype=float32)>