python - Tensorflow 2.0 - SparseCategoricalCrossentropy 的数据形状不正确

Question

使用SparseCategoricalCrossentropy损失时，我不了解有关数据集形状不匹配的错误。根据SparseCategoricalCrossentropy
，输出似乎是正确的形式： 输入：[7 1] 输出：[7 3] 其中批量大小为 7。

我的环境当前运行：
python 3.7.4
tensorflow 2.0.0-rc0
numpy 1.17.2
pandas 0.24.2

---------------------------------------------------------------------------
InvalidArgumentError                      Traceback (most recent call last)
<ipython-input-5-fe1cc3e1dca1> in <module>
     47 print(model.summary())
     48 
---> 49 model.fit(train.batch(BATCH_SIZE), epochs=EPOCHS, verbose=2)
     50 model.evaluate(train, steps=None, verbose=1)

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)
    732         max_queue_size=max_queue_size,
    733         workers=workers,
--> 734         use_multiprocessing=use_multiprocessing)
    735 
    736   def evaluate(self,

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training_v2.py in fit(self, model, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, **kwargs)
    322                 mode=ModeKeys.TRAIN,
    323                 training_context=training_context,
--> 324                 total_epochs=epochs)
    325             cbks.make_logs(model, epoch_logs, training_result, ModeKeys.TRAIN)
    326 

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training_v2.py in run_one_epoch(model, iterator, execution_function, dataset_size, batch_size, strategy, steps_per_epoch, num_samples, mode, training_context, total_epochs)
    121         step=step, mode=mode, size=current_batch_size) as batch_logs:
    122       try:
--> 123         batch_outs = execution_function(iterator)
    124       except (StopIteration, errors.OutOfRangeError):
    125         # TODO(kaftan): File bug about tf function and errors.OutOfRangeError?

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training_v2_utils.py in execution_function(input_fn)
     84     # `numpy` translates Tensors to values in Eager mode.
     85     return nest.map_structure(_non_none_constant_value,
---> 86                               distributed_function(input_fn))
     87 
     88   return execution_function

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/eager/def_function.py in __call__(self, *args, **kwds)
    437         # Lifting succeeded, so variables are initialized and we can run the
    438         # stateless function.
--> 439         return self._stateless_fn(*args, **kwds)
    440     else:
    441       canon_args, canon_kwds = \

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in __call__(self, *args, **kwargs)
   1820     """Calls a graph function specialized to the inputs."""
   1821     graph_function, args, kwargs = self._maybe_define_function(args, kwargs)
-> 1822     return graph_function._filtered_call(args, kwargs)  # pylint: disable=protected-access
   1823 
   1824   @property

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in _filtered_call(self, args, kwargs)
   1139          if isinstance(t, (ops.Tensor,
   1140                            resource_variable_ops.BaseResourceVariable))),
-> 1141         self.captured_inputs)
   1142 
   1143   def _call_flat(self, args, captured_inputs, cancellation_manager=None):

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in _call_flat(self, args, captured_inputs, cancellation_manager)
   1222     if executing_eagerly:
   1223       flat_outputs = forward_function.call(
-> 1224           ctx, args, cancellation_manager=cancellation_manager)
   1225     else:
   1226       gradient_name = self._delayed_rewrite_functions.register()

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in call(self, ctx, args, cancellation_manager)
    509               inputs=args,
    510               attrs=("executor_type", executor_type, "config_proto", config),
--> 511               ctx=ctx)
    512         else:
    513           outputs = execute.execute_with_cancellation(

/usr/local/lib/python3.7/site-packages/tensorflow_core/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name)
     65     else:
     66       message = e.message
---> 67     six.raise_from(core._status_to_exception(e.code, message), None)
     68   except TypeError as e:
     69     keras_symbolic_tensors = [

~/Library/Python/3.7/lib/python/site-packages/six.py in raise_from(value, from_value)

InvalidArgumentError:  assertion failed: [] [Condition x == y did not hold element-wise:] [x (loss/dense_3_loss/SparseSoftmaxCrossEntropyWithLogits/Shape_1:0) = ] [7 1] [y (loss/dense_3_loss/SparseSoftmaxCrossEntropyWithLogits/strided_slice:0) = ] [7 3]
     [[node loss/dense_3_loss/SparseSoftmaxCrossEntropyWithLogits/assert_equal/Assert/Assert (defined at /usr/local/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py:1751) ]] [Op:__inference_distributed_function_2031]

Function call stack:
distributed_function

使用 tensorflow=2.0.0-beta0：

InvalidArgumentError:  logits and labels must have the same first dimension, got logits shape [34,3] and labels shape [2]
     [[node loss/dense_2_loss/SparseSoftmaxCrossEntropyWithLogits/SparseSoftmaxCrossEntropyWithLogits (defined at <ipython-input-3-4fa88a5fad5e>:70) ]] [Op:__inference_keras_scratch_graph_5038]

有人可以解释或指出我正确的方向吗？

谢谢

以下代码重现了该错误：

import numpy as np
import pandas as pd
import random
import tensorflow as tf

INPUT_SHAPE=[3, 5]
NUM_POINTS=20
BATCH_SIZE=7
EPOCHS=4

def data_gen(num=10, in_shape=[5, 3]):
    for i in range(num):
        x = np.random.rand(in_shape[0], in_shape[1])
        y = random.randint(0,2)
        yield x, y

train = tf.data.Dataset.from_generator(
    generator=data_gen,
    output_types=(tf.float32, tf.int32),
    args=([NUM_POINTS, INPUT_SHAPE])
)

def create_model(input_shape):
    model = tf.keras.models.Sequential([
        tf.keras.layers.Dense(100, activation="tanh",input_shape=input_shape),        
        tf.keras.layers.Dense(3, activation="softmax", kernel_regularizer= tf.keras.regularizers.l2(0.001))
    ])
    return model

model = create_model(input_shape=INPUT_SHAPE)

model.compile(
    optimizer=tf.keras.optimizers.Adam(learning_rate=1e-4, clipvalue=1.0),
    loss= tf.keras.losses.SparseCategoricalCrossentropy())
print(model.summary())

model.fit(train.batch(BATCH_SIZE), epochs=EPOCHS, verbose=2)
model.evaluate(train, steps=None, verbose=1)

型号总结：

模型：“sequential_1” _________________________________________________________________ 层（类型）输出形状参数#
==================================== ============================= dense_2（密集）（无，3、100）600
_________________________________________________________________密集_3（密集）（无，3 , 3) 303
============================================== =================== 总参数：903 可训练参数：903 不可训练参数：0

---

标签： pythontensorflowkeras