我正在使用LSTM和带有tensorflow后端的keras学习NER。 Tensorflow == 2.20r4和keras == 2.3.1。
我一直在研究一些示例，但是遇到了一个我似乎无法弄清的错误。错误消息是：ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type int)."
我已将列表输入转换为Numpy数组，如下所示：X = np.asarray(X), Y = np.asarray(Y)
然后我使用from tensorflow.keras.utils import to_categorical对Y进行编码。前置编码和后置编码
形状以及模型说明如下：
数据处理
# print pre-encoding x_train, y_train shapes
print(X_train.shape, y_train.shape)
(8296, 1302) (8296, 1302)
# convert y_train to_categorical
y_train = to_categorical(y_train)
# print post-encoding x_train, y_train shapes
print(X_train.shape, y_train.shape)
(8296, 1302) (8296, 1302, 32)
# print pre-encoding x_test, y_test shapes
print(X_test.shape, y_test.shape)
(1464, 1302) (1464, 1302)
# convert y_test to categorical
y_test = to_categorical(y_test)
# print post-encoding X_test, y_test
print(X_test.shape, y_test.shape)
(1464, 1302) (1464, 1302, 32)
模型声明
input = Input(shape=(max_len,))
model = Embedding(input_dim=num_words, output_dim=50, input_length=max_len)(input)
model = SpatialDropout1D(0.1)(model)
model = Bidirectional(LSTM(units=100, return_sequences=True, recurrent_dropout=0.1))(model)
output = TimeDistributed(Dense(num_tags, activation="softmax"))(model)
model = Model(input, output)
model.summary()
model.compile(optimizer="adam",
loss="categorical_crossentropy",
metrics=["accuracy"])
训练模型
checkpointer = ModelCheckpoint(filepath = 'model.h5',
verbose = 0,
mode = 'auto',
save_best_only = True,
monitor='val_loss')
# the model
early_stopping = EarlyStopping(monitor='val_accuracy', patience=1, verbose=0, mode='max', restore_best_weights=False)
callbacks =[checkpointer, early_stopping]
history = model.fit(
X_train,
y_train,
validation_split=0.2,
batch_size=32,
epochs=3,
verbose=1,
callbacks=callbacks
)
当我尝试训练模型时，出现以下错误消息：
ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type int).
我已经提供了详细的错误消息，以获取可能提供的任何见解
Error message
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-107-482ccaecbe1c> in <module>
3 callbacks =[checkpointer, early_stopping]
4
----> 5 history = model.fit(
6 X_train,
7 y_train,
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\keras\engine\training.py in _method_wrapper(self, *args, **kwargs)
64 def _method_wrapper(self, *args, **kwargs):
65 if not self._in_multi_worker_mode(): # pylint: disable=protected-access
---> 66 return method(self, *args, **kwargs)
67
68 # Running inside `run_distribute_coordinator` already.
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\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_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)
793 # `Tensor` and `NumPy` input.
794 (x, y, sample_weight), validation_data = (
--> 795 data_adapter.train_validation_split((x, y, sample_weight),
796 validation_split=validation_split,
797 shuffle=False))
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py in train_validation_split(arrays, validation_split, shuffle)
1335 return array_ops.gather_v2(t, indices)
1336
-> 1337 train_arrays = nest.map_structure(
1338 functools.partial(_split, indices=train_indices), arrays)
1339 val_arrays = nest.map_structure(
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\util\nest.py in map_structure(func, *structure, **kwargs)
615
616 return pack_sequence_as(
--> 617 structure[0], [func(*x) for x in entries],
618 expand_composites=expand_composites)
619
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\util\nest.py in <listcomp>(.0)
615
616 return pack_sequence_as(
--> 617 structure[0], [func(*x) for x in entries],
618 expand_composites=expand_composites)
619
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py in _split(t, indices)
1332 if t is None:
1333 return t
-> 1334 t = ops.convert_to_tensor_v2(t)
1335 return array_ops.gather_v2(t, indices)
1336
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\framework\ops.py in convert_to_tensor_v2(value, dtype, dtype_hint, name)
1276 ValueError: If the `value` is a tensor not of given `dtype` in graph mode.
1277 """
-> 1278 return convert_to_tensor(
1279 value=value,
1280 dtype=dtype,
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\framework\ops.py in convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, dtype_hint, ctx, accepted_result_types)
1339
1340 if ret is None:
-> 1341 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
1342
1343 if ret is NotImplemented:
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\framework\tensor_conversion_registry.py in _default_conversion_function(***failed resolving arguments***)
50 def _default_conversion_function(value, dtype, name, as_ref):
51 del as_ref # Unused.
---> 52 return constant_op.constant(value, dtype, name=name)
53
54
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\framework\constant_op.py in constant(value, dtype, shape, name)
259 ValueError: if called on a symbolic tensor.
260 """
--> 261 return _constant_impl(value, dtype, shape, name, verify_shape=False,
262 allow_broadcast=True)
263
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\framework\constant_op.py in _constant_impl(value, dtype, shape, name, verify_shape, allow_broadcast)
268 ctx = context.context()
269 if ctx.executing_eagerly():
--> 270 t = convert_to_eager_tensor(value, ctx, dtype)
271 if shape is None:
272 return t
c:\users\dt\appdata\local\programs\python38\lib\site-packages\tensorflow\python\framework\constant_op.py in convert_to_eager_tensor(value, ctx, dtype)
94 dtype = dtypes.as_dtype(dtype).as_datatype_enum
95 ctx.ensure_initialized()
---> 96 return ops.EagerTensor(value, ctx.device_name, dtype)
97
98
ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type int).