IndexError：列表索引超出范圍，而 keras model.predict()

Question

如果我嘗試使用帶有 3 個輸入的 model 使用此代碼進行預測：

predicted_seaLevel = []

prediction = model2.predict(
    [[np.array(X[0][0])],[np.array(X[1][0])],[np.array(X[2][0])]],
    batch_size=None,
    verbose='auto',
    steps=None,
    callbacks=None,
    max_queue_size=10,
    workers=1,
    use_multiprocessing=False
    )
predicted_seaLevel.append(prediction)

我收到此錯誤：

IndexError                                Traceback (most recent call last)
Input In [119], in <cell line: 3>()
      1 predicted_seaLevel = []
----> 3 prediction = model2.predict(
      4     [[np.array(X[0][0])],[np.array(X[1][0])],[np.array(X[2][0])]],
      5     batch_size=None,
      6     verbose='auto',
      7     steps=None,
      8     callbacks=None,
      9     max_queue_size=10,
     10     workers=1,
     11     use_multiprocessing=False
     12     )
     13 predicted_seaLevel.append(prediction)

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\training.py:1720, in Model.predict(self, x, batch_size, verbose, steps, callbacks, max_queue_size, workers, use_multiprocessing)
   1714   except ValueError:
   1715     warnings.warn('Using Model.predict with '
   1716                   'MultiWorkerDistributionStrategy or TPUStrategy and '
   1717                   'AutoShardPolicy.FILE might lead to out-of-order result'
   1718                   '. Consider setting it to AutoShardPolicy.DATA.')
-> 1720 data_handler = data_adapter.get_data_handler(
   1721     x=x,
   1722     batch_size=batch_size,
   1723     steps_per_epoch=steps,
   1724     initial_epoch=0,
   1725     epochs=1,
   1726     max_queue_size=max_queue_size,
   1727     workers=workers,
   1728     use_multiprocessing=use_multiprocessing,
   1729     model=self,
   1730     steps_per_execution=self._steps_per_execution)
   1732 # Container that configures and calls `tf.keras.Callback`s.
   1733 if not isinstance(callbacks, callbacks_module.CallbackList):

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\data_adapter.py:1383, in get_data_handler(*args, **kwargs)
   1381 if getattr(kwargs["model"], "_cluster_coordinator", None):
   1382   return _ClusterCoordinatorDataHandler(*args, **kwargs)
-> 1383 return DataHandler(*args, **kwargs)

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\data_adapter.py:1138, in DataHandler.__init__(self, x, y, sample_weight, batch_size, steps_per_epoch, initial_epoch, epochs, shuffle, class_weight, max_queue_size, workers, use_multiprocessing, model, steps_per_execution, distribute)
   1135   self._steps_per_execution_value = steps_per_execution.numpy().item()
   1137 adapter_cls = select_data_adapter(x, y)
-> 1138 self._adapter = adapter_cls(
   1139     x,
   1140     y,
   1141     batch_size=batch_size,
   1142     steps=steps_per_epoch,
   1143     epochs=epochs - initial_epoch,
   1144     sample_weights=sample_weight,
   1145     shuffle=shuffle,
   1146     max_queue_size=max_queue_size,
   1147     workers=workers,
   1148     use_multiprocessing=use_multiprocessing,
   1149     distribution_strategy=tf.distribute.get_strategy(),
   1150     model=model)
   1152 strategy = tf.distribute.get_strategy()
   1154 self._current_step = 0

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\data_adapter.py:240, in TensorLikeDataAdapter.__init__(self, x, y, sample_weights, sample_weight_modes, batch_size, epochs, steps, shuffle, **kwargs)
    235 (sample_weights, _, _) = training_utils.handle_partial_sample_weights(
    236     y, sample_weights, sample_weight_modes, check_all_flat=True)
    238 inputs = pack_x_y_sample_weight(x, y, sample_weights)
--> 240 num_samples = set(int(i.shape[0]) for i in tf.nest.flatten(inputs)).pop()
    241 _check_data_cardinality(inputs)
    243 # If batch_size is not passed but steps is, calculate from the input data.
    244 # Default to 32 for backwards compat.

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\data_adapter.py:240, in <genexpr>(.0)
    235 (sample_weights, _, _) = training_utils.handle_partial_sample_weights(
    236     y, sample_weights, sample_weight_modes, check_all_flat=True)
    238 inputs = pack_x_y_sample_weight(x, y, sample_weights)
--> 240 num_samples = set(int(i.shape[0]) for i in tf.nest.flatten(inputs)).pop()
    241 _check_data_cardinality(inputs)
    243 # If batch_size is not passed but steps is, calculate from the input data.
    244 # Default to 32 for backwards compat.

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\framework\tensor_shape.py:896, in TensorShape.__getitem__(self, key)
    894 else:
    895   if self._v2_behavior:
--> 896     return self._dims[key].value
    897   else:
    898     return self._dims[key]

IndexError: list index out of range

這是我的 Model：

def simpleFunctional():
    
    input1 = keras.layers.Input(shape=(1,))
    input2 = keras.layers.Input(shape=(1,))
    input3 = keras.layers.Input(shape=(1,))
    merged = keras.layers.Concatenate(axis=1)([input1, input2, input3])
    dense1 = keras.layers.Dense(16, input_dim=2, activation=keras.activations.relu, use_bias=True)(merged)
    dense2 = keras.layers.Dense(32, input_dim=2, activation=keras.activations.relu, use_bias=True)(dense1)
    dense3 = keras.layers.Dense(64, input_dim=2, activation=keras.activations.relu, use_bias=True)(dense2)
    dense4 = keras.layers.Dense(32, input_dim=2, activation=keras.activations.relu, use_bias=True)(dense3)
    dense5 = keras.layers.Dense(32, input_dim=2, activation=keras.activations.relu, use_bias=True)(dense4)
    output = keras.layers.Dense(1, activation=keras.activations.relu, use_bias=True)(dense5)
    model = keras.models.Model(inputs=[input1, input2, input3], outputs=output)
    
    
    #opt = SGD(learning_rate=0.001)
    model.compile(
        loss='binary_crossentropy',
        optimizer='adam',
        metrics=["accuracy"],)

    history = model.fit(X_train, Y_train, batch_size=45, epochs=900)

    test_scores = model.evaluate(X_test, Y_test, verbose=2)
    print("Test loss:", test_scores[0])
    print("Test accuracy:", test_scores[1])
    model.save("../AI_Lilly")
    return model

但這工作得很好：

predicted_seaLevel = []

prediction = model2.predict(
    [[np.array([0.21468531468531468])],[np.array([0.4228191581323462])],[np.array([0.17076839761718698])]],
    batch_size=None,
    verbose='auto',
    steps=None,
    callbacks=None,
    max_queue_size=10,
    workers=1,
    use_multiprocessing=False
    )
predicted_seaLevel.append(prediction)

盡管：

[[np.array(X[0][0])],[np.array(X[1][0])],[np.array(X[2][0])]] == [[np.array([0.21468531468531468])],[np.array([0.4228191581323462])],[np.array([0.17076839761718698])]]

是真的。

我也試過：

X_0 = []
X_1 = []
X_2 = []

for i in range(54):
    X_0.append(np.array(X[0][i]))
    X_1.append(np.array(X[1][i]))
    X_2.append(np.array(X[2][i]))

predicted_seaLevel = []

prediction = model2.predict(
    [[X_0[0]],[X_1[0]],[X_2[0]]],
    batch_size=None,
    verbose='auto',
    steps=None,
    callbacks=None,
    max_queue_size=10,
    workers=1,
    use_multiprocessing=False
    )
predicted_seaLevel.append(prediction)

但它給了我同樣的錯誤。

我也試過：

predicted_seaLevel = []

prediction = model2.predict(
    [X[0][0],X[1][0],X[2][0]],
    batch_size=None,
    verbose='auto',
    steps=None,
    callbacks=None,
    max_queue_size=10,
    workers=1,
    use_multiprocessing=False
    )
predicted_seaLevel.append(prediction)

但它給了我這個錯誤：

Input In [146], in <cell line: 3>()
      1 predicted_seaLevel = []
----> 3 prediction = model2.predict(
      4     [X[0][0],X[1][0],X[2][0]],
      5     batch_size=None,
      6     verbose='auto',
      7     steps=None,
      8     callbacks=None,
      9     max_queue_size=10,
     10     workers=1,
     11     use_multiprocessing=False
     12     )
     13 predicted_seaLevel.append(prediction)

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\training.py:1751, in Model.predict(self, x, batch_size, verbose, steps, callbacks, max_queue_size, workers, use_multiprocessing)
   1749 for step in data_handler.steps():
   1750   callbacks.on_predict_batch_begin(step)
-> 1751   tmp_batch_outputs = self.predict_function(iterator)
   1752   if data_handler.should_sync:
   1753     context.async_wait()

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\eager\def_function.py:885, in Function.__call__(self, *args, **kwds)
    882 compiler = "xla" if self._jit_compile else "nonXla"
    884 with OptionalXlaContext(self._jit_compile):
--> 885   result = self._call(*args, **kwds)
    887 new_tracing_count = self.experimental_get_tracing_count()
    888 without_tracing = (tracing_count == new_tracing_count)

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\eager\def_function.py:924, in Function._call(self, *args, **kwds)
    921 self._lock.release()
    922 # In this case we have not created variables on the first call. So we can
    923 # run the first trace but we should fail if variables are created.
--> 924 results = self._stateful_fn(*args, **kwds)
    925 if self._created_variables and not ALLOW_DYNAMIC_VARIABLE_CREATION:
    926   raise ValueError("Creating variables on a non-first call to a function"
    927                    " decorated with tf.function.")

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\eager\function.py:3038, in Function.__call__(self, *args, **kwargs)
   3035 """Calls a graph function specialized to the inputs."""
   3036 with self._lock:
   3037   (graph_function,
-> 3038    filtered_flat_args) = self._maybe_define_function(args, kwargs)
   3039 return graph_function._call_flat(
   3040     filtered_flat_args, captured_inputs=graph_function.captured_inputs)

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\eager\function.py:3459, in Function._maybe_define_function(self, args, kwargs)
   3449 with ag_ctx.ControlStatusCtx(
   3450     status=ag_status, options=self._autograph_options):
   3451 
   (...)
   3454   # and 2. there's no provided input signature
   3455   # and 3. there's been a cache miss for this calling context
   3456   if (self._experimental_relax_shapes and
   3457       self.input_signature is None and
   3458       call_context_key in self._function_cache.missed):
-> 3459     return self._define_function_with_shape_relaxation(
   3460         args, kwargs, flat_args, filtered_flat_args, cache_key_context)
   3462   self._function_cache.missed.add(call_context_key)
   3463   graph_function = self._create_graph_function(args, kwargs)

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\eager\function.py:3381, in Function._define_function_with_shape_relaxation(self, args, kwargs, flat_args, filtered_flat_args, cache_key_context)
   3374   (relaxed_arg_specs, relaxed_kwarg_specs) = nest.pack_sequence_as(
   3375       (args, kwargs), relaxed_arg_specs, expand_composites=False)
   3376   (args, kwargs) = nest.pack_sequence_as(
   3377       (relaxed_arg_specs, relaxed_kwarg_specs),
   3378       flat_args,
   3379       expand_composites=True)
-> 3381 graph_function = self._create_graph_function(
   3382     args, kwargs, override_flat_arg_shapes=relaxed_arg_shapes)
   3383 self._function_cache.arg_relaxed[rank_only_cache_key] = graph_function
   3385 return (graph_function, [
   3386     t for t in nest.flatten((args, kwargs), expand_composites=True)
   3387     if isinstance(t, (ops.Tensor,
   3388                       resource_variable_ops.BaseResourceVariable))
   3389 ])

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\eager\function.py:3298, in Function._create_graph_function(self, args, kwargs, override_flat_arg_shapes)
   3293 missing_arg_names = [
   3294     "%s_%d" % (arg, i) for i, arg in enumerate(missing_arg_names)
   3295 ]
   3296 arg_names = base_arg_names + missing_arg_names
   3297 graph_function = ConcreteFunction(
-> 3298     func_graph_module.func_graph_from_py_func(
   3299         self._name,
   3300         self._python_function,
   3301         args,
   3302         kwargs,
   3303         self.input_signature,
   3304         autograph=self._autograph,
   3305         autograph_options=self._autograph_options,
   3306         arg_names=arg_names,
   3307         override_flat_arg_shapes=override_flat_arg_shapes,
   3308         capture_by_value=self._capture_by_value),
   3309     self._function_attributes,
   3310     function_spec=self.function_spec,
   3311     # Tell the ConcreteFunction to clean up its graph once it goes out of
   3312     # scope. This is not the default behavior since it gets used in some
   3313     # places (like Keras) where the FuncGraph lives longer than the
   3314     # ConcreteFunction.
   3315     shared_func_graph=False)
   3316 return graph_function

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\framework\func_graph.py:1007, 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, acd_record_initial_resource_uses)
   1004 else:
   1005   _, original_func = tf_decorator.unwrap(python_func)
-> 1007 func_outputs = python_func(*func_args, **func_kwargs)
   1009 # invariant: `func_outputs` contains only Tensors, CompositeTensors,
   1010 # TensorArrays and `None`s.
   1011 func_outputs = nest.map_structure(convert, func_outputs,
   1012                                   expand_composites=True)

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\eager\def_function.py:668, in Function._defun_with_scope.<locals>.wrapped_fn(*args, **kwds)
    664 with default_graph._variable_creator_scope(scope, priority=50):  # pylint: disable=protected-access
    665   # __wrapped__ allows AutoGraph to swap in a converted function. We give
    666   # the function a weak reference to itself to avoid a reference cycle.
    667   with OptionalXlaContext(compile_with_xla):
--> 668     out = weak_wrapped_fn().__wrapped__(*args, **kwds)
    669   return out

File ~\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\framework\func_graph.py:994, in func_graph_from_py_func.<locals>.wrapper(*args, **kwargs)
    992 except Exception as e:  # pylint:disable=broad-except
    993   if hasattr(e, "ag_error_metadata"):
--> 994     raise e.ag_error_metadata.to_exception(e)
    995   else:
    996     raise

ValueError: in user code:

    C:\Users\FutureSpace\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\training.py:1586 predict_function  *
        return step_function(self, iterator)
    C:\Users\FutureSpace\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\training.py:1576 step_function  **
        outputs = model.distribute_strategy.run(run_step, args=(data,))
    C:\Users\FutureSpace\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:1286 run
        return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
    C:\Users\FutureSpace\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:2849 call_for_each_replica
        return self._call_for_each_replica(fn, args, kwargs)
    C:\Users\FutureSpace\.conda\envs\handwritingNumbersAI\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:3632 _call_for_each_replica
        return fn(*args, **kwargs)
    C:\Users\FutureSpace\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\training.py:1569 run_step  **
        outputs = model.predict_step(data)
    C:\Users\FutureSpace\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\training.py:1537 predict_step
        return self(x, training=False)
    C:\Users\FutureSpace\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\base_layer.py:1020 __call__
        input_spec.assert_input_compatibility(self.input_spec, inputs, self.name)
    C:\Users\FutureSpace\.conda\envs\handwritingNumbersAI\lib\site-packages\keras\engine\input_spec.py:199 assert_input_compatibility
        raise ValueError('Layer ' + layer_name + ' expects ' +

    ValueError: Layer model_18 expects 3 input(s), but it received 1 input tensors. Inputs received: [<tf.Tensor 'ExpandDims:0' shape=(None, 1) dtype=float32>]

我真的不知道如何解決，也無法向自己解釋一件事有效，但另一件事等於第一件事無效。

請幫忙！

Answer 1

您需要為 model 提供 3 個輸入，因此您的預測輸入必須具有如下形狀：

[(None, 1), (None, 1), (None, 1)]

要解決您的問題，只需使用以下命令：

model.predict(
    [
        np.array([[X[0][0]],]), # inputs of input1
        np.array([[X[1][0]],]), # inputs of input2
        np.array([[X[2][0]],]), # inputs of input3
    ],
    ...
)