Python 如何为Keras（tensorflow）中的LSTM正确塑造2D输入和输出？将输入设置为3D并将输出设置为热编码时生成错误

对于类项目，我们必须获取2D数据集并使用LSTM NN进行预测。我们将其与简单的DNN进行比较

我需要重塑我的数据，使其与NN一起工作，我很难找到输入和输出的正确形状。我可能也设置了错误的NN——但它解决了我在处理3D数据时遇到的另一个问题

我的

x\u列车形状

是

（2340590）

，我的

y

形状是

（2340，）

我将

x_列车

改造为

（12340590）

I一个热编码的

y

——称为

binary\u标签

二进制标签。形状
是
（2340,2）

总结相关输入形状：

x_train.shape = (1, 2340, 590)

binary_labels.shape = (2340, 2)

> --------------------------------------------------------------------------- ValueError                                Traceback (most recent call
> last) <ipython-input-224-36b1852bd7a7> in <module>
>      11           metrics=['accuracy'])
>      12 
> ---> 13 rnn_model.fit(X_train, binary_labels, epochs=5, validation_split = 0.2)
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/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_freq, max_queue_size, workers, use_multiprocessing,
> **kwargs)
>     641             `tf.data` dataset or a dataset iterator, and 'steps_per_epoch'
>     642             is None, the epoch will run until the input dataset is exhausted.
> --> 643         validation_steps: Only relevant if `validation_data` is provided and
>     644             is a dataset or dataset iterator. Total number of steps (batches of
>     645             samples) to draw before stopping when performing validation
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training_arrays.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)
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training.py
> in _standardize_user_data(self, x, y, sample_weight, class_weight,
> batch_size, check_steps, steps_name, steps, validation_split, shuffle,
> extract_tensors_from_dataset)    2463     if not self.inputs:    2464 
> # We need to use `x_input` to set the model inputs.
> -> 2465     2466       # If input data is a dataset iterator in graph mode or if it is an eager    2467       # iterator and only one batch
> of samples is required, we fetch the data
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training_utils.py
> in check_array_lengths(inputs, targets, weights)
>     617   """
>     618   batch_count = int(len(index_array) / batch_size)
> --> 619   # to reshape we need to be cleanly divisible by batch size
>     620   # we stash extra items and reappend them after shuffling
>     621   last_batch = index_array[batch_count * batch_size:]
> 
> ValueError: Input arrays should have the same number of samples as
> target arrays. Found 1 input samples and 2340 target samples

问题：

x_train.shape = (1, 2340, 590)

binary_labels.shape = (2340, 2)

> --------------------------------------------------------------------------- ValueError                                Traceback (most recent call
> last) <ipython-input-224-36b1852bd7a7> in <module>
>      11           metrics=['accuracy'])
>      12 
> ---> 13 rnn_model.fit(X_train, binary_labels, epochs=5, validation_split = 0.2)
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/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_freq, max_queue_size, workers, use_multiprocessing,
> **kwargs)
>     641             `tf.data` dataset or a dataset iterator, and 'steps_per_epoch'
>     642             is None, the epoch will run until the input dataset is exhausted.
> --> 643         validation_steps: Only relevant if `validation_data` is provided and
>     644             is a dataset or dataset iterator. Total number of steps (batches of
>     645             samples) to draw before stopping when performing validation
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training_arrays.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)
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training.py
> in _standardize_user_data(self, x, y, sample_weight, class_weight,
> batch_size, check_steps, steps_name, steps, validation_split, shuffle,
> extract_tensors_from_dataset)    2463     if not self.inputs:    2464 
> # We need to use `x_input` to set the model inputs.
> -> 2465     2466       # If input data is a dataset iterator in graph mode or if it is an eager    2467       # iterator and only one batch
> of samples is required, we fetch the data
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training_utils.py
> in check_array_lengths(inputs, targets, weights)
>     617   """
>     618   batch_count = int(len(index_array) / batch_size)
> --> 619   # to reshape we need to be cleanly divisible by batch size
>     620   # we stash extra items and reappend them after shuffling
>     621   last_batch = index_array[batch_count * batch_size:]
> 
> ValueError: Input arrays should have the same number of samples as
> target arrays. Found 1 input samples and 2340 target samples

运行模型会生成一个错误，即输入数组应具有相同数量的样本

我尝试将
二进制标签重新设置为
（12340,2）
-但运行NN时，我得到
值错误
：

空的训练数据

rnn\u模型=keras.Sequential([
keras.layers.LSTM（2，输入_-shape=（X_-train.shape[1]，X_-train.shape[2]），返回_-sequences=True），
keras.layers.LSTM（590，返回序列=True，激活=tf.nn.relu），
keras.layers.LSTM（590，返回序列=True，激活=tf.nn.relu），
keras.layers.LSTM（590，返回序列=True，激活=tf.nn.relu），
keras。层。致密（2，激活='sigmoid'）
])
rnn_model.compile（optimizer='adam'，
损失='binary\u交叉熵'，
指标=[‘准确度’]）
rnn_模型拟合（X_序列，二进制标签，历元=5，验证_分割=0.2）

我希望模型能够运行！相反，我收到一条无法解决的错误消息

有人知道我如何解决这个问题吗

错误消息的相关部分：
ValueError:输入数组的样本数应与目标数组的样本数相同。找到1个输入样本和2340个目标样本

整个错误消息：

x_train.shape = (1, 2340, 590)

binary_labels.shape = (2340, 2)

> --------------------------------------------------------------------------- ValueError                                Traceback (most recent call
> last) <ipython-input-224-36b1852bd7a7> in <module>
>      11           metrics=['accuracy'])
>      12 
> ---> 13 rnn_model.fit(X_train, binary_labels, epochs=5, validation_split = 0.2)
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/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_freq, max_queue_size, workers, use_multiprocessing,
> **kwargs)
>     641             `tf.data` dataset or a dataset iterator, and 'steps_per_epoch'
>     642             is None, the epoch will run until the input dataset is exhausted.
> --> 643         validation_steps: Only relevant if `validation_data` is provided and
>     644             is a dataset or dataset iterator. Total number of steps (batches of
>     645             samples) to draw before stopping when performing validation
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training_arrays.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)
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training.py
> in _standardize_user_data(self, x, y, sample_weight, class_weight,
> batch_size, check_steps, steps_name, steps, validation_split, shuffle,
> extract_tensors_from_dataset)    2463     if not self.inputs:    2464 
> # We need to use `x_input` to set the model inputs.
> -> 2465     2466       # If input data is a dataset iterator in graph mode or if it is an eager    2467       # iterator and only one batch
> of samples is required, we fetch the data
> 
> ~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training_utils.py
> in check_array_lengths(inputs, targets, weights)
>     617   """
>     618   batch_count = int(len(index_array) / batch_size)
> --> 619   # to reshape we need to be cleanly divisible by batch size
>     620   # we stash extra items and reappend them after shuffling
>     621   last_batch = index_array[batch_count * batch_size:]
> 
> ValueError: Input arrays should have the same number of samples as
> target arrays. Found 1 input samples and 2340 target samples

--------------------------------------------------------------------------------------ValueError回溯（最近的调用）
>最后）在
>11个指标=[‘准确度’]）
>      12 
>-->13 rnn_模型.fit（X_序列，二进制标签，历元=5，验证_分割=0.2）
> 
>~/Desktop/Program_下载/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training.py
>适合（自我、x、y、批量大小、时代、详细、回调、，
>验证分割、验证数据、洗牌、等级权重、，
>样本重量、初始历元、每历元步骤、验证步骤、，
>验证频率、最大队列大小、工人、使用多处理、，
>**kwargs）
>641`tf.data`dataset或dataset迭代器，以及'steps_per_epoch'
>642为“无”，将运行历元，直到输入数据集耗尽。
>-->643验证步骤：仅当提供了“验证数据”且
>644是数据集或数据集迭代器。步骤总数（批数）
>645个样本）在执行验证时停止之前提取
> 
>~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training_arrays.py
>适合（自我、型号、x、y、批量大小、年代、详细、回调、，
>验证分割、验证数据、洗牌、等级权重、，
>样本重量、初始历元、每历元步骤、验证步骤、，
>验证频率，**kwargs）
> 
>~/Desktop/Program_下载/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training.py
>标准化用户数据（自身、x、y、样本重量、类别重量、，
>批量大小、检查步骤、步骤名称、步骤、验证、拆分、洗牌、，
>从_数据集中提取_张量_）2463如果不是自输入：2464
>#我们需要使用'x_input'来设置模型输入。
>->2465 2466#如果输入数据是图形模式下的数据集迭代器，或者如果它是一个渴望的2467#迭代器，并且只有一批
>如果需要大量样本，我们将获取数据
> 
>~/Desktop/Program_Downloads/anaconda3/envs/uwdatasci420/lib/python3.6/site-packages/tensorflow/python/keras/engine/training_utils.py
>检查数组长度（输入、目标、权重）
>     617   """
>618批次计数=整数（len（索引数组）/批次大小）
>-->619#要重塑，我们需要用批量大小清楚地划分
>620#我们把多余的物品藏起来，洗牌后重新出售
>621最后一批=索引数组[批次计数*批次大小：]
> 
>ValueError:输入数组的样本数应与
>目标数组。找到1个输入样本和2340个目标样本
如果我很理解你的问题，问题是：

x\u train.shape
是
（2340590）
所以您有2340个大小（590，）的样本

如果您像以前那样重塑数据：
（12340590）
，您将只输入一个大小为
（2340590）
，因为keras模型输入形状的定义如下：
（批次大小，大小1，大小2）

因此，为了使模型正常工作，您只需对数据进行如下重塑：

x_train = np.expand_dims(x_train, -1) #new shape = (2340, 590, 1)

试试看，告诉我是否更好！
我认为将最后的
致密层包装进去会为您解决问题

#。。。
keras.layers.LSTM（590，返回序列=True，激活=tf.nn.relu），
keras.layers.TimeDistributed（keras.layers.Dense（2，激活='sigmoid'））
]

也就是说，如果要对序列中的每个时间步进行分类，则可能不希望返回上一个LSTM中的序列，而只希望在看到整个输入序列后，根据LSTM的最终状态进行分类：