TF Keras: ValueError: Input 0 of layer sequence is in compatible with the layer: expected ndim=3, found ndim=2

Question

所以我有形成模式的二维向量序列。我想预测序列如何继续。 我有一个 start_xy 数组，由数组组成，顺序为 start_x 和 start_y： 例如[1、2.4、3.8] end_xy 也一样。

我想训练一个模型一个序列预测模型：

import numpy as np
import pickle
import keras
from keras.models import Sequential
from keras.layers import LSTM, Dense
from keras.callbacks import ModelCheckpoint
import training_data_generator
tdg = training_data_generator.training_data_generator(500)
trainingdata = tdg.produceTrainingSequences()
print("Printing DATA!:")
start_xy =[]
end_xy =[]

for batch in trainingdata:
    for pattern in batch:
        order = 1
        for sequence in pattern:
            start = [order,sequence[0],sequence[1]] 
            start_xy.append(start)
            end = [order,sequence[2],sequence[3]]
            end_xy.append(end)
            order = order +1

    
    model = Sequential()
    model.add(LSTM(64, return_sequences=False, input_shape=(2,len(start_xy))))
    model.add(Dense(2, activation='relu'))
    model.compile(loss='mse', optimizer='adam')
    model.fit(start_xy,end_xy,batch_size=len(start_xy), epochs=5000,  verbose=2)

但我收到错误消息：

 ValueError: Input 0 of layer sequential is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: [320, 3]

我怀疑我必须以某种方式重塑我的输入，但我还不明白如何。 我该如何进行这项工作？ 我这样做是否正确？

Answer 1

您通常只需将数据转换为 numpy 数组并对该数据进行一些重塑，以便模型能够接受它。

首先将 start_xy 转换为 numpy 数组，然后将其重塑为 3 个暗淡：

start_xy = np.array(start_xy)
start_xy = start_xy.reshape(*start_xy.shape, 1)

接下来将 LSTM 层的输入形状固定为 [3, 1]：

model.add(LSTM(64, return_sequences=False, input_shape=start_xy.shape[1:]))

让我知道错误是否仍然存在或是否出现另一个错误！