python - Keras - LSTM 如何重塑输入值
问题描述
我正在将 LSTM 与 keras 一起使用，我的测试和训练工作正常，但是当我尝试输入不同的输入时，我收到错误无法将大小 20 重塑为(1,20,30)
这是我的 model.fit() 代码
PositiveOrNegativeLabel=np.array([[1]])
PositiveOrNegativeLabel=PositiveOrNegativeLabel.reshape(1,-1)
PositiveOrNegativeLabel.shape
inputBatch =inputBatch.reshape(1,24,30)
testBatch =testBatch.reshape(1,24,30)
model=Sequential()
model.add(LSTM(100,input_shape=(24,30)))
model.add(Dense(1,activation="relu"))
model.compile(loss='mean_absolute_error',optimizer='adam')
model.fit(inputBatch,PositiveOrNegativeLabel,batch_size=24,epochs=9,verbose=1)
我的输入是 firstSentence 数组，就像这样
[ 174 11501 420 4242 12111 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0]
firstSentence 的形状是 (20,)
我用来拟合模型的输入类型
[[ 0. 12184. 420. 4636. 0. 8840. 0. 0. 10499. 11508.
7511. 0. 5468. 2879. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 6689. 2818. 12003. 6480. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 3045. 11087. 2710. 0. 494. 1087. 420. 4995.
11516. 3637. 5842. 0. 9963. 7015. 11090. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 1287. 420. 4070. 11087. 7410. 12186. 2387. 12111. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 3395. 1087. 11904. 7232. 8840. 10115. 4494. 11516. 7441.
8535. 12106. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 494. 0. 0. 6541. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 8744. 11105. 1570. 5842. 174. 11266. 2929. 10438. 2879.
0. 10936. 6330. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 11956. 5222. 0. 0. 12106. 6481. 0. 7093. 13756.
12152. 0. 0. 0. 0. 10173. 0. 5173. 13756. 9371.
0. 9956. 0. 0. 9716. 0. 0. 0. 0. 0.]
[ 0. 3395. 1087. 11904. 7232. 8840. 10115. 4494. 11516. 7441.
8535. 12106. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 420. 5842. 3058. 11875. 2879. 1087. 11105. 4995. 8840.
0. 11100. 11875. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 5419. 420. 2250. 1299. 2151. 12111. 6454. 0. 11501.
8094. 5842. 942. 7503. 7410. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 420. 5842. 3058. 11875. 2879. 1087. 11105. 4995. 8840.
0. 11100. 11875. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 3395. 1087. 11904. 7232. 8840. 10115. 4494. 11516. 7441.
8535. 12106. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 1287. 420. 4070. 11087. 7410. 12186. 2387. 12111. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 3395. 1087. 11904. 7232. 8840. 10115. 4494. 11516. 7441.
8535. 12106. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 11501. 1592. 10603. 11102. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 174. 5842. 2387. 10453. 11090. 0. 7531. 11956. 450.
420. 11516. 6693. 2624. 9963. 11992. 9322. 11090. 12106. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 7544. 0. 1709. 420. 10936. 5222. 5842. 10407. 6937.
11329. 2937. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 1520. 1295. 0. 8396. 9322. 12715. 0. 5172. 7232.
11266. 0. 11266. 2757. 4416. 12020. 12111. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 7544. 0. 1709. 420. 10936. 5222. 5842. 10407. 6937.
11329. 2937. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 9191. 5952. 0. 0. 11516. 9413. 3081. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 11516. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 3395. 1087. 11904. 7232. 8840. 10115. 4494. 11516. 7441.
8535. 12106. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 9371. 10412. 2356. 5412. 11502. 0. 1087. 228. 0.
2937. 11480. 10412. 5412. 420. 9435. 2937. 228. 1057. 9435.
12111. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
我得到的价值错误是这样的
cannot reshape array of size 20 into shape (1,20,30)
我正在运行这段代码来预测它
predict=model.predict(firstSentence, batch_size=24, verbose=1, steps=None)
# make a prediction
ynew = model.predict_classes(firstSentence)
# show the inputs and predicted outputs
for i in range(len(predict)):
print("X=%s, Predicted=%s" % (predict[i], ynew[i]))
解决方案
我认为您将batch_size与时间步数混淆了。Kerasinput_shape中的 LSTM 层应该是 ( num_timesteps, num_features)。根据使用的 inbatch_size推断。据我所知，您的训练数据有 24 个示例，每个示例有 30 个时间步长，每个时间步长都有 1 个特征。batch_sizefit
此外，您的代码只有一个标签，但据我所知，您有 24 个示例，因此需要 24 个标签。
import numpy as np
from keras.models import Sequential
from keras.layers import LSTM, Dense
# Need same number of labels as examples
PositiveOrNegativeLabel = np.ones(shape = (24, 1))
# 24 examples, each with 30 timesteps, 1 feature at each timestep
inputBatch = result.reshape(24, 30, 1)
model=Sequential()
# input_shape is (num_timesteps, num_features)
model.add(LSTM(100, input_shape=(30, 1)))
model.add(Dense(1, activation="relu"))
model.compile(loss='mean_absolute_error', optimizer='adam')
model.fit(inputBatch, PositiveOrNegativeLabel,
batch_size=24, epochs=9, verbose=1)
您的测试句子只有一个示例，有 20 个时间步，每个时间步有 1 个特征。因此，您需要用零填充测试语句。
from keras.preprocessing.sequence import pad_sequences
# Need 30 timesteps
firstSentence = pad_sequences([firstSentence], maxlen = 30)[0]
firstSentence = firstSentence.reshape((1, 30, 1))
predict=model.predict(firstSentence, batch_size=1, verbose=1, steps=None)
# make a prediction
ynew = model.predict_classes(firstSentence)
# show the inputs and predicted outputs
for i in range(len(predict)):
print("X=%s, Predicted=%s" % (predict[i], ynew[i]))
此代码有效，但我可能错误地解释了您的设置。因此，如果您的训练标签建议您只有 1 个训练示例，那么输入批次的形状应该是(1, 24, 30)，并且您的测试序列也需要 24 个时间步长，每个时间步长有 30 个特征。
有关在 Keras 中塑造 LSTM 输入的更多信息，请参阅这篇文章。
推荐阅读
- javascript - 如何从对象值中获取 HTML 标签？
- django - 如何在模板 Django 中显示 MultiSelectField
- isabelle - 在 Isabelle/Isar 中使用时累积结果
- sql - 在 ODBC 中获取一个月的最后一天（2 个月前）
- node.js - 无法使用 express 从后端获取数据
- java - 使用 Java 以输入的精度递增浮点数
- c++ - 为什么 C++20 概念与“const auto&”不兼容？
- azure - 使用标头而不是 Cookie 定位特定的 Azure Web 应用实例
- architecture - USB 可以处理不同速度的 I/O 设备吗？
- javascript - Javascript函数中是否存在错误？