LSTM 时间序列分类的特征缩放

Question

我正在尝试为 LSTM 时间序列分类 model 准备输入数据。 我想使用sklearn MaxAbsScaler()来缩放功能，但是我很难在我的以下代码中加入缩放。 我正在使用 LSTM 网络的tensorflow实现。

import keras
from keras.layers import LSTM, Dropout, Dense
import tensorflow as tf
import numpy as np
import pandas as pd
df = pd.read_excel('/content/hdds.xlsx')

def generate_data(X, y, sequence_length=2, step = 1):
    X_local = []
    y_local = []
    for start in range(0, len(df) - sequence_length, step):
        end = start + sequence_length
        X_local.append(X[start:end])
        y_local.append(y[end-1])
    return np.array(X_local), np.array(y_local)

X_sequence, y = generate_data(df.loc[:, "V1":"V3"].values, df.Class)

X_sequence.shape, y.shape
((16, 2, 3), (16,))

training_size = int(len(X_sequence) * 0.7)
X_train, y_train = X_sequence[:training_size], y[:training_size]
X_test, y_test = X_sequence[training_size:], y[training_size:]

X_train.shape, X_test.shape
((11, 2, 3), (5, 2, 3))

我打算使用以下代码来缩放输入数据，但它不起作用。 我很欣赏你的建议。 谢谢！

from sklearn.preprocessing import MaxAbsScaler
scaler =  MaxAbsScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)
X_train = np.clip(X_train, -5, 5)
X_test = np.clip(X_test, -5, 5)

样本数据：