Python keras lstm的输出形状错误

我需要预测一系列数字的k值。为此，我决定使用Fibonacci序列mod 15，并为每个预测值（n+1，n+2，…n+k）建立一个模型。除此之外，一切正常：当我使用特性测试集预测值时，模型给了我错误的形状。我希望a的形状是（20,1），但我得到的是（20,200,1）。你能帮帮我吗？这是我的密码：

# Importing all the required libraries

import pandas as pd
import numpy as np
from sklearn import metrics
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers

nterms = 1000

fibonacci=[]

# Program to display the Fibonacci sequence up to n-th term

nterms = 10000

# first two terms
n1, n2 = 0, 1
count = 0

# check if the number of terms is valid
if nterms <= 0:
    print("Please enter a positive integer")
elif nterms == 1:
    print("Fibonacci sequence upto",nterms,":")
    print(n1)
else:
    while count < nterms:
        fibonacci.append(n1)
        nth = (n1 + n2)%15
       # update values
        n1 = n2
        n2 = nth
        count += 1

df_fib=pd.DataFrame(fibonacci)

from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler(feature_range = (0, 1))
df_fib_scaled = scaler.fit_transform(df_fib)

# Dictionary for storing generated models and values
models = {}
predicted_values={}

numbers_of_predictions=2

for k in range(0,numbers_of_predictions):

    features_set = []
    labels = []
    for i in range(200, len(df_fib_scaled)-k-20):
        features_set.append(df_fib_scaled[i-200:i, 0])
        labels.append(df_fib_scaled[i+k, 0])

    features_set, labels = np.array(features_set), np.array(labels)
    features_set = np.reshape(features_set, (features_set.shape[0], features_set.shape[1], 1))
    
    # Training the model
    model = keras.Sequential()
    model.add(layers.LSTM(units=50, return_sequences=True, input_shape=(features_set.shape[1], 1)))
    model.add(layers.Dropout(0.2))
    model.add(layers.LSTM(units=50, return_sequences=True))
    model.add(layers.Dropout(0.2))
    model.add(layers.Dense(units = 1))
    model.compile(optimizer = 'adam', loss = 'mean_squared_error')
    model.fit(features_set, labels, epochs = 30, batch_size = 50)
    models[k+1]=model
    
    test_features = []
    real_values=[]
    for i in range(len(df_fib_scaled)-k-20, len(df_fib_scaled)-k):
        test_features.append(df_fib_scaled[i-200:i, 0])
        real_values.append(df_fib_scaled[i+k, 0])
        
    test_features = np.array(test_features)
    test_features = np.reshape(test_features, (test_features.shape[0], test_features.shape[1], 1))
    
    predictions = models[k+1].predict(test_features)

#导入所有必需的库
作为pd进口熊猫
将numpy作为np导入
从SKM学习导入度量
导入tensorflow作为tf
从tensorflow进口keras
从tensorflow.keras导入图层
nterms=1000
斐波那契=[]
#显示斐波那契序列（最多第n项）的程序
nterms=10000
#前两届
n1，n2=0，1
计数=0
#检查条款数量是否有效
如果nterms则预测是最后一个序列元素。添加此行：

predictions = predictions[:, -1]