基于LSTM的双色预测Python程序报错求助：解决IndexError问题

我希望开发一个基于LSTM技术的Python程序，借助AI与机器学习库，依据最近40次双色（r/g）抽取结果的模式，预测下一次的抽取结果或概率。我已编写对应程序，但运行时持续触发IndexError错误，报错位置为X = np.reshape(X, (X.shape[0], X.shape[1], 1))，错误信息为IndexError: tuple index out of range，请求排查解决。

程序代码 #

from keras.models import Sequential
from keras.layers import LSTM, Dense
import numpy as np


def predict_next_color_lstm(outcomes):
    if len(outcomes) < 40:
        return "Error: Number of outcomes provided is less than 40."

    # Convert string input to integer sequence
    seq = [0 if x == 'r' else 1 for x in outcomes]

    # Create rolling window of 40 outcomes
    X = []
    y = []
    for i in range(len(seq) - 40):
        X.append(seq[i:i + 40])
        y.append(seq[i + 40])
    X = np.array(X)
    y = np.array(y)

    # Reshape X to fit LSTM input shape
    X = np.reshape(X, (X.shape[0], X.shape[1], 1))

    # Create LSTM model
    model = Sequential()
    model.add(LSTM(50, input_shape=(40, 1)))
    model.add(Dense(1, activation='sigmoid'))

    # Compile the model
    model.compile(loss='binary_crossentropy', optimizer='adam')

    # Train the model
    model.fit(X, y, epochs=50, batch_size=32)

    # Predict the next outcome
    last_40 = seq[-40:]
    pred = model.predict(np.array([last_40]))
    return 'r' if pred < 0.5 else 'g'


def get_input():
    # Ask the user to enter a ball color sequence of length 40
    ball_seq = input("Enter the ball color sequence of length 40 (e.g. rrggrrgrrgggrgrgrrggggrgrgrrgrgggrrgggg): ")
    return ball_seq


# _main_
ball_seq = get_input()
print("Prediction : ", predict_next_color_lstm(ball_seq))

报错信息 #

C:\Users\Ashish\miniconda3\python.exe C:\Users\Ashish\Desktop\pyt_pract\test_prob1.py
Enter the ball color sequence of length 40 (e.g. rrggrrgrrgggrgrgrrggggrgrgrrgrgggrrgggg): rgggrrgrgrggrrgrgrgrgrggggrrrrggrrggrgrg
Traceback (most recent call last):
  File "C:\Users\Ashish\Desktop\pyt_pract\test_prob1.py", line 50, in <module>
    print("Prediction : ", predict_next_color_lstm(ball_seq))
  File "C:\Users\Ashish\Desktop\pyt_pract\test_prob1.py", line 23, in predict_next_color_lstm
    X = np.reshape(X, (X.shape[0], X.shape[1], 1))
IndexError: tuple index out of range

问题原因与解决方法 #

问题根源 #

当输入的序列长度恰好为40时，len(seq) - 40等于0，导致循环for i in range(len(seq) - 40)完全不会执行，最终X和y都是空数组。此时np.array(X)是一个0维空数组，其shape为()，没有第二个索引，调用X.shape[1]就会触发“元组索引超出范围”的错误。

另外，当前逻辑需要用前40个结果预测第41个，因此至少需要41个历史结果才能生成一组训练数据，仅40个样本无法完成模型训练。

解决步骤 #

1. 修正输入长度检查：将输入长度的判断条件改为要求长度大于40，确保能生成至少一组训练数据：

   if len(outcomes) <= 40:
       return "Error: Number of outcomes provided must be greater than 40 (need at least 41 to generate training data)."
   

2. 调整输入提示：修改get_input函数里的提示文本，明确告诉用户需要输入长度大于40的序列：

   ball_seq = input("Enter the ball color sequence (length must be at least 41, e.g. rrggrrgrrgggrgrgrrggggrgrgrrgrgggrrgggg...): ")
   

修改后的完整代码 #

from keras.models import Sequential
from keras.layers import LSTM, Dense
import numpy as np


def predict_next_color_lstm(outcomes):
    if len(outcomes) <= 40:
        return "Error: Number of outcomes provided must be greater than 40 (need at least 41 to generate training data)."

    # Convert string input to integer sequence
    seq = [0 if x == 'r' else 1 for x in outcomes]

    # Create rolling window of 40 outcomes
    X = []
    y = []
    for i in range(len(seq) - 40):
        X.append(seq[i:i + 40])
        y.append(seq[i + 40])
    X = np.array(X)
    y = np.array(y)

    # Reshape X to fit LSTM input shape
    X = np.reshape(X, (X.shape[0], X.shape[1], 1))

    # Create LSTM model
    model = Sequential()
    model.add(LSTM(50, input_shape=(40, 1)))
    model.add(Dense(1, activation='sigmoid'))

    # Compile the model
    model.compile(loss='binary_crossentropy', optimizer='adam')

    # Train the model
    model.fit(X, y, epochs=50, batch_size=32)

    # Predict the next outcome
    last_40 = seq[-40:]
    pred = model.predict(np.array([last_40]))
    return 'r' if pred < 0.5 else 'g'


def get_input():
    # Ask the user to enter a ball color sequence with length >=41
    ball_seq = input("Enter the ball color sequence (length must be at least 41, e.g. rrggrrgrrgggrgrgrrggggrgrgrrgrgggrrgggg...): ")
    return ball_seq


# _main_
ball_seq = get_input()
print("Prediction : ", predict_next_color_lstm(ball_seq))

内容的提问来源于stack exchange，提问作者Ashish