请使用Python实现以下代码：Use the first 100 rows as training data to construct 90%, 80%, 60% prediction intervals for waiting time for the remaining 172 data trough conformal prediction. Is your prediction correct 90%, 80%, 60% of the time? (Use any nonconformity measure you like.)

这个问题需要使用到非参数置信度算法，例如Conformal Prediction或Transductive Conformal Prediction。下面是一个使用Transductive Conformal Prediction的Python实现：

import numpy as np
import pandas as pd
from sklearn.neighbors import NearestNeighbors
from sklearn.linear_model import LinearRegression
from nonconformist.tc import TcpClassifier
from nonconformist.nc import MarginErrFunc

# 读取数据集
data = pd.read_csv('data.csv')

# 划分训练集和测试集
train_data = data[:100]
test_data = data[100:]

# 训练模型
model = LinearRegression()
model.fit(train_data[['X']], train_data[['Y']])

# 计算测试集的预测值和非符合性
X_test = test_data[['X']].values
y_test = test_data[['Y']].values
predictions = model.predict(X_test)
errors = np.abs(predictions - y_test)

# 计算最近邻
k = 5
nbrs = NearestNeighbors(n_neighbors=k, algorithm='ball_tree').fit(X_test)
distances, indices = nbrs.kneighbors(X_test)

# 使用Transductive Conformal Prediction构造置信区间
tcp = TcpClassifier(MarginErrFunc(), k=k)
tcp.fit(X_test, y_test, indices, distances)
interval90 = tcp.predict(X_test, significance=0.1)
interval80 = tcp.predict(X_test, significance=0.2)
interval60 = tcp.predict(X_test, significance=0.4)

# 计算正确率
correct90 = 0
correct80 = 0
correct60 = 0
for i in range(len(X_test)):
    if y_test[i] >= interval90[i][0] and y_test[i] <= interval90[i][1]:
        correct90 += 1
    if y_test[i] >= interval80[i][0] and y_test[i] <= interval80[i][1]:
        correct80 += 1
    if y_test[i] >= interval60[i][0] and y_test[i] <= interval60[i][1]:
        correct60 += 1

accuracy90 = correct90 / len(X_test)
accuracy80 = correct80 / len(X_test)
accuracy60 = correct60 / len(X_test)

print('90% prediction interval accuracy:', accuracy90)
print('80% prediction interval accuracy:', accuracy80)
print('60% prediction interval accuracy:', accuracy60)

这个实现使用了scikit-learn和nonconformist库来进行Transductive Conformal Prediction，它可以计算测试集的预测值和非符合性，计算最近邻，并使用Transductive Conformal Prediction构造置信区间。最后，它计算正确率并输出结果。请注意，这个实现可能需要额外的库函数，你可以使用 pip 命令来安装它们。