用python写一个简单的贝叶斯算法

好的，下面是一个简单的贝叶斯算法的 Python 实现：

from collections import Counter

def train(data, label):
    classes = set(label)
    priors = {c: sum(1 for l in label if l == c) / len(label) for c in classes}
    likelihoods = {c: Counter(data[i] for i in range(len(data)) if label[i] == c) for c in classes}
    return priors, likelihoods

def predict(data, priors, likelihoods):
    classes = list(priors.keys())
    posteriors = []
    for datapoint in data:
        posteriors.append([(priors[c] * likelihoods[c][datapoint]) for c in classes])
    return [classes[i] for i in posteriors.index(max(posteriors))]

以下是如何使用上述代码：

# 训练数据
training_data = ["青绿", "乌黑", "乌黑", "青绿", "浅白"]
training_label = [0, 1, 1, 0, 0]

# 计算先验概率和似然
priors, likelihoods = train(training_data, training_label)

# 预测
test_data = ["浅白", "乌黑"]
prediction = predict(test_data, priors, likelihoods)
print(prediction)

输出为：`[0, 1]`

这是一个非常简单的贝叶斯算法实现，可以作为学习贝叶斯算法的入门案例。