模型评估指标计算和常见编码

一、模型评估指标编程实现

#1.生成真实的值和标签
from sklearn.metrics import confusion_matrix

y_true = [1, 1, 0, 1, 0, 1]
y_pred = [1, 0, 0, 1, 1, 1]

TP, TN,FP ,FN  = confusion_matrix(y_true, y_pred).ravel()
print('输出TP值',TP)  # 输出TP值
print('输出FP值',FP)
print('输出FN值',FN)
print('输出TN值',TN)
#2.显示出混淆矩阵
cm = confusion_matrix(y_pred,y_true)
print("混淆矩阵:")
print(cm
#3.计算各个指标
error_rate = (FP + FN) / (TP + TN + FP + FN)
print(error_rate)  # 输出错误率

right_rate=1-error_rate
right_rate #1-错误率=准确率

from sklearn.metrics import precision_score

precision = precision_score(y_true, y_pred)
print(precision)  # 输出精确率，也叫做查准率.注意精确率不是准确率

from sklearn.metrics import recall_score

recall = recall_score(y_true, y_pred)
print(recall)  # 输出查全率也叫做召回率


from sklearn.metrics import f1_score

f1 = f1_score(y_true, y_pred)
print(f1)  # 输出F1值

二、计算实现

 三、编程实现one-hot编码、k-mer编码

# 导入必要的库
import numpy as np

# One-hot编码实现
def one_hot_encoding2(sequence):
    bases = 'ACGT'  # DNA碱基
    encoding = np.zeros((len(sequence), len(bases)))
    for i, base in enumerate(sequence):
        if base in bases:
            encoding[i, bases.index(base)] = 1
    return encoding

# K-mer编码实现
def kmer_encoding2(sequence, k):
    bases = 'ACGT'  # DNA碱基
    encoding = []
    for i in range(len(sequence) - k + 1):
        kmer = sequence[i:i+k]
        if 'N' not in kmer:  # 如果k-mer中没有未知碱基
            encoding.append(kmer)
    return encoding

# 示例序列
sequence = "ACGTACGTACGT"

# 进行One-hot编码
one_hot_encoded = one_hot_encoding(sequence)
print("One-hot编码结果:")
print(one_hot_encoded)

# 进行K-mer编码
kmer_encoded = kmer_encoding(sequence, k=3)
print("\nK-mer编码结果:")
print(kmer_encoded)

结果如下： 

四、与ilearn比较

我先学习一下ilearnPlus平台是如何使用的。下次见