generated from LufsX/Experiment-LogReg
This commit is contained in:
+25
-4
@@ -56,9 +56,11 @@ def predict_logreg(inX, weights):
|
|||||||
# 将概率转为标签
|
# 将概率转为标签
|
||||||
return 1 if prob >= 0.5 else 0
|
return 1 if prob >= 0.5 else 0
|
||||||
|
|
||||||
|
import numpy as np
|
||||||
|
|
||||||
def load_data(filepath):
|
def load_data(filepath):
|
||||||
"""
|
"""
|
||||||
从文件中加载数据。
|
从文件中加载数据,处理缺失数据,并将 '?' 替换为列的均值。
|
||||||
|
|
||||||
Args:
|
Args:
|
||||||
- filepath (str): 数据文件路径。
|
- filepath (str): 数据文件路径。
|
||||||
@@ -70,18 +72,37 @@ def load_data(filepath):
|
|||||||
data = []
|
data = []
|
||||||
labels = []
|
labels = []
|
||||||
|
|
||||||
|
# 读取数据
|
||||||
with open(filepath, 'r') as file:
|
with open(filepath, 'r') as file:
|
||||||
for line in file:
|
for line in file:
|
||||||
line = line.strip().split()
|
line = line.strip().split()
|
||||||
if not line:
|
if not line:
|
||||||
continue
|
continue
|
||||||
# 将特征转换为浮点型
|
|
||||||
features = list(map(float, line[:-1]))
|
# 将标签提取出来
|
||||||
label = int(line[-1])
|
label = int(line[-1])
|
||||||
|
|
||||||
|
# 将特征转换为浮动数组
|
||||||
|
features = line[:-1] # 去除标签部分
|
||||||
|
# 用特征中的均值替换 '?',并转换为 float
|
||||||
|
features = [float(f) if f != '?' else None for f in features]
|
||||||
|
|
||||||
|
# 将数据和标签分别加入到列表
|
||||||
data.append(features)
|
data.append(features)
|
||||||
labels.append(label)
|
labels.append(label)
|
||||||
|
|
||||||
return np.array(data), np.array(labels)
|
# 将列表转换为numpy数组
|
||||||
|
data = np.array(data)
|
||||||
|
labels = np.array(labels)
|
||||||
|
|
||||||
|
# 计算每列的均值,并用均值替换 None
|
||||||
|
col_means = np.nanmean(data, axis=0) # 计算每列的均值,忽略 NaN
|
||||||
|
# 将 None 替换为对应的列均值
|
||||||
|
for i in range(data.shape[1]):
|
||||||
|
data[:, i] = np.where(np.isnan(data[:, i]), col_means[i], data[:, i])
|
||||||
|
|
||||||
|
return data, labels
|
||||||
|
|
||||||
|
|
||||||
if __name__ == "__main__":
|
if __name__ == "__main__":
|
||||||
# 加载训练集
|
# 加载训练集
|
||||||
|
|||||||
Reference in New Issue
Block a user