分类任务绘制混淆矩阵

最新推荐文章于 2024-07-22 20:35:33 发布

遥感加油呀

最新推荐文章于 2024-07-22 20:35:33 发布

阅读量1.8k

点赞数 1

文章标签：分类矩阵 matplotlib

本文链接：https://blog.csdn.net/weixin_45276304/article/details/130735493

版权

该代码片段导入了numpy和matplotlib库，用以读取CSV文件中的混淆矩阵数据并生成蓝调热力图。它设置了字体大小和类型，定义了类标签，然后创建了一个14x12大小的图像，并以lower为原点显示了热力图。此外，还添加了颜色条，轴标签，标题以及每个单元格的值，并反转了y轴。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

import numpy as np
import matplotlib.pyplot as plt
import matplotlib as mpl

# Set font size and family
mpl.rcParams['font.size'] =12
mpl.rcParams['font.family'] = 'Times New Roman'

# Load confusion matrix from CSV file
confusion_matrix = np.loadtxt('../confusion_matrix.csv', delimiter=',')

# Define class labels
labels = ['Airport', 'BareLand', 'BaseballField', 'Beach', 'Bridge', 'Center', 'Church', 'Commercial', 'DenseResidential', 'Desert', 'Farmland', 'Forest', 'Industrial', 'Meadow', 'MediumResidential', 'Mountain', 'Park', 'Parking', 'Playground', 'Pond', 'Port', 'RailwayStation', 'Resort', 'River', 'School', 'SparseResidential', 'Square', 'Stadium', 'StorageTanks', 'Viaduct']

# Set figure size
fig, ax = plt.subplots(figsize=(14,12))

# Plot heatmap with origin set to 'lower'
im = ax.imshow(confusion_matrix, cmap='Blues', extent=[0, len(labels), 0, len(labels)], origin='lower')

# Set ticks at the middle of each cell
ticks = np.arange(len(labels))
plt.xticks(ticks+0.5, labels, rotation=90, fontsize=12)
plt.yticks(ticks+0.5, labels, fontsize=12)
# Hide tick marks
plt.tick_params(axis='both', which='both', length=0)

# Add color bar
cbar = ax.figure.colorbar(im, ax=ax, shrink=0.8)




# Set axis labels and title
plt.xlabel('Predicted Labels', fontsize=16)
plt.ylabel('True Labels', fontsize=16)
plt.title('Confusion Matrix',fontsize=17)



# Add text labels to each cell
for i in range(len(labels)):
    for j in range(len(labels)):
        if confusion_matrix[i, j] != 0:
            text = ax.text(j + 0.5, i + 0.5, '{:.1f}'.format(confusion_matrix[i, j]), ha='center', va='center', color='black', fontsize=10)

# Invert y-axis
ax.invert_yaxis()

# Show plot
plt.show()