提供两个代码：
代码1：在YOLOv5根目录下创建一个py文件，代码如下：
import pandas as pd
import matplotlib.pyplot as plt
# Function to clean column names
def clean_column_names(df):
df.columns = df.columns.str.strip()
df.columns = df.columns.str.replace('\s+', '_', regex=True)
#nonoresult.csv表示原始的结果图,csv文件在runs/train/exp中
original_results = pd.read_csv("D:/pycharm/PythonProject/yolov5/runs/train/exp/result1.csv")
#yesyesresult.csv表示提高后的结果图，csv文件在runs/train/exp中
improved_results = pd.read_csv("D:/pycharm/PythonProject/yolov5/runs/train/exp/result2.csv")
# Clean column names
clean_column_names(original_results)
clean_column_names(improved_results)
# Plot mAP@0.5 curves
plt.figure()
#lable属性为曲线名称，自己可以定义
plt.plot(original_results['metrics/mAP_0.5'], label="Original YOLOv5")
plt.plot(improved_results['metrics/mAP_0.5'], label="Improved YOLOv5")
plt.xlabel("Epoch")
plt.ylabel("mAP@0.5")
plt.legend()
plt.title("mAP@0.5 Comparison")
plt.savefig("mAP_0.5_comparison.png")
# Plot mAP@0.5:0.95 curves
plt.figure()
plt.plot(original_results['metrics/mAP_0.5:0.95'], label="Original YOLOv5")
plt.plot(improved_results['metrics/mAP_0.5:0.95'], label="Improved YOLOv5")
plt.xlabel("Epoch")
plt.ylabel("mAP@0.5:0.95")
plt.legend()
#图的标题
plt.title("mAP@0.5:0.95 Comparison")
#图片名称
plt.savefig("mAP_0.5_0.95_comparison.png")
其中：result1.csv是原始结果；result2.csv是改进后的结果
运行该代码，结果如下：
代码2：新建一个.py文件，代码如下：
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
if __name__ == '__main__':
# 列出待获取数据内容的文件位置
# v5、v8都是csv格式的，v7是txt格式的
result_dict = {
'CIOU': r'D:/pycharm/Python练习项目/picture/pictur/result/CIOU.csv',
'EIOU': r'D:/pycharm/Python练习项目/picture/pictur/result/EIOU.csv',
'DIOU': r'D:/pycharm/Python练习项目/picture/pictur/result/DIOU.csv',
'SIOU': r'D:/pycharm/Python练习项目/picture/pictur/result/SIOU.csv',
}
# 绘制map50
for modelname in result_dict:
res_path = result_dict[modelname]
ext = res_path.split('.')[-1]
if ext == 'csv':
data = pd.read_csv(res_path, usecols=[6]).values.ravel() # 6是指map50的下标（每行从0开始向右数）
else: # 文件后缀是txt
with open(res_path, 'r') as f:
datalist = f.readlines()
data = []
for d in datalist:
data.append(float(d.strip().split()[10])) # 10是指map50的下标（每行从0开始向右数）
data = np.array(data)
x = range(len(data))
plt.plot(x, data, label=modelname, linewidth='1') # 线条粗细设为1
# 添加x轴和y轴标签
plt.xlabel('Epochs')
plt.ylabel('mAP@0.5')
plt.legend()
plt.grid()
# 显示图像
plt.savefig("mAP50.png", dpi=600) # dpi可设为300/600/900，表示存为更高清的矢量图
plt.show()
# 绘制map50-95
for modelname in result_dict:
res_path = result_dict[modelname]
ext = res_path.split('.')[-1]
if ext == 'csv':
data = pd.read_csv(res_path, usecols=[7]).values.ravel() # 7是指map50-95的下标（每行从0开始向右数）
else:
with open(res_path, 'r') as f:
datalist = f.readlines()
data = []
for d in datalist:
data.append(float(d.strip().split()[11])) # 11是指map50-95的下标（每行从0开始向右数）
data = np.array(data)
x = range(len(data))
plt.plot(x, data, label=modelname, linewidth='1')
# 添加x轴和y轴标签
plt.xlabel('Epochs')
plt.ylabel('mAP@0.5:0.95')
plt.legend()
plt.grid()
# 显示图像
plt.savefig("mAP50-95.png", dpi=600)
plt.show()
# 绘制训练的总loss
for modelname in result_dict:
res_path = result_dict[modelname]
ext = res_path.split('.')[-1]
if ext == 'csv':
box_loss = pd.read_csv(res_path, usecols=[1]).values.ravel()
obj_loss = pd.read_csv(res_path, usecols=[2]).values.ravel()
cls_loss = pd.read_csv(res_path, usecols=[3]).values.ravel()
data = np.round(box_loss + obj_loss + cls_loss, 5) # 3个loss相加并且保留小数点后5位（与v7一致）
else:
with open(res_path, 'r') as f:
datalist = f.readlines()
data = []
for d in datalist:
data.append(float(d.strip().split()[5]))
data = np.array(data)
x = range(len(data))
plt.plot(x, data, label=modelname, linewidth='1')
# 添加x轴和y轴标签
plt.xlabel('Epochs')
plt.ylabel('Loss')
plt.legend()
plt.grid()
# 显示图像
plt.savefig("loss.png", dpi=600)
plt.show()
运行该代码：结果如下：
参考博客：用代码yolov5生成改进前后map曲线对比图，map0.5，map0.5:0.95，很简单，小白都能看懂！_map yolov5_qq_42366162的博客-CSDN博客
加油！！