方法一:
思路:将txt文件读成一个字符串,在整个字符串中进行匹配,匹配到的是数组,再从数组的对应位置取数据
txt样例:
------begin checkaccuracy---------
Validating batch 10
Validating batch 20
Validating batch 30
Validating batch 40
Validating batch 50
Validating batch 60
Total Top1 Accuracy: 70.40%
Total Top5 Accuracy: 89.20%
FPS is: 1144.3161883555188
Run with precision fp32, batchsize 16
---------06221903 ---> 0622-190406 Total:42 seconds
目标:需要提取TOP1、TOP5、FPS、以及最后一行的时长数据
代码实现:
top1=re.findall(r"Total Top1 Accuracy: (.+?)\n", content)[0]top5=re.findall(r"Total Top5 Accuracy: (.+?)\n", content)[0]fps=re.findall(r"FPS is: (.+?)\n", content)[0]time = re.findall(r"Total:(.+?)\n", content)[0]
实现结果(写入到了excel中):
方法二:
思路:将txt文件逐行读出,按照字符串中数据的特定切分字符串,如冒号,切分成多部分后取对应位置的数据,如果数据中包含不想要的字符,可以用空字符替代对应字符
txt样例-同上:
------begin checkaccuracy---------
Validating batch 10
Validating batch 20
Validating batch 30
Validating batch 40
Validating batch 50
Validating batch 60
Total Top1 Accuracy: 70.40%
Total Top5 Accuracy: 89.20%
FPS is: 1144.3161883555188
Run with precision fp32, batchsize 16
---------06221903 ---> 0622-190406 Total:42 seconds
代码:
with open(file, 'r') as f:lines = f.readlines() # 读取所有行first_line = lines[0] # 取第一行last_line = lines[-1] # 取最后一行last5_line = lines[-5] # 取最后一行str="FPS"if str in last5_line:FPS=last5_line.split(": ")[1]top1 = lines[-7].split(": ")[-1]top5 = lines[-6].split(": ")[-1]#top1 = last_line.split(' ')[-2].split(":")[1].replace('Prec', '')#top5 = last_line.split(' ')[-1].split(":")[-1]sheet.write(i, 3, FPS)sheet.write(i, 4, top1)sheet.write(i, 5, top5)
处理结果:同上
方法三:正则匹配--todo