计算模型 FPS

无论是图像分类也好，目标检测/实例分割也罢，检测速度都是一个重要指标。

计算模型FPS过程：
只需要在模型运行前后各添加一个 time记录时间即可，然后通过时间差计算出模型 FPS

start = time.time()
with torch.no_grad():
	seg_result = model.forward(img, target)
img_show = show_result_ins(imgpath, seg_result)
end = time.time()
t_all = end - start
# print("spend time: ", t_all ,"s")
print('average time:{:.02f} s'.format(np.mean(t_all) / 1))
print('average FPS :{:.02f} fps'.format(1 / np.mean(t_all)))

完整代码可以参考如下

import time
import torch
import numpy as np

net = build_model
net.eval()

# x是输入图片的大小
x = torch.zeros((1,3,H,W)).cuda()
t_all = []

for i in range(100):
    t1 = time.time()
    y = net(x)
    t2 = time.time()
    t_all.append(t2 - t1)

print('average time:', np.mean(t_all) / 1)
print('average fps:',1 / np.mean(t_all))

print('fastest time:', min(t_all) / 1)
print('fastest fps:',1 / min(t_all))

print('slowest time:', max(t_all) / 1)
print('slowest fps:',1 / max(t_all))