計算模型FLOPs和參數量

在pytorch環境下，有兩個計算FLOPs和參數量的包thop和ptflops，結果基本是一致的。

thop

參考https://github.com/Lyken17/pytorch-OpCounter

安裝方法：pip install thop

使用方法：

from torchvision.models import resnet18
from thop import profile
model = resnet18()
input = torch.randn(1, 3, 224, 224) #模型輸入的形狀,batch_size=1
flops, params = profile(model, inputs=(input, ))
print(flops/1e9,params/1e6) #flops單位G，para單位M

用來測試3d resnet18的FLOPs：

model =C3D_Hash_Model(48)
input = torch.randn(1, 3,10, 112, 112) #視頻取10幀
flops, params = profile(model, inputs=(input, ))
print(flops/1e9,params/1e6)

ptflops
參考https://github.com/sovrasov/flops-counter.pytorch

安裝方法：pip install ptflops

或者 pip install git+https://github.com/sovrasov/flops-counter.pytorch.git

import torchvision.models as models
import torch
from ptflops import get_model_complexity_info

with torch.cuda.device(0):
  net = models.resnet18()
  flops, params = get_model_complexity_info(net, (3, 224, 224), as_strings=True, 		                        print_per_layer_stat=True) #不用寫batch_size大小，默認batch_size=1
  print('Flops:  ' + flops)
  print('Params: ' + params)

用來測試3d resnet18的FLOPs：

import torch
from ptflops.flops_counter import get_model_complexity_info
with torch.cuda.device(0):
    net = C3D_Hash_Model(48)
    flops, params = get_model_complexity_info(net, (3,10, 112, 112), as_strings=True, 	                print_per_layer_stat=True)
    print('Flops:  ' + flops)
    print('Params: ' + params)

如果安裝ptflops出問題，可以直接到https://github.com/sovrasov/flops-counter.pytorch.git下載代碼，然后直接把目錄ptflops復制到項目代碼中，通過from ptflops.flops_counter import get_model_complexity_info來調用函數計算FLOPs

python計時程序運行時間

import time

time_start=time.time()
#在這里運行模型
time_end=time.time()
print('totally cost',time_end-time_start)