计算Transformer的Flops

计算GCN的Flops

计算CNN的Flops

计算ResNet的Flops

计算MobileNet的Flops

计算Transformer的Flops

核心程序：

    model = torch.load(path)calculate_flops = 1if calculate_flops:total_params = sum(p.numel() for p in model.parameters())print("Total parameters:", total_params)trainable_params = sum(p.numel() for p in model.parameters() if p.requires_grad)print("Trainable parameters:", trainable_params)# 确保模型构建完成input_shape = (1, x.shape[1])  # batch size = 1macs, params = get_model_complexity_info(model, input_shape, as_strings=True,print_per_layer_stat=True, verbose=True)print(f"FLOPs: {macs}")print(f"Parameters: {params}")

每一模块的Flops计算结果，其中1Flops=2MACs=1次乘法+1次加法

最终结果：

完整程序，仅供参考：

import argparse
import torch
import h5py
import numpy as np
from transformer.Models import Transformer, MLP
import os
import re
import torch.nn as nn
from torch.nn import functional as F
from tensorboardX import SummaryWriter
import time
from ptflops import get_model_complexity_infoN = 30
snr = 40
learning_rate = 0.003
M = 3000
epochs = 10000
dataset_n = 100def graph_normalize(signals):for i in range(signals.shape[0]):signals[i, :] = signals[i, :]/np.max(np.abs(signals[i, :]))def generate_dataset():base_dir = "D:\无线通信网络认知\论文1\大修意见\Reviewer2-7 多种深度学习方法对比实验\\test data 30 (mat)\\"mat_file = h5py.File(base_dir  + '30_nodes_dataset.mat', 'r')# 获取数据集signals = mat_file["Signals"][()]tp = mat_file["Tp"][()]tp_list = mat_file["Tp_list"][()]Signals = np.swapaxes(signals, 2, 0)Tp = np.swapaxes(tp, 2, 0)# tp_list = tp_list - 1# 关闭文件mat_file.close()#把每张图的数据和标签都装进去x_list = []y_list = []for n in range(0,dataset_n,1):# print("n: ",n)signals = Signals[n,:,:]graph_normalize(signals)L = 2 * signals.shape[1]  # 待分析的2个信号拼在一起tp = Tp[n,:,:]# x 的形状为 (100000, 3600)x = np.zeros((N * N, L))# y 的形状为 (100000, 1)y = np.zeros((N * N, 1))# 生成 x 和 yindex = 0#处理正样本for i in range(N):for j in range(N):if i!=j and tp[i, j]==1:combined_signal = np.concatenate((signals[i, :], signals[j, :]))x[index, :] = combined_signaly[index, 0] = 1index += 1#算负样本总共有多少n_pair_list = []for i in range(N):for j in range(N):if i!=j and tp[i, j] == 0:n_pair_list.append((i,j))np.random.seed(42)indices = np.arange(len(n_pair_list))np.random.shuffle(indices)n_pair_list = np.array(n_pair_list)n_pair_list = n_pair_list[indices]#根据正样本数取负样本n_pair = n_pair_list[:index,:]for k in range(n_pair.shape[0]):i = n_pair[k, 0]j = n_pair[k, 1]combined_signal = np.concatenate((signals[i, :], signals[j, :]))x[index, :] = combined_signaly[index, 0] = 0index += 1x = x[:index,:]y = y[:index, :]# x = np.expand_dims(x, axis=-1)# plt.plot(np.linspace(0, 1, 2520), np.hstack((x[1, :1260], x[2, 1260:])))# c = 70# plot_sig = np.vstack((x[c, :1260], x[c, 1260:]))# print("y: ",1-y[c,0])# plt.plot(np.linspace(0, 1, 1260), plot_sig.T)x_list.append(x)y_list.append(y)x = np.vstack(x_list)y = np.vstack(y_list)return x, ydef generate_time_dataset():base_dir = "D:\无线通信网络认知\论文1\大修意见\Reviewer2-7 多种深度学习方法对比实验\\test data 30 (mat)\\"mat_file = h5py.File(base_dir + '30_nodes_dataset.mat', 'r')signals = mat_file["Signals"][()]tp = mat_file["Tp"][()]# 获取数据集Signals = np.swapaxes(signals, 2, 0)Tp = np.swapaxes(tp, 2, 0)# tp_list = tp_list - 1# 关闭文件mat_file.close()x_list = []for n in range(0,dataset_n,1):# print("n: ",n)signals = Signals[n, :, :]L = 2 * signals.shape[1]  # 待分析的2个信号拼在一起tp = Tp[n, :, :]# x 的形状为 (100000, 3600)x = np.zeros((N * N, L))# y 的形状为 (100000, 1)y = np.zeros((N * N, 1))# 生成 x 和 yindex = 0for i in range(N):for j in range(N):combined_signal = np.concatenate((signals[i, :], signals[j, :]))x[index, :] = combined_signalindex += 1x_list.append(x)x = np.vstack(x_list)return xdef cal_performance(tra_pred,tra_true):return F.mse_loss(tra_pred,tra_true)def train(model, data, label, optimizer):best_acc = 0count = 0# model = torch.load(model_dir + 'model.pt')for epoch in range(epochs):# if epoch!=0:count = count + 1if count>50:breakoptimizer.zero_grad()  # 清零优化器梯度，梯度不清零会一直存在# score = score.to(device)correct_count = 0# pred = model(before_track_data.get(p).to(device), after_track_data.get(q).to(device))pre = model(data)loss = loss_function(pre, label)  # 计算一次损失# loss = loss_function(pre_1.float(), data_1.y.float())# loss = loss_function(pre_1, data_1.y)# loss反向传播就行，这里没有acc监视器loss.backward()# print(" ")# 用反向传播得到的梯度进行参数更新optimizer.step()# 计算准确率with torch.no_grad():# 输出是概率，转换成0/1预测值pred_class = (pre >= 0.5).float()correct = (pred_class == label).sum().item()total = label.size(0)accuracy = correct / totalif accuracy > best_acc:best_acc = accuracytorch.save(model, model_dir + 'epoch '+ str(round(epoch)) +' accuracy '+str(round(accuracy,3))+'.pt')count = 0print("epoch: ", epoch, "  loss: ", loss.item(), "accuracy: ", accuracy)def find_max_val_acc_file(folder_path):max_val_acc = -1max_file = None# 正则表达式匹配 'val_acc_' 后面的数字# pattern = re.compile(r'val_accuracy_([0-9]+\.[0-9]+)')pattern = re.compile(r'accuracy ([0-9]+\.[0-9]+)')for filename in os.listdir(folder_path):match = pattern.search(filename)if match:val_acc = float(match.group(1))if val_acc > max_val_acc:max_val_acc = val_accmax_file = filenamereturn max_file, max_val_accdef test(data, label):base_path = r'D:\english\WCNA\Transformer\fn 59 8\\'max_file, max_val_acc = find_max_val_acc_file(base_path)path = base_path + max_filemodel = torch.load(path)calculate_flops = 1if calculate_flops:# model.summary()# print("Total parameters:", model.count_params())total_params = sum(p.numel() for p in model.parameters())print("Total parameters:", total_params)trainable_params = sum(p.numel() for p in model.parameters() if p.requires_grad)print("Trainable parameters:", trainable_params)# 确保模型构建完成input_shape = (1, x.shape[1])  # batch size = 1# input_shape = x.shape[1]# input_shape = (1, x_val.shape[1])  # 例如 (batch_size=1, sequence_length=3600)macs, params = get_model_complexity_info(model, input_shape, as_strings=True,print_per_layer_stat=True, verbose=True)print(f"FLOPs: {macs}")print(f"Parameters: {params}")# print(f"FLOPs: {flops / 10 ** 6:.2f} MFLOPs")pre = model(data)# loss = loss_function(pre, label)with torch.no_grad():# 输出是概率，转换成0/1预测值pred_class = (pre >= 0.5).float()correct = (pred_class == label).sum().item()total = label.size(0)accuracy = correct / totalprint("accuracy: ", accuracy)def calculate(data):base_path = r'D:\englist\WCNA\Transformer\fn 59 8\\'max_file, max_val_acc = find_max_val_acc_file(base_path)path = base_path + max_filemodel = torch.load(path)t1 =time.time()pre = model(data)t2 = time.time()delta_t = (t2 - t1)/100 #100个样本print("time: ", delta_t)if __name__ == '__main__':x, y = generate_dataset()indices = np.arange(y.shape[0])np.random.shuffle(indices)x = x[indices]y = y[indices]# x_down = x[:, ::2]a = round(0.8 * len(indices))# b = 12000x_train = x[:a, :]y_train = y[:a, :]x_val = x[a:, :]y_val = y[a:, :]x_train = torch.tensor(x_train).float()y_train = torch.tensor(y_train).float()x_val = torch.tensor(x_val).float()y_val = torch.tensor(y_val).float()device = "cuda:0"x_train = x_train.to(device)y_train = y_train.to(device)x_val = x_val.to(device)y_val = y_val.to(device)log_writer = SummaryWriter()# os.environ['CUDA_LAUNCH_BLOCKING'] = '1'model_dir = "D:\englist\WCNA\Transformer\\fn 7 4\\"loss_function = nn.BCELoss()  # 损失函数parser = argparse.ArgumentParser()parser.add_argument('-epoch', type=int, default=epochs)parser.add_argument('-b', '--batch_size', type=int, default=40)parser.add_argument('-d_model', type=int, default=7)parser.add_argument('-d_inner_hid', type=int, default=236)parser.add_argument('-d_k', type=int, default=64)parser.add_argument('-d_v', type=int, default=64)parser.add_argument('-warmup', '--n_warmup_steps', type=int, default=2000)parser.add_argument('-lr_mul', type=float, default=2.0)parser.add_argument('-lr', type=float, default=0.001)parser.add_argument('-n_head', type=int, default=2)parser.add_argument('-n_layers', type=int, default=1)parser.add_argument('-dropout', type=float, default=0.1)parser.add_argument('-do_train', type=bool, default=False)parser.add_argument('-do_retrain', type=bool, default=False)parser.add_argument('-do_eval', type=bool, default=True)parser.add_argument('-use_mlp', type=bool, default=False)parser.add_argument('-calculate', type=bool, default=False)opt = parser.parse_args()opt.d_word_vec = opt.d_model# device = "gpu:0"# device="cpu"transformer = Transformer(2000,2000,d_k=opt.d_k,d_v=opt.d_v,d_model=opt.d_model,d_word_vec=opt.d_word_vec,d_inner=opt.d_inner_hid,n_layers=opt.n_layers,n_head=opt.n_head,dropout=opt.dropout,n_position = 250).to(device)mlp = MLP(10,10,25,50,use_extra_input=False).to(device)model_train = transformerif opt.use_mlp:model_train = mlpif opt.do_train == True:parameters = mlp.parameters() if opt.use_mlp else transformer.parameters()# optimizer = ScheduledOptim(#     optim.Adam(parameters, betas=(0.9, 0.98), eps=1e-09),#     opt.lr, opt.d_model, opt.n_warmup_steps, opt.use_mlp)lr = 0.001optimizer = torch.optim.Adam(model_train.parameters(), lr=lr)if opt.do_retrain == True: # only used for transformercheckpoint = torch.load("./checkpoint/ckpt.pth")transformer.load_state_dict(checkpoint['net'])optimizer.load_state_dict(checkpoint['optimizer'])train(model=model_train,data=x_train,label=y_train,optimizer=optimizer)if opt.do_eval == True:test(data=x_val,label=y_val)if opt.calculate == True:x = generate_time_dataset()x = torch.tensor(x).float()x = x.to("cuda:0")calculate(data=x)# model = torch.load(model_dir + 'model.pt')