YOLO源码详解(一)-训练_数据之美的博客-CSDN博客

来自 : lcqbit11.blog.csdn.net/article 发布时间：2021-03-25

int classes l.classes; //jitter是什么意思呢 可以参考这篇博客 [非均衡数据集处理 利用抖动(jittering)生成额外数据](http://weibo.com/1402400261/EgMr4vCC2?type comment#_rnd1478833653326) float jitter l.jitter; list *plist get_paths(train_images); //int N plist- size; char **paths (char **)list_to_array(plist); load_args args {0}; args.w net.w; args.h net.h; args.paths paths; //n就是一次加载到内存中的图片数量 args.n imgs; //m是待训练图片的总数量 args.m plist- size; args.classes classes; args.jitter jitter; //7*7个网格 args.num_boxes side; args.d buffer; args.type REGION_DATA; //调节图片旋转角度、曝光度、饱和度、色调等 来增加图片数量 args.angle net.angle; args.exposure net.exposure; args.saturation net.saturation; args.hue net.hue; //声明线程ID pthread_t load_thread load_data_in_thread(args); clock_t time; //while(i*imgs N*120){ while(get_current_batch(net) net.max_batches){ i 1; time clock(); /*pthread_join()函数 以阻塞的方式等待thread指定的线程结束。当函数返回时 被等待线程的资源被 收回。如果线程已经结束 那么该函数会立即返回。*/ pthread_join(load_thread, 0); train buffer; load_thread load_data_in_thread(args); printf( Loaded: %lf seconds\\n , sec(clock()-time)); time clock(); //开始训练 float loss train_network(net, train); if (avg_loss 0) avg_loss loss; avg_loss avg_loss*.9 loss*.1; printf( %d: %f, %f avg, %f rate, %lf seconds, %d images\\n , i, loss, avg_loss, get_current_rate(net), sec(clock()-time), i*imgs); //每100次或者1000次保存一次权重 if(i%1000 0 || (i 1000 i%100 0)){ char buff[256]; sprintf(buff, %s/%s_%d.weights , backup_directory, base, i); save_weights(net, buff); free_data(train); char buff[256]; sprintf(buff, %s/%s_final.weights , backup_directory, base); save_weights(net, buff); 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117

看到这里 我想你已经能可以自己来训练VOC数据集了。但这只是刚刚开始 我们要想将YOLO移植到mxnet下 就需要了解其具体函数的实现过程 请看下一节内容。

(END)