Taylor King

Results 7 comments of Taylor King

@Springlee-P I guess you can use this code `python train.py -m MNet(or other model ) `to generate a checkpoint_epoch_0 file.

Could you please release the train.prototxt and pre-train model?Thanks a lot!

@xiaoweihappy123 Me too.My result is `Single Query: [cmc1: 45.07%], [cmc5: 66.80%], [cmc10: 76.07%] mAP = 0.50221015378 Done, 3.83s ` I guess if my mAP calculation method is wrong.I wonder what...

@xiaoweihappy123 Could you please tell me the paper's title?Thanks a lot!

看起来像是你的cudnn版本没有匹配上

> 我发现有个地方的确有问题。我不知道怎么改? > Training文件夹下还是“151_tr.jpg”,而程序似乎要“151.jpg”。 > ![image](https://user-images.githubusercontent.com/10806595/55875248-33f7ad00-5bc7-11e9-9a9c-23604f363990.png) > “ir_train.txt”下截图可以看出,运行fileList.py后,它是“151.jpg”,如下。 > ![image](https://user-images.githubusercontent.com/10806595/55875305-60abc480-5bc7-11e9-88bc-0c1006686825.png) > 问题应该是这个,我不知道怎么回事。@SoftwareGift 我通过rgb.write(pwd + list[0].split('.jpg')[0]+ '_tr' + '.jpg' +'\n')修改了文件名,但是仍然运行不了训练指令,请问您最后是如何解决的呢?

> @ TianSong1991 > > https://competitions.codalab.org/competitions/20853 > > 你应该注册。然后发送电子邮件给组织者,他们将发送下载URL或数据集。 这个网站已经挂了,在CSDN上我已经找到了训练集和测试集,请问如何获取验证集呢?