Seven.Hsu

谢谢，问题解决了 Sent from my iPhone > On Apr 8, 2018, at 10:54, withstars wrote: > > 1.启动数据库服务，导入项目根目录下library.sql文件，数据库创建完成。 > 2.修改src/main/resources/book-context.xml文件里的数据源配置。 > > — > You are receiving this because you authored...

https://github.com/sevenHsu/LicensePlateChars Here are some simulated chinese license plate character images.It can be pasted on the CCPD images for License plate recognize.

@getsanjeev Yep!Just Replace 15th line code `temp = os.path.abspath(str(np.random.randint( 1, 1000000)) + 'ntl.txt').replace('\\', '/')` in this file "pylsd/pylsd/lsd.py" to `temp='a'`. I just run this code on windows

line90 和 line101的两个with torch.no_grad()代码段看来确实没实际意义，后面更新meta时也没用到这两段的loss，单纯记录了下loss和correct数量。然后就是计算train acc 使用了这两个部分的correct 数量，让大家看看从随机参数的acc到迭代1、2 ... K轮后的acc变化，再到后面每个batch初始化上一个batch更新的meta参数初始化的acc变化。

Actually，your test set will be finetuned (train 10 steps for each support set) after every 500 steps training on training set. So you can see the acc growth from 0,2615...

They were randomly selected from each task test dataset.