yy2yy

> 您好，尚未测试；近期会进行实验 请问您现在进行了“MQbench提供的工具量化yolov5”的测试了吗？请问效果如何呢？我用您提供的命令进行NNIE量化训练测试，会出现一些代码错误，请问您那里是否有新的代码需要更新呢？

> (chineseocr-web) root@cqls:~/chinese-ocr/chineseocr-web# ipython app.py 8090 > Nms engine gpu_nms: None ,cython_nms: None ,python_nms: > 2019-04-18 15:49:18.422556: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled...

> No description provided. 请问你找到了train-mtcnn的资源了吗？

> Hi, > In order to get the `all_all_all.mat` file, according to [LSFM documentation](https://github.com/menpo/lsfm/blob/master/MODELS.md), **YOUR SUPERVISOR** or **FACULTY MEMBER** have to access [this link](https://xip.uclb.com/i/healthcare_tools/LSFM.html?item_id=203&licence_id%5B414%5D=2&basket-action:add=1) and make the request. You can...

Does anyone use MobileNetV3 as backbone get a better result?

谢谢你提供的这个，请问你的handnet模型是在哪里找的？是自己训练的吗？用的什么数据？