先mark几个地址
看了几个faster rcnn的代码,注释,觉得还好,目前没时间细看,先mark一下
https://www.cnblogs.com/wind-chaser/p/11359948.html
https://blog.csdn.net/qq_32678471/article/details/85678921
https://blog.csdn.net/e01528/article/details/84026513
https://www.cnblogs.com/kerwins-AC/p/9728731.html
https://www.cnblogs.com/houkai/p/6824455.html
还有一个mobilnet的代码
https://github.com/marvis/pytorch-mobilenet/blob/master/main.py
yolo4的
https://blog.csdn.net/weixin_44791964/article/details/106214657
crnn的
keras的https://github.com/felixBrave/ocr_chinese/blob/master/crnn/net/network.py
tensorflow的https://github.com/bai-shang/crnn_ctc_ocr_tf/blob/master/crnn_model/model.py
pytorch:https://github.com/s3nh/text-detector/blob/master/text_reco/models/crnn/crnn.py(是craft+crnn,所以mark)
文本的east和crnn:https://www.cnblogs.com/Allen-rg/p/10420553.html
不定长字符的crnnhttps://www.ctolib.com/topics-142186.html
实战之PCA降维(写的很多都仔细看)
https://blog.csdn.net/wuzhongqiang/article/details/104607832
还有一个‘pytorch基础六(保存、加载模型)’,也都仔细看
https://www.jianshu.com/p/aee6a3d72014
训练tricks:(如使用fast.ai库)https://zhuanlan.zhihu.com/p/41379279
还有一个写tricks https://juejin.im/post/5be5b0d7e51d4543b365da51
一直很想看的MRF,CRF https://juejin.im/post/5b61272ee51d4519721bd228
PPT准备:
大的目录:https://juejin.im/post/5eca4b5ae51d45784f7ffbac
DL为什么需要大量数据:https://juejin.im/post/5b87cda6e51d456e1b5003ae
pytorch资源(有时间一定要看,比如resnet,https://github.com/yunjey/pytorch-tutorial/blob/master/tutorials/02-intermediate/deep_residual_network/main.py#L76-L113)https://zhuanlan.zhihu.com/p/28475866
tensorboard一定要学
https://blog.csdn.net/wuzhongqiang/article/details/105612607