Anchor free series network yolox source code line by line explanation four (a total of ten, ensure line by line explanation, after reading, you can change the network at will, not just as a participan

The whole series includes ：Demo Explain the source code line by line ->train The script source code is explained line by line ->backbone Explain the source code line by line ->FPN Explain the source code line by line ->Head Explain the source code line by line ->loss Explain the source code line by line -> The source code of data loading is explained line by line -> The source code of data enhancement is explained line by line ->simOTA Explain the source code line by line . Ensure line by line , Note that line by line , Include python grammar ,tensor Role and application of dimension and line by line code . In fact, there is no mystery about the network structure , It's a stack of modules , There is no reason why you can modify any module . After reading this series, I can do anything about any network structure at will , It's not just limited to one caller .

This article is about YOLOX Data loading and data preprocessing in . When we learn about the network, we always focus on the construction of model structure , And ignore the data processing part . The consequence of this is that there is no accurate understanding of the overall changes in the data . Construct two things ： Data parsers ： Load annotation files and images locally and parse them into the data format required by the network . Sampler ： Complete random sampling in each batch batchsize Operation of images .

---

First of all, I'll go to yolox\exp\yolox_base.py Of get_data_loader Method .

Parameters is_distributed： Whether to conduct distributed training .no_aug：YOLOX In the end 15 individual epoch Whether to enhance the data