建筑数据集

在计算机视觉任务中，大多数时候都涉及到图片的加载、训练以及训练结果的可视化。在实操过程中，经常会遇到图片各种类型之间的转换问题。本文主要是对笔者在学习过程中所遇到的一些问题的总结，以便自己日后查阅方便。如有错误，请不吝赐教。

In most computer vision tasks, core operations involve image loading, model training, and visualization of training results. During practical implementation, issues related to format conversion between different image types are frequently encountered. This paper mainly summarizes the problems encountered by the author during their learning process, for the convenience of future reference. Any corrections or constructive suggestions are welcome if there are any errors.