Learning Transferable Visual Models From Natural Language Supervision (CLIP) 逐行源码阅读中文笔记。 blog解读:https://nakaizura.blog.csdn.net/article/details/116903995 原code: https://github.com/openai/CLIP