20230406 AI 多模态

1、模型数据

Hugging Face – The AI community building the future.

2、谷歌github

Meta Research · GitHub

3、deit

 GitHub - facebookresearch/deit at colab

4、SAM

GitHub - IDEA-Research/Grounded-Segment-Anything: Marrying Grounding DINO with Segment Anything & Stable Diffusion & BLIP - Automatically Detect , Segment and Generate Anything with Image and Text Inputs

5、智能问答

Slack

6、 vild:

tpu/models/official/detection/projects/vild at master · tensorflow/tpu · GitHub

7.出现了一些基于自监督的方法，这包括基于对比学习的方法如MoCo和SimCLR，和基于图像掩码的方法如MAE和BeiT  
8、Transformer结构及其应用详解--GPT、BERT、MT-DNN、GPT-2 - 知乎

9、clip

重要：

基于CLIP实现的以文搜图 - 飞桨AI Studio

GitHub - yzhuoning/Awesome-CLIP: Awesome list for research on CLIP (Contrastive Language-Image Pre-Training).

GitHub - mlfoundations/open_clip: An open source implementation of CLIP.

GitHub - openai/CLIP: CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

自动学习： 

Prompt工程：自动学习Clip的Prompt - 知乎

10、GPT搭建使用

GitHub - modelscope/modelscope: ModelScope: bring the notion of Model-as-a-Service to life.

wgetcloud->注册openai