AI_ChatGPT_Research

Website Link

Question_answering_using_embeddings

https://github.com/openai/openai-cookbook/blob/main/examples/Question_answering_using_embeddings.ipynb?ref=mlq.ai

fine-tuning-gpt-3

https://www.mlq.ai/gpt-3-fine-tuning-key-concepts/ https://www.mlq.ai/fine-tuning-gpt-3-question-answer-bot/

langchain

https://www.mlq.ai/getting-started-with-langchain/

https://python.langchain.com/en/latest/index.html

langchain project guide

https://dagster.io/blog/chatgpt-langchain https://github.com/namuan/dr-doc-search https://simonwillison.net/2023/Jan/13/semantic-search-answers/ https://github.com/slavingia/askmybook

python library : pydantic: 数据验证: 参考: https://www.cnblogs.com/fengqiang626/p/13307771.html typing: 类型约束 manifest-ml: 多个大模型(不包含所有)的统一调用客户端 promptlayer: 记录对openapi的调用, 搜索历史, 性能追踪. unstructured: 若用户在使用 0.4.9 以下版本, 读取到的信息没有metadata字段, 而如果大于等于0.4.9则有 tenacity: 重试

python : dataclass: 合于存储数据对象（data object）的Python类, 参考

prompt-engineering

https://www.mlq.ai/introduction-to-prompt-engineering/

https://www.mlq.ai/prompt-engineering-techniques-improve-reliability/

Microsoft Semantic Kernel

https://github.com/microsoft/semantic-kernel