Working and explanation of LangChain, an open-source framework that allows users to build applications using LLMs like GPT4 and GPT 3.5 Turbo.
LangChain is an open-source framework that allows developing end-to-end applications powered by large language models.
The framework currently offers a Python or a JavaScript package, Typescript to be specific.
If we want to build an app on top of ChatGPT or any other LLM and then we wish to combine our own data (eg. a database of emails, PDFs, excel sheets) to that while also constructing prompts based on basic user input.
Langchain also offers to store conversation history.
LangChain allows building applications with LLMs through composability.
Currently, it provides 6 different key functionality:
- Models
- Prompts
- Chains
- Memory
- Indexes
- Agents and Tools
Learn about each one in ./Notes
directory
Access the notes. The notes contain code as well as it's explaination
- Clone the repository
git clone https://github.com/vaniseth/LangChain-Explained.git
- Change the directory to the project directory
cd LangChain-Explained
- Intall the dependencies
pip install -r requirements.txt