LangChain-Yoda is a collection of various agents and tools built based on Langchain to explore the capabilities of
- natural language processing using open-source libraries and APIs.
- kinds of tools and agents build by Langchain
- Embeddings and index capability
Slides
export GOOGLE_CSE_ID=""
export GOOGLE_API_KEY=""
export SERPER_API_KEY=""
export MODEL=""pip install -r requirements.txtThe repository has the following file structure:
LangChain/
├── agent/
│ ├── bashbot.py
│ ├── googlesearch.py
│ ├── googlesearch_chatbot.py
│ └── langchain_agent.py
├── gpt4all/
│ └── gpt4all.py
├── index/
│ ├── blog.txt
│ ├── langchain_index.py
│ └── state_of_the_union.txt
├── lib/
│ └── proxy.py
├── llama-cpp/
│ ├── llama.py
│ └── llama_embed.py
├── main.py
├── openai/
│ └── langchain_openai.py
├── requirements.txt
└── start/
└── langchain_start.py
The agent/ directory contains various agents built for specific tasks in natural language processing, including a Bashbot that executes shell commands, a Googlesearch that provides search results from Google, and a Langchain Agent that performs various NLP-related tasks.
The gpt4all/ directory contains Python scripts for interacting with the OpenAI GPT-3 API.
The index/ directory contains sample text files designed for language processing, including a blog post and a transcript of President Barack Obama's 2009 inaugural address.
The lib/ directory contains helper modules.
The llama-cpp/ directory contains Python scripts that use the Llama Embeddings Module to perform NLP-related tasks.
The openai/ directory contains Python scripts for interacting with the OpenAI GPT-3 API.
requirements.txt contains a list of Python packages required to run the scripts in the repository.
The start/ directory contains Python scripts for launching the various agents and tools in LangChain.
To get started with LangChain, clone the repository and install the required packages by running:
pip install -r requirements.txt
Then, navigate to the start/ directory and choose one of the Python scripts to launch an agent or tool. For example, to launch the Langchain Agent, run:
python langchain_start.py
The code in this repository is licensed under the MIT License. Please read the LICENSE file in the root directory for more information.