LangChain and Vector Databases in Production

ActiveLoop, Towards AI and the Intel Disruptor Initiative

This contains most of the material from ActiveLoop's Course on LangChain and Vector Databases in Production: https://learn.activeloop.ai/courses/langchain

Authentication/API Keys

While going through the course, I created a keys.py file in the same directory and within the file, constants were set as follows:

OPENAI_API_KEY = '<Your OpenAI API Key>'
ACTIVELOOP_TOKEN = '<Your ActiveLoop Token>'
GOOGLE_API_KEY = '<Your Google API Key>'
GOOGLE_CSE_ID = '<Your Google App ID>'
HUGGINGFACEHUB_API_TOKEN = '<Your HuggingFace Token>'
ELEVEN_API_KEY = '<Your ElevenLabs API Key>'

Other API keys may also be required while going through the course.

The GPT4All model was used in a small portion, and the model weights used are present here

NOTE 1: Some of the generated assets are added under Assets and most of the code is under the Source folders. However, note that during the course, I had placed it all in the main directory itself, so file paths will have to be changed in the respective code notebooks.

NOTE 2: I had taken this course using the free version of the GPT APIs, so there will be some places I was rate limited at 3 reqs/min (A RateLimitError will be raised) - These sections will run properly on paid access to the respective APIs.