The necessary dependencies have been listed in the Requirements.txt file ...just create a virtual environment and install all the dependencies and then follow along.
To install all the dependencies from Requirements.txt....activate your virtual environment..and then run the following command in the terminal
pip install -r \Requirements.txt
To create the Requirements.txt file ..run the command in the terminall pip freeze > Requirements.txt
LangChain - https://astra.datastax.com/org/f86072ff-9b0c-4f97-bcea-b3224186edd2/integrations/LangChain
The method we are using here is called as RAG(Retrieval Augmented Generation) ....Traditional Language Models generate Text from Scratch, whereas RAG model first retrieve relevant information and then generate responses based on that retrieved context....so in this way RAG models are designed to provide more accuracte responses...
So what basically happening in this project is - we define a template(adventure) based on which our game revolves...then
user can respond with literally anything....and this all chat between AI and User is saved in Cassandra's database...
now whatever user replies...i want my AI model to give a next prompt based on the user reply and the prompt should not
break the template we are using...that's why we again pass the whole template...in which chat_history will keep on updating
as well as human_input...these two will decide what will be the next Prompt...now a particular session is stored at max for
60 mins....after which it automatically gets deleted....now ...when the game ends..we are also instructed our LLM to
include "The End." at the last....so we know when to stop the game...
- Go to the following Link - https://www.datastax.com/lp/astra-registration?utm_medium=youtube_video&utm_source=datastax&utm_campaign=yt_influencers&utm_content=vector_search_tim_ruscica
- SignUp your account
- Click on create database & create a vector database
- Now go to
Connectsection and then generate token and save it in the same directory
- Now download the bundle and save it in the same directory as well
- Activation of the database after it's creattion takes some time..
7. After this follow along the code file `main.py` in the repository.
NOTE - OpenAI API is not free....So once you get the key you can run this game
- Visit the website - https://platform.openai.com/api-keys
- Click on generate secret key and get your key