Steps to install the project in the local machine - cyber-fraud-detection-chatbot/chatbot GitHub Wiki

The following are the steps to install the project in the local machine.

  1. Qdrant is an open-source vector database optimized for storing and querying high-dimensional vectors. Since it is open-source, you can easily use Docker to self-host a Qdrant vector database (more on this in the code walkthrough section).

That said, like other vector databases, it provides fast similarity search even with billions of vectors and also offers extended filtering, which makes it suitable for building RAG systems.

To get started with this step, you first need to install Docker.

Since installation depends on your machine, please head over to the following link and download docker accordingly to your local machine:

https://docs.docker.com/get-started/get-docker/

The step is quite simple. The above link will allow you to download docker. Install it, and you're done!

  1. Once done, run the following command to start a docker container for Qdrant:

docker run -p 6333:6333 -p 6334:6334 -v $(pwd)/qdrant_storage:/qdrant/storage:z qdrant/qdrant

  1. Next, if you head over to this link http://localhost:6333/dashboard, you can access the Qdrant Dashboard. If you are able to see the dashboard means that you have been able to run QDrant successfully in your local machine.

  2. Lastly, Ollama provides a platform to run LLMs locally, giving you control over your data and model usage.

  3. Go to Ollama.com, select your operating system, and follow the instructions.

  4. To verify if Ollama is working fine, run the following command

ollama run llama3.2:1b

At the prompt type something like "What is the capital of America?" and see if you are able to get a response

  1. Finally, run the following pip commands to make sure everything is set up on your machine:

pip install llama-index

pip install llama-index qdrant_client torch transformers

pip install llama-index-embeddings-huggingface

pip install llama-index-llms-ollama

pip install llama-index-vector-stores-qdrant

  1. Launch the Jupyter notebook in a safe environment in Mac and through Anaconda in Windows

  2. Run the commands in project.ipynb