Examples - davidmarsoni/llog GitHub Wiki

This section contains examples of our research and development process concerning the LlamaIndex framework, the Notion API, and more.

To make it easier to understand, we have divided our research into multiple IPYNB notebooks. Each notebook contains a specific example of a part of our research or a specific use case. The notebooks are organized in a way that allows you to follow our research process step by step.

:warning: Important Notes

Some notebooks can ask you to have some API keys to run the code. If you don't know how to get them, you can refer to the main readme file of the project available by clicking on the link below:

Main page of the project

We recommend viewing the different notebooks in the order they are presented below. This will allow you to understand the different parts of our research and how they are related to each other.

Pre-setup

In order to help you to run all the notebooks that require the Notion integration token you can copy the following token that will allow you to run the notebooks without having to create your own token.

NOTION_INTEGRATION_TOKEN=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

[!NOTE] This token allows you to only read the content of the pages needed for the notebooks. It does not allow you to write or modify the content of the pages. This token is only used for testing purposes and should not be used anywhere else.

:books: Notion Connection

The first notebook is about the connection to the Notion API. This notebook will show you how to connect to the Notion API and get the content of a page. It will also show you how to use the Notion API to get the content of a page or database and how to use it with LlamaIndex.

This notebook was one of our first tests with the Notion API. It was used to test the connection to the Notion API and how to get the content of a page.

Here is the link to the notebook:

Notion Connection

:books: Simple Caching System

The second notebook talks about a more simplified caching system than the one we have in the project. This notebook will show you how to create a simple caching system that can be used to cache the content of a page. Then it will use this caching system to get the content of a page and use it with LlamaIndex.

Here is the link to the notebook:

Simple caching system

:books: Simple PDF query

This notebook covers a test we have done with PDF files. We have tried to use the LlamaIndex library to query a PDF file by creating sub-questions.

Here is the link to the notebook:

Simple PDF query

:books: Simple agent system

This notebook will help you to understand how to create a simple agent with LlamaIndex. This notebook will show you how to create a simple agent that can be used to perform a specific task. It will also show you how to use this agent with LlamaIndex.

Here is the link to the notebook:

Simple agent system

:books: Multi-agent collaboration

Our last notebook is about multi-agent collaboration. This notebook will show you how to create a multi-agent system that can be used to perform complex tasks.

Here is the link to the notebook:

Multi agent collaboration

:checkered_flag: Conclusion

These notebooks are a good highlight of our research and the different main parts of our project. They are not exhaustive and do not cover all the parts of our project, but they are a good starting point to understand the different parts of our project and how they are related to each other.

For example, we have not covered how we connect to a Google Cloud bucket or how we manage the files in the Google Cloud bucket. This is because we think this isn't the core part of our project, and we wanted to focus on the main parts of our project.