library

This project explores the applications of LLMs. It uses a "library (bibliotheca)" as the context.

I. Prerequisite

click to see details

• Python 11

II. How to use this repo?

click to see details

1. Download this repo

git clone https://github.com/dujm/library.git

# remove my git directory
rm -rf .git/

# create a new git repository if you need
#git init

2. Create a conda environment (named as e.g.library)

# create an env with Python 11 (see file `environment.yml`)
conda env create --name library --file=environment.yml

# activate env
conda activate library

3. Add conda environment to your jupyter lab (or jupyter notebook)

# add conda environment to jupyter lab
ipython kernel install --user --name=library

# open jupyter lab
jupyter lab

4. Set up Ollama

• Below is for MacOS. Find more instructions on Ollama if you use other operating systems.

a. First-time using Ollma (for Mac users)

• Download file from Ollama website
• Open Ollama app
• Select a model from Model library.
• I selectedllama2 model.
• Download it in the terminal

# pull llama2 model
ollama pull llama2

b. For future use of Ollama

• Open Ollama app
• Or run the bash script in the terminal

bash scripts/ollama_serve.sh

# if you want to stop Ollma in the Mac terminal 
pkill ollama

5. Run notebooks

• Go to notebooks/
• Open a notebook
• Select the kernel library

III. Project Organization

click to see details

├── LICENSE
├── Makefile           <- Makefile with commands like `make data` or `make train`
├── README.md          <- The top-level README for developers using this project.
├── data/              <- Data directory
│
├── docs/              <- A default Sphinx project; see sphinx-doc.org for details
│
├── models/            <- Trained and serialized models, model predictions, or model summaries
│
├── notebooks/         <- Jupyter notebooks. Naming convention is a number (for ordering),
│                         the creator's initials, and a short `-` delimited description, 
│
├── reports/           <- Generated analysis as HTML, PDF, LaTeX, etc.
│
├── requirements.txt   <- The requirements file for reproducing the Python environment 
│
├── environment.yml    <- The environment file for reproducing the conda environment
│
├── setup.py           <- makes project pip installable (pip install -e .) so src can be imported
│
├── src/               <- Source code for use in this project.
│
└── tox.ini            <- tox file with settings for running tox; see tox.readthedocs.io

Thanks to

click to see details

A

• Agentchat RetrieveChat
• Autogen Ollama
• Chromadb
• Langroid
• Two Agent Debates with Tools
• LangChain AI Image Recognition
• Crewai Framework
• AI Crew for Jobs Postings

B

• Cookiecutter Data Science
• Project Gutenberg
• GIFs For Readme

Top