This is the multi-page printable view of this section. Click here to print.

Return to the regular view of this page.

NeuroDesk Copilot

NeuroDesk Copilot: How to use LLMs for code autocompletion, chat support in NeuroDesk ecosystem

1 - NeuroDesk Copilot using Github Copilot

NeuroDesk Copilot: How to use LLMs for code autocompletion, chat support in NeuroDesk ecosystem

Neurodesk Copilot: Using Github Copilot inside NeuroDesk Environment

Step 1: Install Jupyter AI and Jupyter lab

conda create -n jupyter-ai-intelligence python=3.12
conda activate jupyter-ai-intelligence
pip install jupyterlab gitpython
pip install notebook_intelligence

Start Jupyter lab

jupyter lab

Step 2: Login with Github and follow the Instruction below:

Login with Github

Step 3: Use chat interface

Chat feature

Step 4: Code completion

Code completion

Step 5: Generate code

Generate Code

Configuring LLM Provider and models

You can configure the model provider and model options using the Notebook Intelligence Settings dialog. You can access this dialog from JupyterLab Settings menu -> Notebook Intelligence Settings, using /settings command in Copilot Chat or by using the command palette.

2 - NeuroDesk Copilot using Local LLMs

NeuroDesk Copilot: How to use LLMs for code autocompletion, chat support in NeuroDesk ecosystem

Neurodesk Copilot: Using Locally hosted LLMs inside Neurodesk Environment

Step 1: Install Jupyter AI and Jupyter lab

conda create -n jupyter-ai python=3.12
conda activate jupyter-ai
pip install jupyterlab gitpython
pip install 'jupyter-ai[all]'

Step 2: Deploying Neurodesk Copilot model with Ollama for Linux (For Windows and Mac users: download Ollama from https://ollama.com/)

curl -fsSL https://ollama.com/install.sh | sh 
git clone https://github.com/jnikhilreddy/jupyter-ai-neurodesk.git
cd jupyter-ai-neurodesk/
wget https://huggingface.co/jnikhilreddy/neurodesk-gguf/resolve/main/neurodesk.gguf?download=true -O neurodesk.gguf
ollama create  neurodesk -f ./Modelfile  
#Optional: To make Neurodesk copilot faster (quantize the Neurodesk model): ollama create --quantize q4_K_M neurodesk -f ./Modelfile 
ollama run neurodesk

Start Jupyter lab with Neurodesk Copilot

jupyter lab

Step 3: Choose Neurodesk copilot in Jupyter AI: Press on settings and choose Ollama and neurodesk model and save settings

Choose Jupyter-AI settings Enable Neurodesk copilot in Jupyter-AI settings

Feel free to update the settings to disable auto completer to manual invocation in Settings -> Settings Editor -> Inline Completer