Run Jupyter Notebooks
Fast-track AI development with Jupyter Notebooks from the NVIDIA NGC catalog.
What Is a Jupyter Notebook?
A Jupyter Notebook, much like a laboratory notebook, provides the easiest way for data scientists and developers to iterate, implement, and document their code in multiple programming languages, including Python. Jupyter Notebooks can be easily shared, allowing anyone to run the code on their end with minimal effort.
Jupyter Notebooks from the NVIDIA NGC Catalog
The NVIDIA® NGC™ catalog, a hub for GPU-optimized AI and high-performance software, offers hundreds of Python-based Jupyter Notebooks for various use cases, including machine learning, computer vision, and conversational AI.
These Jupyter Notebooks allow data scientists to examine, understand, customize, test, and build their own models faster, while taking advantage of best practices in the Notebooks to build high-throughput, low-latency models.
Explore NGC's Jupyter Notebooks
Quick Deploy
The quick deploy feature in the NGC catalog automatically sets up the Vertex AI instance with an optimal configuration, preloads the dependencies, runs the software from NGC without any need to set up the infrastructure.
Deploy popular DL and ML containers, models, and SDKs directly from the NGC catalog.
Download a Jupyter Notebook for Your Use Case
Get started today with Jupyter Notebooks that span diverse use cases, including medical imaging, conversational AI, computer vision, and more.
BERT for TensorFlow
Explore a worked example for utilizing the BERT for TensorFlow model scripts for question answering.
Question Answering
Build an end-to-end workflow for question answering starting with training in NVIDIA TAO Toolkit and deploying using NVIDIA Riva.
Image Classification
Learn how to install the Docker Engine on your system, pull and run the TensorFlow container, and execute image classification tasks.
Text Classification
See an end-to-end sample workflow for text classification starting with training in NVIDIA TAO Toolkit and deploying using NVIDIA Riva.
Medical Imaging
Learn how 3D-UNet medical image segmentation allows for seamless segmentation of 3D volumes with high accuracy and performance.
Image Segmentation
Explore how an NVIDIA pretrained U-Net model is adapted from the original version, which is a convolutional auto-encoder for 2D image segmentation.
Conversational AI
Learn how to use PyTorch Lightning and NVIDIA Nemo™ to create new conversational AI models.
Computer Vision
Access a collection of Notebook examples that can detect people, recognize human action, detect gaze, and more with the NVIDIA TAO Toolkit.
How to Run Jupyter Notebooks
Jupyter Notebooks from the NGC catalog can be deployed in the cloud or on premises.
Run in the Cloud
The NGC catalog provides a one-click deploy approach for setting up a Jupyter environment on Google Cloud Vertex AI, simplifying deployment so data scientists can focus on AI development.
Instructions for running a Jupyter Notebook from the NGC catalog.
- Log in to the NGC catalog.
- Identify the deep learning framework, SDK, or AI model to deploy from the catalog, and open the product page.
- Select “Vertex AI.”
- Click “Deploy on JupyterLab.”
This will launch the JupyterLab instance on the selected infrastructure with optimal configuration, preload the software dependencies as a kernel, and download the Jupyter Notebook from the NGC catalog in essentially one click.
Run Anywhere
Jupyter Notebooks from the NGC catalog can run on GPU-powered on-prem systems, including NVIDIA DGX™, as well as on cloud instances.
Instructions for running a Jupyter Notebook from the NGC catalog.
- Download the Jupyter Notebook from NGC.
- Make sure that your system has the requirements mentioned in the NGC resource.
- Upload the Jupyter Notebook inside the JupyterLab.
- Execute the Notebook.
Here are a few step-by-step guides on getting started with NGC’s Jupyter Notebooks: image segmentation, recommender system, medical imaging.
Jupyter Notebook Tutorials
Learn how to run a Jupyter Notebook for your use case with these step-by-step, technical tutorials.
PyTorch Lightning Tutorial
Build speech models with PyTorch Lightning on NVIDIA GPU-powered AWS instances using the Grid, NGC, and PyTorch Lightning Jupyter Notebook tutorial.
Image Segmentation Tutorial
Learn how to use a sample image segmentation Notebook to identify defective parts in a manufacturing assembly line.
Recommender System Tutorial
Use a sample recommendation system Notebook to predict a user’s rating of a movie and recommend movies.
Medical Image Segmentation Tutorial
Learn how you can use the Medical 3D Image Segmentation Notebook to predict brain tumors in MRI images.
NGC Catalog Resources
Technical Blogs
Learn how to use the NGC catalog with these step-by-step instructions.
Read technical blogs