Nautilus JupyterHub - Image & Resource Guide
Choose the right environment for your project
When launching a JupyterHub server on Nautilus, you must choose an Image. This image determines your computing environment, including pre-installed libraries and tools. Below is a guide to help you pick the right one based on your needs.
General Science & Programming
| Image Name | Use Case | Includes |
|---|---|---|
| Jupyter Stack Minimal Notebook | Lightweight or custom environments | Minimal Python and Jupyter setup |
| Jupyter Stack R Notebook | R programming in Jupyter | R language and IRKernel |
| Jupyter Stack Scipy Notebook | Scientific computing | NumPy, SciPy, Matplotlib |
| Jupyter Stack Tensorflow Notebook | TensorFlow-based ML | TensorFlow, Keras |
| Jupyter Stack PyTorch Notebook | PyTorch-based ML | PyTorch, TorchVision |
| Jupyter Stack Julia Notebook | Julia development | Julia language in Jupyter |
| Jupyter Stack Datascience Notebook | General data science | Pandas, Scikit-learn, Seaborn, Matplotlib |
Domain-Specific & Specialized Images
| Image Name | Use Case | Includes |
|---|---|---|
| NRP R Studio Notebook | R users preferring RStudio | RStudio IDE, R packages |
| Eclipse C/C++ Notebook | C/C++ development | Eclipse IDE, compilers |
| Kube Notebook | Kubernetes workflows | kubectl, K8s tools |
| SageMath Notebook | Symbolic math, algebra | SageMath, math libraries |
| GIS Notebook | Geospatial analysis | GeoPandas, GDAL, QGIS |
| Health Informatics Notebook | Biomedical/health data | Bioinformatics libraries |
| Astronomy Notebook | Astronomy & astrophysics | Astropy, astronomy tools |
| LLM Notebook | Language models & NLP | Hugging Face, NLP tools |
| Molecular Dynamics Notebook | Molecular simulation | GROMACS, NAMD, VMD |
Deep Learning & GPU-Accelerated Images
| Image Name | Use Case | Includes |
|---|---|---|
| NRP Deep Learning & Data Science Full, PyTorch (CUDA) | GPU-based deep learning (PyTorch) | PyTorch, CUDA, full data science stack |
| NRP Deep Learning & Data Science Full, TensorFlow (CUDA) | GPU-based deep learning (TensorFlow) | TensorFlow, CUDA, full data science stack |
Resource Selection Reminders
GPUs: Use only if required (e.g., training deep learning models)
Cores & RAM: Start small (e.g., 2 cores, 4-8 GB RAM) and scale as needed
GPU Type: Leave as Any unless you need specific hardware