Important

You are viewing the NeMo 2.0 documentation. This release introduces significant changes to the API and a new library, NeMo Run. We are currently porting all features from NeMo 1.0 to 2.0. For documentation on previous versions or features not yet available in 2.0, please refer to the NeMo 24.07 documentation.

Get Started#

NeMo Curator provides many tools for curating large scale text-image pair datasets for training generative image models.

Install NeMo Curator#

To install the image curation modules of NeMo Curator, ensure you meet the following requirements:

  • Python 3.10

  • Ubuntu 22.04/20.04

  • NVIDIA GPU * Volta™ or higher (compute capability 7.0+) * CUDA 12 (or above)

Note: While some of the text-based NeMo Curator modules do not require a GPU, all image curation modules require a GPU.

You can get NeMo Curator in 3 ways.

  1. PyPi

  2. Source

  3. NeMo Framework Container

PyPi#

NeMo Curator’s PyPi page can be found here.

pip install cython
pip install nemo-curator[image]

Source#

NeMo Curator’s GitHub can be found here.

git clone https://github.com/NVIDIA/NeMo-Curator.git
pip install cython
pip install ./NeMo-Curator[image]

NeMo Framework Container#

NeMo Curator comes preinstalled in the NeMo Framework container. You can find a list of all the NeMo Framework container tags here.

Use NeMo Curator#

NeMo Curator can be run locally, or on a variety of compute platforms (Slurm, k8s, and more).

To get started using the image modules in NeMo Curator, we recommend you check out the following resources:

On this page