This month the NGC catalog added new containers, model resumes, container security scan reports, and more to help identify and deploy AI software faster.
The NVIDIA NGC catalog is a hub for GPU-optimized deep learning, machine learning, and HPC applications. With highly performant software containers, pretrained models, industry-specific SDKs, and Jupyter Notebooks the content helps simplify and accelerate end-to-end workflows.
New features, software, and updates to help you streamline your workflow and build your solutions faster on NGC include:
Model resumes
The NGC catalog offers state-of-the-art pretrained models that help you build your custom models faster with just a fraction of the training data.
Now, every model comes with a resume that provides information on model architecture, training parameters, training datasets, performance, and limitations to help you make informed decisions before downloading the model. They also include instructions on how to use the model so you can focus on AI development.
View the demo video and explore models for applications like speech and computer vision in various industries including Retail, Healthcare, Smart Cities, and Manufacturing.
Container security scan reports
All the container images in the NGC catalog are scanned for CVEs, malware, crypto keys, open ports, and more.
Now, the containers come with a security scan report, which provides a security rating of that image, breakdown of CVE severity by package, and links to detailed information on CVEs.
The scan reports are available on the latest as well as the previous versions of the images and with the entire NGC catalog scanned every 30 days. If you’re using an older version with high or critical severity, the scan report will flag the vulnerabilities and suggest remedies.
View the demo video for more details and explore application containers for deep learning, machine learning, and HPC.
TAO Toolkit
The latest version of the TAO Toolkit is now available for download. The TAO Toolkit, a CLI, and Jupyter notebook-based version of TAO, brings together several new capabilities to help you speed up your model creation process.
Key highlights include:
- New computer vision model for 2D and 3D action recognition.
- New text-to-speech models: HifiGAN and FastPitch.
- New computer vision model architectures: EfficientDet and YoloV4 Tiny.
- Increased utilization of GPUs for faster, more efficient training.
Deep learning software
The most popular deep learning frameworks for training and inference are updated monthly. Pull the latest version (v22.01) of:
M-Star CFD
M-Star CFD is a multiphysics modeling package used to simulate fluid flow, heat transfer, species transport, chemical reactions, particle transport, and rigid-body dynamics.
M-Star CFD contains M-Star Build (to prepare models and specify simulation parameters), M-Star Solve (to run simulations), and M-Star Post (to render and plot data.)
HPC applications
Latest versions of the popular HPC applications are also available in the NGC catalog.
Visit the NGC catalog to see how the GPU-optimized software can help simplify workflows and speedup solution times.