One of the biggest and most powerful hardware companies in the world is NVIDIA. The company offers free courses to assist you learn more about generative AI, GPUs, robots, chips, and other topics in addition to its highly sought-after GPUs.
Above all, these can all be done in less than a day and are completely free of charge. Let’s examine them now.
- Illustrated Generative AI
The principles of generative AI are introduced in this free, self-paced online course. Generative AI is the process of producing new information depending on various inputs. Participants will gain an understanding of the ideas, uses, difficulties, and future possibilities of generative AI through this course. Learning objectives cover generative AI’s definition and operation, a summary of its various applications, and a discussion of the opportunities and challenges that come with it. To take part, all you need to know is the fundamentals of deep learning and machine learning.
- Digital Fingerprinting with Morpheus
In just one hour, participants will learn how to create and implement the NVIDIA digital fingerprinting AI workflow, which offers total data visibility and drastically cuts down on the amount of time needed to detect threats. The NVIDIA Morpheus AI Framework, which is intended to speed up GPU-based AI applications for filtering, processing, and categorizing massive amounts of streaming cybersecurity data, will be made practical for participants to use.
They will also be introduced to the NVIDIA Triton Inference Server, an open-source program that enables the uniform deployment and operation of AI models across a range of workloads. Although knowledge of the Linux command line and defensive cybersecurity concepts is helpful, there are no prerequisites for this lesson.
- Building A Brain in 10 Minutes
This course explores the fundamentals of neural networks using concepts from psychology and biology. Understanding how neural networks use input for learning and the mathematical concepts that underpin a neuron’s operation are its two main goals.
Although anyone can run the provided code and watch its activities, it is recommended to have a firm understanding of basic Python 3 programming principles, such as functions, loops, dictionaries, and arrays. It’s also advised to be knowledgeable about computing regression lines.
- Building RAG Agents with LLMs
The ability of retrieval-based LLM systems to analyze documents, plan strategically, use tools, and hold educated conversations has made them popular. The deployment of agent systems and scaling them to satisfy user and client demands are the main topics of this course.
The main learning goals include investigating scalable deployment techniques for vector databases and LLMs, comprehending microservices and their interactions, and experimenting with modern LangChain paradigms for document retrieval and dialogue management. Moreover, you can learn about productionalization and framework exploration while gaining hands-on experience with cutting-edge models.
This is great if you know your way around LLMs and related composition frameworks like LangChain and you know Python at an intermediate level.
- Augment your LLM Using RAG
Facebook AI Research developed Retrieval Augmented Generation (RAG) in 2020 as a way to improve an LLM output without requiring model retraining by integrating real-time, domain-specific data. RAG creates an end-to-end architecture by integrating a response generator and an information retrieval module. This introduction attempts to give a basic knowledge of RAG, including its retrieval process and the key elements inside NVIDIA’s AI Foundations framework, by drawing on NVIDIA’s internal practices. Once you have a firm grasp of these principles, you can begin researching LLM and RAG applications.
- Developing Video AI Use Cases in the Periphery with Jetson Nano
The goal of this self-paced online course is to give students the tools they need to understand videos using AI utilizing the NVIDIA Jetson Nano Developer Kit. Using the NVIDIA DeepStream SDK, participants will investigate intelligent video analytics (IVA) applications through hands-on exercises and Python application samples in JupyterLab notebooks. The Jetson Nano setup, building end-to-end DeepStream pipelines for video analysis, including different input and output sources, configuring multiple video streams, and using alternative inference engines such as YOLO are all covered in this course.
Basic knowledge of the Linux command line and comprehension of Python 3 programming fundamentals are prerequisites. The course requires specific hardware, such as the Jetson Nano Developer Kit, and makes use of technologies like TensorRT and DeepStream. Multiple-choice questions are used for assessment, and when finished, a certificate is given out. You will need a compatible power source, microSD card, USB data cable, USB webcam, and an NVIDIA Jetson Nano Developer Kit or the 2GB version for this course.
- How to Build Custom 3D Scene Manipulator Tools on NVIDIA Omniverse
This course provides useful advice on how to use the flexible Omniverse platform to extend and improve 3D tools. Participants will learn how to construct advanced tools for generating physically correct virtual environments from the Omniverse developer ecosystem team.
Learners will delve into Python code to create new scene manipulator tools within Omniverse through self-paced assignments. Launching Omniverse Code, installing and enabling extensions, traversing the USD stage hierarchy, and building scale-controlling widget manipulators are among the main learning goals.
In addition, the training covers creating specialized scale manipulators and repairing malfunctioning manipulators. The Python Extension, Visual Studio Code, and Omniverse Code are necessary tools. The bare minimum of hardware consists of a desktop or laptop computer with an AMD Ryzen or Intel i7 Gen 5 processor and an NVIDIA RTX Enabled GPU with 16GB RAM.
- Assemble a Simple Robot in Isaac Sim
This course provides a hands-on lesson for putting together a basic mobile robot with two wheels using the Isaac Sim GPU platform’s “Assemble a Simple Robot” guide. Throughout the course of the 30-minute lesson, important topics are covered, including setting up joint drives and properties for the robot’s mobility, importing a USD dummy robot into the simulation environment, and connecting a local streaming client to an Omniverse Isaac Sim server. Participants will also learn how to give the robot more articulations.
Participants will be more comfortable with the Isaac Sim interface and the documentation needed to start their own robot simulation projects by the end of the course.
A Windows or Linux computer with sufficient internet connectivity for client/server streaming and the ability to install Omniverse Launcher and other apps are requirements for this course. This 30-minute, free lesson on Omniverse technology will cover several topics.
- Monitoring Disaster Risk with Satellite Images
The course, which was developed in partnership with the United Nations Satellite Centre, teaches participants how to develop and apply deep learning models for automated flood detection with a focus on disaster risk monitoring using satellite data. The acquired competencies are intended to lower expenses, boost productivity, and raise the efficacy of disaster relief initiatives. The course will cover how to implement a machine learning workflow, use hardware-accelerated tools to handle massive amounts of satellite imagery data, and use transfer-learning to create deep learning models at a reasonable cost.
The course also addresses the use of deep learning-based inference for flood event identification and response, as well as the deployment of models for analysis in close to real-time. Proficiency in Python 3, a fundamental comprehension of machine learning and deep learning principles, and an enthusiasm for manipulating satellite photos are prerequisites.
- Introduction to AI in the Data Center
This course will cover the architecture and background of GPUs as well as AI application cases, machine learning, and deep learning procedures. The course also addresses deployment issues for AI workloads in data centers, including multi-system clusters and infrastructure design, in an approachable style for beginners.
Professionals in data centers, DevOps, system and network administrators, and IT are the target audience for this course.
