When it comes to solving the world’s most profound computational challenges, scientists and researchers need the most powerful and accessible tools at their fingertips. With the ROCm™ open software platform built for GPU computing, HPC and ML developers can now gain access to an array of different open compute languages, compilers, libraries and tools that are both open and portable.

ROCm™ Learning Center offers resources to developers looking to tap the power of accelerated computing. No matter where they are in their journey, from those just getting started to experts in GPU programming, a broad range of technical resources below are designed to meet developers where they are at.

Getting Started


This module offers an introduction and sets you up for success with using ROCm™. The tutorials provides an overview of what ROCm is and how to install the software to get started.

Introduction to ROCm™ Watch Video Download Presentation
ROCm™ Installation Watch Video Download Presentation

Fundamentals of HIP Programming


This module provides in-depth training on programming with HIP. HIP is a high performance, CUDA-like programming model that is built on an open and portable framework. You will learn everything ranging from the basics of GPU programming to profiling GPU applications to an in-depth knowledge of programming with HIP.

Introduction to HIP Watch Video Download Presentation
Deep Dive into GPU and Performance Optimizations Watch Video Download Presentation
Your First HIP Code: Vector Add Watch Video Download Presentation
   Lab: Vector Add Watch Video Download Lab
   Lab: Vector Add using Printf Watch Video Download Lab
HIP using ROCm Profiler: Matrix Transpose Watch Video Download Presentation
   Lab: Matrix Transpose with ROCm Profiler Watch Video Download Lab
Matrix Transpose Part 2: Naïve Version Watch Video Download Presentation
   Lab: Matrix Transpose Naive Watch Video Download Lab
Matrix Transpose Part 3: Optimized LDS Version Watch Video Download Presentation
   Lab: Matrix Transpose Optimized LDS Watch Video Download Lab
Matrix Transpose Part 4: Key Takeaways Watch Video Download Presentation
Debugging Tips and Tricks Watch Video Download Presentation
   Lab: Debugging Tips and Tricks Watch Video Download Lab
Debugging Tips and Tricks: Wrap-up Watch Video Download Presentation

From CUDA to HIP


In this module, you will learn best practices in porting your CUDA code to HIP.

Getting Started from CUDA to HIP Watch Video Download Presentation
   Lab: Getting Started with Vector Add Watch Video Download Lab
Porting Deep Learning CUDA-CNN to HIP Watch Video Download Presentation
   Lab: CUDA-CNN to HIP Watch Video Download Lab
Porting Machine Learning K-means to HIP Watch Video Download Presentation
   Lab: K-means to HIP Watch Video Download Lab
Wrap-up: Porting from CUDA to HIP Watch Video Download Presentation

Deep Learning on ROCm


This module includes hands on training for Deep Learning and equips you with the necessary knowledge on optimal usage of ROCm™ based systems.

Introduction to Deep Learning on ROCm Watch Video Download Presentation
Running TensorFlow for MNIST Example Watch Video Download Presentation
   Lab: TensorFlow & MNIST Watch Video Download Lab
Running PyTorch & LSTM Example Watch Video Download Presentation
   Lab: PyTorch & LSTM Watch Video Download Lab
Multi-GPU Deep Learning Watch Video Download Presentation
   Lab: Multi-GPU Deep Learning Watch Video Download Lab
Wrap-up: Deep Learning on ROCm Watch Video Download Presentation

Multi GPU Programing


This last module deepens your knowledge of ROCm™ and scaling across platforms through hands-on labs, examples, and trainings.

Introduction to Multi-GPU Programming on ROCm Watch Video Download Presentation
Using RCCL Communication Collectives Library Watch Video Download Presentation
   Lab: RCCL Library Watch Video Download Lab
Multi-GPU with MPI Watch Video Download Presentation
   Lab: Multi-GPU with MPI Watch Video Download Lab
Multi GPU Summary Watch Video Download Presentation