DGL compatibility

DGL compatibility#

2025-07-22

6 min read time

Applies to Linux

Deep Graph Library (DGL) is an easy-to-use, high-performance and scalable Python package for deep learning on graphs. DGL is framework agnostic, meaning if a deep graph model is a component in an end-to-end application, the rest of the logic is implemented using PyTorch.

  • ROCm support for DGL is hosted in the ROCm/dgl repository.

  • Due to independent compatibility considerations, this location differs from the dmlc/dgl upstream repository.

  • Use the prebuilt Docker images with DGL, PyTorch, and ROCm preinstalled.

  • See the ROCm DGL installation guide to install and get started.

Supported devices#

  • Officially Supported: TF32 with AMD Instinct MI300X (through hipblaslt)

  • Partially Supported: TF32 with AMD Instinct MI250X

Use cases and recommendations#

DGL can be used for Graph Learning, and building popular graph models like GAT, GCN and GraphSage. Using these we can support a variety of use-cases such as:

  • Recommender systems

  • Network Optimization and Analysis

  • 1D (Temporal) and 2D (Image) Classification

  • Drug Discovery

Multiple use cases of DGL have been tested and verified. However, a recommended example follows a drug discovery pipeline using the SE3Transformer. Refer to the AMD ROCm blog, where you can search for DGL examples and best practices to optimize your training workflows on AMD GPUs.

Coverage includes:

  • Single-GPU training/inference

  • Multi-GPU training

Docker image compatibility#

AMD validates and publishes DGL images with ROCm and Pytorch backends on Docker Hub. The following Docker image tags and associated inventories were tested on ROCm 6.4.0. Click the to view the image on Docker Hub.

DGL Docker image components#

Docker

DGL

PyTorch

Ubuntu

Python

2.4.0

2.6.0

24.04

3.12.9

2.4.0

2.4.1

24.04

3.12.9

2.4.0

2.4.1

22.04

3.10.16

2.4.0

2.3.0

22.04

3.10.16

Key ROCm libraries for DGL#

DGL on ROCm depends on specific libraries that affect its features and performance. Using the DGL Docker container or building it with the provided docker file or a ROCm base image is recommended. If you prefer to build it yourself, ensure the following dependencies are installed:

ROCm library

Version

Purpose

Composable Kernel

1.1.0

Enables faster execution of core operations like matrix multiplication (GEMM), convolutions and transformations.

hipBLAS

2.4.0

Provides GPU-accelerated Basic Linear Algebra Subprograms (BLAS) for matrix and vector operations.

hipBLASLt

0.12.1

hipBLASLt is an extension of the hipBLAS library, providing additional features like epilogues fused into the matrix multiplication kernel or use of integer tensor cores.

hipCUB

3.4.0

Provides a C++ template library for parallel algorithms for reduction, scan, sort and select.

hipFFT

1.0.18

Provides GPU-accelerated Fast Fourier Transform (FFT) operations.

hipRAND

2.12.0

Provides fast random number generation for GPUs.

hipSOLVER

2.4.0

Provides GPU-accelerated solvers for linear systems, eigenvalues, and singular value decompositions (SVD).

hipSPARSE

3.2.0

Accelerates operations on sparse matrices, such as sparse matrix-vector or matrix-matrix products.

hipSPARSELt

0.2.3

Accelerates operations on sparse matrices, such as sparse matrix-vector or matrix-matrix products.

hipTensor

1.5.0

Optimizes for high-performance tensor operations, such as contractions.

MIOpen

3.4.0

Optimizes deep learning primitives such as convolutions, pooling, normalization, and activation functions.

MIGraphX

2.12.0

Adds graph-level optimizations, ONNX models and mixed precision support and enable Ahead-of-Time (AOT) Compilation.

MIVisionX

3.2.0

Optimizes acceleration for computer vision and AI workloads like preprocessing, augmentation, and inferencing.

rocAL

2.2.0

Accelerates the data pipeline by offloading intensive preprocessing and augmentation tasks. rocAL is part of MIVisionX.

RCCL

2.22.3

Optimizes for multi-GPU communication for operations like AllReduce and Broadcast.

rocDecode

0.10.0

Provides hardware-accelerated data decoding capabilities, particularly for image, video, and other dataset formats.

rocJPEG

0.8.0

Provides hardware-accelerated JPEG image decoding and encoding.

RPP

1.9.10

Speeds up data augmentation, transformation, and other preprocessing steps.

rocThrust

3.3.0

Provides a C++ template library for parallel algorithms like sorting, reduction, and scanning.

rocWMMA

1.7.0

Accelerates warp-level matrix-multiply and matrix-accumulate to speed up matrix multiplication (GEMM) and accumulation operations with mixed precision support.

Supported features#

Many functions and methods available in DGL Upstream are also supported in DGL ROCm. Instead of listing them all, support is grouped into the following categories to provide a general overview.

  • DGL Base

  • DGL Backend

  • DGL Data

  • DGL Dataloading

  • DGL DGLGraph

  • DGL Function

  • DGL Ops

  • DGL Sampling

  • DGL Transforms

  • DGL Utils

  • DGL Distributed

  • DGL Geometry

  • DGL Mpops

  • DGL NN

  • DGL Optim

  • DGL Sparse

Unsupported features#

  • Graphbolt

  • Partial TF32 Support (MI250x only)

  • Kineto/ ROCTracer integration

Unsupported functions#

  • more_nnz

  • format

  • multiprocess_sparse_adam_state_dict

  • record_stream_ndarray

  • half_spmm

  • segment_mm

  • gather_mm_idx_b

  • pgexplainer

  • sample_labors_prob

  • sample_labors_noprob