Optimization

PyTorch Memory Tuning

July 20, 2023
Pytorch, Optimization

This article will focus on minimizing GPU memory footprint — for both optimization and inference workloads. Throughput and latency usually get all the attention, but reducing memory consumption without making architecture sacrifices is often just as valuable.

PyTorch Performance Features and How They Interact

April 14, 2023
Machine Learning Productionization
Pytorch, Optimization

PyTorch in 2023 is a complex beast, with many great performance features hidden away. Simple top-N lists are weak content, so I’ve empirically tested the most important PyTorch tuning techniques and settings in all combinations. I’ve benchmarked inference across a handful of different model architectures and sizes, different versions of PyTorch and even different Docker containers.

Solving Machine Learning Performance Anti-Patterns: a Systematic Approach

June 24, 2021
Machine Learning Productionization
Nsight Systems, NVTX, Optimization, TensorRT, TorchScript, Quantization

This article is a high-level introduction to an efficient worfklow for optimizing runtime performance of machine learning systems running on the GPU. Using traces from Nsight Systems to show real production scenarios, I introduce a set of common utilization patterns and outline effective approaches to improve performance.

Object Detection at 2530 FPS with TensorRT and 8-Bit Quantization

December 31, 2020
Visual Analytics, Machine Learning Productionization
SSD300, Pytorch, Object Detection, Optimization, TensorRT, Quantization, ONNX, Nsight Systems

This article is a deep dive into the techniques needed to get SSD300 object detection throughput to 2530 FPS. We will rewrite Pytorch model code, perform ONNX graph surgery, optimize a TensorRT plugin and finally we’ll quantize the model to an 8-bit representation. We will also examine divergence from the accuracy of the full-precision model.

Object Detection at 1840 FPS with TorchScript, TensorRT and DeepStream

October 17, 2020
Visual Analytics, Machine Learning Productionization
SSD300, Pytorch, Object Detection, Optimization, DeepStream, TorchScript, TensorRT, ONNX, NVTX, Nsight Systems

In this article we take performance of the SSD300 model even further, leaving Python behind and moving towards true production deployment technologies: TorchScript, TensorRT and DeepStream. We also identify and understand several limitations in Nvidia’s DeepStream framework, and then remove them by modifying how the nvinfer element works.

Object Detection from 9 FPS to 650 FPS in 6 Steps

September 30, 2020
Visual Analytics, Machine Learning Productionization
SSD300, Pytorch, Object Detection, Gstreamer, NVTX, Optimization, Nsight Systems

Making code run fast on GPUs requires a very different approach to making code run fast on CPUs because the hardware architecture is fundamentally different. Machine learning engineers of all kinds should care about squeezing performance from their models and hardware — not just for production purposes, but also for research and training. In research as in development, a fast iteration loop leads to faster improvement. This article is a practical deep dive into making a specific deep learning model (Nvidia’s SSD300) run fast on a powerful GPU server, but the general principles apply to all GPU programming.


© Paul Bridger 2020