Table of Contents

Introduction

TensorBoard is a web-based visualisation tool that is used to visualise and analyse various aspects of machine learning models and their training process. It allows users to track and visualise metrics such as loss, accuracy and visualisations of the structure of the model and the model's weights and activations, helping users understand the internal representations learned by the model. PyTorch 1.8 introduces an enhanced profiler API that can record both CPU-side operations and CUDA kernel launches on the GPU side. This profiler enables performance analysis and provides insights into performance bottlenecks. The recorded information can be visualised using the TensorBoard Plugin, allowing for detailed analysis and optimisation of PyTorch models.

...