From 1fd83545b4ea3e7c121fa4196f08897e36efa807 Mon Sep 17 00:00:00 2001 From: jatkinson1000 <109271713+jatkinson1000@users.noreply.github.com> Date: Sat, 28 Oct 2023 14:21:23 +0100 Subject: [PATCH] Tidy statement of need. --- JOSE_paper/paper.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/JOSE_paper/paper.md b/JOSE_paper/paper.md index 52c5f25..a7d6651 100644 --- a/JOSE_paper/paper.md +++ b/JOSE_paper/paper.md @@ -55,14 +55,13 @@ In contrast, much theoretical ML material addresses high-level concepts without discussing coding considerations or details of how to actually use popular frameworks to implement the models. - -Practical aspects can only really be learnt through trial-and-error and -practical experience. Many know how ML works in an abstract sense, but will be unfamiliar with lower-level practicalities such as image transforms and other preprocessing techniques required to present data to neural networks. They can describe how something works, but would have no idea where to start if asked to do it. +Such practical aspects are ideally learnt through trial-and-error and +hands-on experience. Many machine learning frameworks are accessed using a Python framework. One such commonly used framework is [PyTorch](https://pytorch.org/) [@paszke2019pytorch].