8000 Home · jhagelback/ml-workshop Wiki · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content
Johan Hagelbäck edited this page Apr 3, 2020 · 26 revisions

What to do?

  • Select programming tasks that you are interested in from the list below to work on. You can work alone or in small groups.
  • For each task there is a pre-recorded lecture that it is recommended to watch, plus some optional reading if you want to learn even more.
  • The size of different tasks can vary quite a lot. The ... from scratch is typically larger.
  • When you are done with a task, notify @johan.hagelback on Slack to book a short online meeting where you show what you have done.
  • When you are done with the programming tasks, the next step is to define and work on a project. In the project it is recommended that you work on some data from the organisation (it is okay to use other datasets as well). Discuss your project idea with @johan.hagelback on Slack before you start. You can work alone or in small groups.

Development Environment

For machine learning projects I usually recommend using Jupyter Notebooks, see the Jupyter page for more information and example notebooks.

Programming Tasks

 A01 - Recommendation System from scratch
 A02 - Clustering from scratch
 A03 - Search Engine from scratch
 A04 - Naïve Bayes algorithm from scratch
 A05 - Text classification on Wikipedia articles
 A06 - Basic classification on different datasets
 A07 - Regression on GPU benchmark data
 A08 - Classifying hand-written digits
 A09 - Classifying Zalando articles
 A10 - Build a face recognition system
 A11 - Build an object recognition system

Datasets

 Datasets page
Clone this wiki locally
0