Skip to content

FH-Potsdam/ml-prototyping

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ml-prototyping

Applied introduction to machine learning

This repo holds a set of examples to get started with machine learning / artificial intelligence / algorithmic decision making projects. It is separated into three sections browser, node and python. This introduction is for beginners. The order of sections also represents the difficulty of getting started. For the browser version you basically need almost nothing to get started. For the other two sections you need to install libraries and (for efficient execution) make sure your libraries and your graphics card are able to talk to one another.

The projects will not introduce you to training a custom network, but instead use pre-trained models. Transfer-learning is used to modify an existing network for new content. Training your own custom network is ressource intensive and requires large data sets as well as a more indepth understanding of neural networks, therefore, this is not covered in this introductory session.

Content

The examples show how to do:

  • face and pose detection from images and video feeds
  • object detection in images
  • image classification (with an example for custom objects using transfer-learning-like method)
  • hand detection in images & videos
  • pre-trained style-transfer
  • sentiment analysis
  • KNN-cluster
  • transforming words/sentences to vectors for similarity analysis

More information on each section can be found inside the folders.

Recommendation

While everything can be done on a windows machine (I actually wrote and tested everything on a windows machine), I strongly recommend doing this on a unix machine (mac/linux), its a lot easier. If you are on a windows machine, see if the browser examples work for you and ignore the other two.

About

Applied introduction to machine learning

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published