[ANN] EasyML.jl - a foolproof way of doing ML with GUI elements

We would like to announce a package which goal is to simplify machine learning as much as possible so that anyone can start doing it right away.

With our package it is possible to:

  • Design a neural network
  • Train a neural network
  • Validate a neural network
  • Apply a neural network to new data

Classification, regression and segmentation on images are currently supported.

Github link is here: EasyML.jl

40 Likes

It is a nice package!

I have installed, and it is running nicely. This type of software is specially interesting for me, because I teach ML, and this type of GUI packages are could be very useful. If you want, I could help you with the documentation, that it still needs more work.

6 Likes

Hi! I am the main developer of this package.

My background is actually in microbiology and, therefore, help of someone knowledgeable in ML would be awesome. You could DM me with your suggestions so I can implement them.

7 Likes

Found and fixed some elusive bugs. Those who installed the package consider reinstalling it. It will be a good idea in general to try reinstalling the package if you encounter any bugs until the package is added to the registry in 3 days, since the bugs may have already been fixed on main.

2 Likes

The package has been added to the registry and can be installed using add EasyML.

1 Like

Hi,

although I have not tested the package yet, I am interested. On JuliaCon, I will have a talk about Mill and JsonGrinder packages, which are great for creating classifiers for data stored in JSONs (assuming the json has a meaningful structure). Do you think it would be difficult to make those packages to speak to each other? The result might be very interesting.

1 Like

It seems that it should not be hard to implement support for Mill.jl since its output is just a model. I left a note in a plan for the package.

1 Like