Skip to content

mikolajsacha/meta_learning_experiments

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

22 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Experiments with Meta-Learning

Searching for beautiful minima

This project uses python 3.5 with Keras/tensorflow for experiments.
Code works only for Keras with tensorflow backend.

To run experiments:

  1. create conda environment or install required packages:
    conda env create -f config/environment.yml
    or
    pip install -r requirements.txt
    Note: if GPU is not used, change 'tensorflow-gpu' to 'tensorflow'
  2. set hyperparameters. As of now they need to be set by manually editing file train.py (this will be refactored)
  3. run: python train.py

For visualizations, see: notebooks folder

About

Experiments with Meta-Learning in search for beautiful minima

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published