Skip to content

Tutorial that is used for the HDS-LEE course on hyperparameter optimization.

Notifications You must be signed in to change notification settings

DLR-SC/Hyperparameter_tutorial

Repository files navigation

Hyperparameter_tutorial

Tutorial that is used for the HDS-LEE course on hpyerparameter optimization.

The tutorial consists of two parts:

  • Tutorial_Part1 is used for an introduction to the considered regression problem. Furthermore, it gives a very short introduction to Keras.
  • Tutorial_Part2 uses Talos for machine-assisted hyperparameter optimization. There is a further notebook '*_solution' that contains the solution to the exercises in this notebook.

Installation

Before using the notebooks, you need to install talos, keras and all other dependencies. First, create a virtual environment (or conda environment) and activate it:

On macOS and Linux:

python3 -m venv env

source env/bin/activate

On Windows:

py -m venv env

.\env\Scripts\activate

Install all requirements

pip install -r requirements.txt

Running the notebooks

jupyter notebook

Fallback solution (nothing works)

Alternatively, if nothing works, you can also run your Jupyter notebooks in a live environment using Binder. Just click on the following link

Binder

About

Tutorial that is used for the HDS-LEE course on hyperparameter optimization.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •