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