The goal of this project is to understand how K-nearest neigbhors works.
The experiment is implemented using PyTorch.
The post of this experiment can be found following this link.
The K-nearest neighbors is trained on physicochemical data to predict the quality of a red or white wines.
This dataset contains 6497 samples and the following features:
- fixed acidity
- volatile acidity
- citric acid
- residual sugar
- chlorides
- free sulfur dioxide
- total sulfur dioxide
- density
- pH
- sulphates
- alcohol
- quality
In order to install the conda environment needed to run the notebook, run the following line:
conda env create --file requirements.yml
conda activate torch