This work is done for the Digital Product School in Munich. In order to apply for the position of Artificial Intelligence Engineer, this repository is provided. The challenge is to use the Auto MPG dataset from Kaggle (https://www.kaggle.com/uciml/autompg-dataset) to predict the fuel efficiency of a vehicle using a basic regression with TensorFlow.
Create a new conda environment:
conda create -n tensor python=3.8
source activate tensor
Download and install the package:
git clone https://github.com/frommwonderland/dps.git
cd dps
pip install --upgrade pip
pip install -r requirements.txt
Running the little program by simply
python main.py
The main goal is to predict the 'mpg' parameter based on various others as fuel efficiency (MPG) is a function of many different other parameters. The first image shows some general information about the data.
The model is a simple multi-layer-perceptron. One example run with the loss and the result is shown in the following:
MPG prediction: 13.5