This project aims to create a model which predicts the relative performance of a CPU based on some characteristics. The input-output mapping is nonlinear, thus a Multi-Layer Perceptron (MLP) is created using TensorFlow and Keras ML frameworks. The input layer takes a 6-feature input vector. We designed a 2-hidden layer architecture of 64 neurons each. Regarding the output layer, we selected a linear activation function due to the nature of the problem (unbounded output - regression problem).
Libraries/Frameworks used
- Pandas
- NumPy
- Seaborn
- Matplotlib
- scikit-learn
- TensorFlow
- Keras
