Skip to content

Storing of the midterms delivered throughout the semester for the Intelligent Systems for Pattern Recognition [ISPR] course, a.y. 2022/23

Notifications You must be signed in to change notification settings

francescapoli98/pattern-recognition-systems

Repository files navigation

Intelligent Systems for Pattern Recognition [ISPR]

The ISPR couse is an introduction to the analysis and design of advanced machine learning and deep learning models for modern pattern recognition problems and discusses how to realize advanced applications exploiting computational intelligence techniques.

This repository stores the notebooks delivered for the midterms during the semester.

Midterm 1

  • Assignment 3: basic signal processing in audio files

Plotting the spectogram of samples of different instruments from the english Philarmonia dataset in order to check if it is possible to recognize the different instruments by only looking at them. Developed using the librosa library.

Midterm 2

  • Assignment 1: Hidden Markov Models and Viterbi algorithm for signal processing

Fitting and experimenting with an Hidden Markov Model with Gaussian emissions to the data in the dataset, and confront with the resulting time series of performing Viterbi on a reasonably sized subsequence.

Midterm 3

  • Assignment 5: Fake News Classification

Training a a binary classifier to recognize fake news from a Kaggle dataset.

Midterm 4

Reading and summarizing the main findings of a chosen paper through a short presentation.

  • Paper n°3: Bo Chang, Minmin Chen, Eldad Haber, Ed H. Chi, AntisymmetricRNN: A Dynamical System View on Recurrent Neural Networks, ICLR 2019 arXiv:1902.09689