Spyridon Dimitriadis 's repositories
pca_kmeans_matlab
An assignment in which dimension reduction is applied with PCA, clustering with k-means algorithm and classification with Naive Bayes Classifier. Implemented in MATLAB, using the MNIST dataset.
CNN_interpretability
This project is about detecting pneumonia or no pneumonia from x-ray images and trying to explain the predictions of the Network.
MNist_classification
Classification of the famous dataset of Mnist, applying the Support Vector Machine algorithm and a Radial Basis Function Network. These assignments were for the course Neural Networks from Aristotle University of Thessaloníki. February 2017.
DNN_with_keras
Finding the right hyperparameteres for a Deep Neural Network with keras. Classification task with 7 classes. Use cartographic variables to classify forest categories.
Pytorch-mini-Projects
Apply different Neural Network architectures in Pytorch.
NLP-Multilabel-Classification
The scope of this project is to apply a number of tools on solving a machine learning problem. The NLP Multilabel classification problem is chosen.
Utilizing-NLP-transfer-learning-with-ULMFiT-in-Chemoinformatics
This project was carried out as a part of the course Text Mining (732A92) which is part of the Statistics and Machine Learning Master programme at Linköping's University in 2020/2021. The course web page can be found at https://www.ida.liu.se/~732A92/lectures.en.
adv_R_course_lab3
lab 3
apply_sklearn
Apply some of the main machine learning algorithms with sklearn library in different data sets, also did some visualization to explore the data. All the exercises are from Pierian Data.
BookStore_mini_project
A GUI with which you can stores and manipulate book titles in a database. Libraries used: Tkinter and SQLite.
fastbook
The fastai book, published as Jupyter Notebooks
Information_Theory_s_relation_to_Machine_Learning
A theoretical approach of the relation between Information Theory and Machine Learning.
text_classification_on_IMDb_reviews
A sentiment analysis task of classifying the polarity (positive or negative) of IMDb reviews. Using 'traditional' machine learning.