magdalena318 Goto Github PK

cognitivemap

Cognitive map for predicting univariate time-series using ADAM optimization. Coded from scratch without any help of machine learning libraries.

cpg-islands-and-dna-methylation-analysis

dos-script

Python script for Denial-of-Service (DoS) attack on Russian and Belorussian propaganda and key webpages.

file-storage-server

Python Tornado server with functionalities of uploading and downloading files

flightradarwebsite

image-quality-enhancement

Project allows for few ways of improving images quality or other characteristics as needed. Contains 2 types of neural networks: U-net and GAN coded with help of FastAI library. Alternatively one may choose from various OpenCV filters.

imageeditor

Application for editing existing images (parameters changing, applying filters, quantization) and drawing new ones.

irisrecognitiondaugman

Implementation of Daugman's iris recognition algorithm.

nn-from-scratch

Neural network with 3 layers using relu, sigmoid and softmax activation functions

package-delivery-website

A website for package delivery. Front-end is implemented in Javascript and uses Leaflet maps for forward/reverse geocoding. The page allows for adding/looking up packages and vehicles transporting the parcels. Back-end is a .NET web application which assigns new packages to the closest available vehicles and computes transport routes using Dijkstra's algorithm.

qualitative-activity-recognition

This project has been created for Practical machine learning course project from 'Data Science Specialization' on Coursera provided by 'Johns Hopkins University'. The goal is to predict the quality of performed activies based on the data from accelerometers on the belt, forearm, arm, and dumbell of 6 participants. The utilized dataset is the Weight Lifting Exercises Dataset presented in the following paper:

random-forest-classifier-for-finding-enhancers

regression-models-course-project

This analysis is performed for the Regression Models course by John Hopkins University on Coursera. The data used for analysis is the built-in 'mtcars' R dataset.