Skip to content

thomasklimek/Projects_Portfolio

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Projects Portfolio

This folder contains a list of projects demonstrating various coding, data science, ML, and visualization skills.

2-Layer Neural Network and Linear Regression In TensorFlow

Implementation of a 2-layer fully connected neural network using Python, TensorFlow and Jupyter Notebook. Neural network is used to classify CIFAR-10 data.

A-Star Search Algorithm

Implementation of a A-Star search in C++ to solve pancake flipping sample problem.

C++ Neural Network

Implementing neural network from scratch using C++ and standard libraries.

Climate Change and Emissions Visualization

Data visualization of emissions levels in different states using EPA dataset javascript, and D3.js.

Dropout, Batch Norm, and Conv Net

Practice implementations of deep learning techniques such as dropout, batch normalization, and a convolutional neural network. Implemented using Python, TensorFlow, and Jupyter Notebook.

Gambler's Problem Reinforcement Learning

Solving the Gambler's Problem using Reinforcement Learning algorithm of value itteration. Implemented using Python.

Generative Adversarial Network for EEG Data

Implementation of a GAN for generating realistic synthetic EEG data. This was a final project for a deep learning class in which we experimented with creating a GAN to augment small EEG datasets in attempt to increase classification accuracy. Implemented using Python, and PyTorch.

Genetic Search Algorithm

Implementation of a genetic search algorithm in C++ to solve the backpack problem.

Harmony Generation Using Reinforcement Learning

This project uses reinforcement learning algorithms such as q-learning and sarsa to tackle the problem of real time musical accompaniment. Implemented as a final project for my Reinforcement Learning course, we frame the act of playing music as a reinforcement learning problem and use Python to implement a system for creating chords based on feedback from existing songs.

K-Arm Bandit Reinforcement Learning

Implementation of basic reinforcement learning algorithm to learn an optimal policy for the K-Arm Bandit Problem.

Linear Regression and 2-Layer Neural Net

Practice implementation of basic Linear Regression and a 2-layer fully connected neural network in TensorFlow and classification of CIFAR-10 data.

Logistic Regression for classifying Titanic and Mnist

From scratch logistic regression implemented using Python, Numpy, and autograd for calculating gradients.

Movie Recommendations, SVD, and Vector Analysis

Practice implementation of a recommendation system using collaborative filtering and singular value decomposition. Implemented in Python and Surpise Library.