SteliosKliafas
Artificial Intelligence, and Software Engineering
Athens Technology Center (ATC)Athens (Greece) - London (UK)
Pinned Repositories
coursera-gan-specialization
Programming assignments and quizzes from all courses within the GANs specialization offered by deeplearning.ai
Deeplearning.ai-GAN-Specialization-Generative-Adversarial-Networks
This repository contains my full work and notes on Deeplearning.ai GAN Specialization (Generative Adversarial Networks)
django-react-redux-task-app
A simple CRUD app to get comfortable with Django-React-Redux
Finetuning-Transformer-Models-Movie-Plot-Generation-Classification
Used the Netflix Movie & TV Shows dataset to finetune the GPT-2 model for Text Generation, while also finetuning the BERT model for classification to evaluate and compare the model’s accuracy on the GPT-2 generated text and the original
Generating-Deep-Fake-Images-with-DCGAN
In this project we propose the challenge to generate realistic people faces. This is one of the most controversial applications of GANs, as they recently achieved incredibly good results which raised many concerns mostly related to ethical and privacy aspects (thispersondoesnotexists). To achieve this goal, we propose the implementation of the DCGAN model. We then propose a comparison between the results obtained by trying different configurations of the model highlighting the strengths and weaknesses of each one. The dataset that we used is the ‘Labelled Faces in the Wild (LFW)’ dataset, which contains more than 13,000 images of faces collected from the web. Although this dataset is typically used for tasks related to image recognition, we found that the images suited well for the scope of our project. Furthermore, most implementations revolving around face generation use the ‘CelebA’ dataset and achieve decent results, but there are not many examples on different datasets, thus, we decided to experiment and test the performance of the DCGAN model on the LFW dataset.
HopfieldDRQN
Embedding Modern Continuous Hopfield Networks (Ramsauer et al., 2020) in the DRQN algorithm
INM705-Project-Capecchi---Kliafas
numpy_python_ann
The point of this task is to develop a multi-layer neural network for classification using Python and Numpy: Implement sigmoid and relu layers (with forward and backward pass) Implement a softmax output layer Implement a fully parameterizable neural network (number and types of layers, number of units) Implement an optimizer (e.g. SGD or Adam) and a stopping criterion of your choosing Train your Neural Network using backpropagation
shortest_path_algorithms
This repository contains implementations of algorithms that find the shortest path from a point a to point b located at a maze
SteliosKliafas's Repositories
SteliosKliafas/HopfieldDRQN
Embedding Modern Continuous Hopfield Networks (Ramsauer et al., 2020) in the DRQN algorithm
SteliosKliafas/coursera-gan-specialization
Programming assignments and quizzes from all courses within the GANs specialization offered by deeplearning.ai
SteliosKliafas/Deeplearning.ai-GAN-Specialization-Generative-Adversarial-Networks
This repository contains my full work and notes on Deeplearning.ai GAN Specialization (Generative Adversarial Networks)
SteliosKliafas/django-react-redux-task-app
A simple CRUD app to get comfortable with Django-React-Redux
SteliosKliafas/Finetuning-Transformer-Models-Movie-Plot-Generation-Classification
Used the Netflix Movie & TV Shows dataset to finetune the GPT-2 model for Text Generation, while also finetuning the BERT model for classification to evaluate and compare the model’s accuracy on the GPT-2 generated text and the original
SteliosKliafas/Generating-Deep-Fake-Images-with-DCGAN
In this project we propose the challenge to generate realistic people faces. This is one of the most controversial applications of GANs, as they recently achieved incredibly good results which raised many concerns mostly related to ethical and privacy aspects (thispersondoesnotexists). To achieve this goal, we propose the implementation of the DCGAN model. We then propose a comparison between the results obtained by trying different configurations of the model highlighting the strengths and weaknesses of each one. The dataset that we used is the ‘Labelled Faces in the Wild (LFW)’ dataset, which contains more than 13,000 images of faces collected from the web. Although this dataset is typically used for tasks related to image recognition, we found that the images suited well for the scope of our project. Furthermore, most implementations revolving around face generation use the ‘CelebA’ dataset and achieve decent results, but there are not many examples on different datasets, thus, we decided to experiment and test the performance of the DCGAN model on the LFW dataset.
SteliosKliafas/INM705-Project-Capecchi---Kliafas
SteliosKliafas/numpy_python_ann
The point of this task is to develop a multi-layer neural network for classification using Python and Numpy: Implement sigmoid and relu layers (with forward and backward pass) Implement a softmax output layer Implement a fully parameterizable neural network (number and types of layers, number of units) Implement an optimizer (e.g. SGD or Adam) and a stopping criterion of your choosing Train your Neural Network using backpropagation
SteliosKliafas/shortest_path_algorithms
This repository contains implementations of algorithms that find the shortest path from a point a to point b located at a maze
SteliosKliafas/INM706_DL_Sequence_Analysis
SteliosKliafas/PokerJ
SteliosKliafas/SteliosKliafas
Config files for my GitHub profile.
SteliosKliafas/TicTacToe
Local TicTacToe Game