Image-Captioning

Image Captioning using 9 Different Deep Learning models. This project was done as part of the Bilkent University course EEE443. The Keras deep learning library is utilized to build the mentioned models. The word dictionary consists of 1000 words, and the training data contains 82783 images. More information can be found in the report.

NN-Mini-Projects

Mini projects done for the EEE443 Neural Networks course. Includes from scratch implementations (using Numpy) of LSTM, GRU, simple RNN, a MLP structure with SGD + Momentum and more.

bilkent_eee102

Lab and project work done in the course EEE102. The code was written for the BASYS 3 FPGA and was written in VHDL.

PAVR-Preference-Aware-Vehicle-Routing-using-GNNs

RL-Algorithms-from-Scratch

Our goal with this term project was to implement three different (Sarsa, DQN and DDPG) reinforcement learning models from scratch. The project is done as part of the course EEE485 - Statistical Learning and Data Analysis offered at Bilkent University. For more information, you can check out the report which explains the steps taken in detail, and also a small 5-minute presentation in video format which shows how the agents act in each environment.