This patch enables chunked prefill and prefix caching on NVIDIA V100 (Volta, SM 7.0) GPUs running vLLM 0.9.2. These features were previously broken on V100 due to a Triton compiler crash. The fix i…

Implements foundational RL techniques such as Q-learning and Deep Q-Networks (DQN). Supports simulation in multiple environments for agent training.

This project is a prototype for a multi-client chat application built in Java, designed to demonstrate the fundamentals of network programming. The application utilizes client-server architecture, …

Basic streetfighter-like game with plans to implement machine learning into enemy actions.

BoW-WordFrequencyCounter is a Java program that analyzes Notepad text files. It uses bag-of-words to calculate the frequency of each word and outputs the top 10 most frequent words along with proce…

demo