kaan bursa's starred repositories
gpt-engineer
Specify what you want it to build, the AI asks for clarification, and then builds it.
generative_agents
Generative Agents: Interactive Simulacra of Human Behavior
hummingbot
Open source software that helps you create and deploy high-frequency crypto trading bots
instructor
structured outputs for llms
LLM-Agent-Paper-List
The paper list of the 86-page paper "The Rise and Potential of Large Language Model Based Agents: A Survey" by Zhiheng Xi et al.
tree-of-thought-llm
[NeurIPS 2023] Tree of Thoughts: Deliberate Problem Solving with Large Language Models
llm-numbers
Numbers every LLM developer should know
NeMo-Guardrails
NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.
instagraph
Converts text input or URL into knowledge graph and displays
graph-of-thoughts
Official Implementation of "Graph of Thoughts: Solving Elaborate Problems with Large Language Models"
long-range-arena
Long Range Arena for Benchmarking Efficient Transformers
GraphRec-WWW19
Graph Neural Networks for Social Recommendation, WWW'19
modular-assemblies
[NeurIPS 2019] Code for the paper "Learning to Control Self-Assembling Morphologies: A Study of Generalization via Modularity"
Game-theoretic-network-simulator
GTNS is a discrete-event network simulator targeted primarily for research and educational use. GTNS is written in Visual C++ programming language and supports different network topologies. This simulator was first produced to implement locally multipath adaptive routing (LMAR) protocol, classified as a new reactive distance vector routing protocol for MANETs. LMAR can find an ad-hoc path without selfish nodes and wormholes using an exhaustive search algorithm in polynomial time. Also when the primary path fails, it discovers an alternative safe path if network graph remains connected after eliminating selfish/malicious nodes. The key feature of LMAR to seek safe route free of selfish and malicious nodes in polynomial time is its searching algorithm and flooding stage that its generated traffic is equi-loaded compared to single-path routing protocols but its security efficiency to bypass the attacks is much better than the other multi-path routing protocols. LMAR concept is introduced to provide the security feature known as availability and a simulator has been developed to analyze its behavior in complex network environments [1]. Then we have added detection mechanism to the simulator, which can detect selfish nodes in network. The proposed algorithm is resilient against collision and can be used in networks which wireless nodes use directional antennas and it also defend against an attack that malicious nodes try to break communications by relaying the packets in a specific direction. Some game theoretic strategies to enforce cooperation in network have been implemented in GTNS, for example Forwarding-Ratio Strategy, TFT-Strategy and ERTFT. This tutorial helps new users to get familiar with GTNS and run different network scenarios.
CLGridWorld
Configurable Curriculum Learning Domain for Reinforcement Learning Agents. As specified by [Narvekar, Sanmit, Jivko Sinapov, and Peter Stone. "Autonomous Task Sequencing for Customized Curriculum Design in Reinforcement Learning." IJCAI. 2017.]
brian-fine-tuning
Fine tuning repo for hackaton
CFTP-sampling
naive implementation of the CFTP algorithm and a simple Ising model simulation