Official implementation of NeurIPS'24 paper "Reinforcement Learning Policy as Macro Regulator Rather than Macro Placer".

This is the code for our paper "Reinforcement Learning within Tree Search for Fast Macro Placement".

An open-source 3D IC design platform that leverages existing open EDA tools while incorporating tailored abstractions and customizations optimized for 3D chiplet designs.

Reimplementation of the VLSI placement algorithm: ePlace and ePlace-MS

[ICML 2023] An official source code for paper "Dink-Net: Neural Clustering on Large Graphs".

Bag of Tricks for Graph Neural Networks.

Implementations of DeepPlace, PRNet, HubRouter, PreRoutGNN and FlexPlanner.

Github repository of the AIStats 2024 paper: DE-HNN: An effective neural model for Circuit Netlist representation

Alpha Zero Implemented through Libtorch/pytorch

ChiPBench:Benchmarking End-to-End Performance of AI-based Chip Placement Algorithms

A Design Rule Checker with GPU Acceleration

The repository for the ibUNet project: A Lightweight Inception Boosted U-Net Neural Network for Routability Prediction

This repository contains the implementation of paper Online 3D Bin Packing with Constrained Deep Reinforcement Learning.

A parallel global router using the Galois framework

Official implementation of "Learning from Visual Observation via Offline Pretrained State-to-Go Transformer"

C++ implementation of the Python Numpy library

This is a deep-learning based model for Electronic Design Automation(EDA), predicting the Design Rule Check (DRC) violation location.

This is a deep-learning based model for Electronic Design Automation(EDA), predicting the congestion location.

LLM training in simple, raw C/CUDA

RL environment for detailed routing.

An unnecessarily tiny implementation of GPT-2 in NumPy.

The release for ICML 2023 paper

Official implementation of NeurIPS'23 paper "Macro Placement by Wire-Mask-Guided Black-Box Optimization"

Official codebase for Decision Transformer: Reinforcement Learning via Sequence Modeling.

REST, a reinforcement learning framework for constructing rectilinear Steiner Minimum tree (RSMT)

Applying Deep Q-learning for Global Routing

使用ONNXRuntime部署CodeFormer图像清晰修复，包含C++和Python两个版本的程序