Jianan Zhou's starred repositories
step-and-reconsider
Variant of gumbeldore for the ECAI paper "Take a Step and Reconsider: Sequence Decoding for Self-Improved Neural Combinatorial Optimization"
gumbeldore
Official repository for the TMLR paper "Self-Improvement for Neural Combinatorial Optimization: Sample Without Replacement, but Improvement"
CitationMap
A simple pip-installable Python tool to generate your own HTML citation world map from your Google Scholar ID.
routefinder
[ICML'24 FM-Wild Oral] RouteFinder: Towards Foundation Models for Vehicle Routing Problems
rethink_mcts_for_tsp
[ICML'24 Oral] Rethinking Post-Hoc Search-Based Neural Approaches for Solving Large-Scale Traveling Salesman Problems
Chain-of-Experts
Official implementation of the paper "Chain-of-Experts: When LLMs Meet Complex Operation Research Problems"
llama3-from-scratch
llama3 implementation one matrix multiplication at a time
RWKV-LM
RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
pyconcorde
Python wrapper around the Concorde TSP solver
Awesome-Mamba-Papers
Awesome Papers related to Mamba.
calculate-flops.pytorch
The calflops is designed to calculate FLOPs、MACs and Parameters in all various neural networks, such as Linear、 CNN、 RNN、 GCN、Transformer(Bert、LlaMA etc Large Language Model)
learning-generic-csp
Representing CP problems as graphs
flash-linear-attention
Efficient implementations of state-of-the-art linear attention models in Pytorch and Triton