Skip to content

We're a 5 person team building a fraud detection tool for Etherium using HiPerGator(UF's super computer)

Notifications You must be signed in to change notification settings

blopezleon/Front_Running_Detection_Displacement

Repository files navigation

Image

this project was hard, and vibe coding it made me want to off myself.

examples:

steps to get ts working:

  1. create a virtual environment (preferably with conda, but python -m venv .venv will be fine) && run pip install -r requirements.txt.
  2. run python download_mempool_historical.py to download most recent 7 days of mempool dumps. this will be like 40+ gigs of data (like 8-9 million txs).
  3. run python train_mev_detector_improved.py. this will train a lightGBM model off of data labeled by the flashboys algorithm.
  4. run python analyze_auction_timing.py to see how early detections can be made. don't get too excited.
  5. run python live_mempool_scorer.py. when it doesn't that well, don't ask me cuz i don't fucking know how to fix it either.

About

We're a 5 person team building a fraud detection tool for Etherium using HiPerGator(UF's super computer)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages