🚀 Deep CTF. CTF Team Name Generator

So, you're looking for a name for your CTF team? You've come to the right place! This is an evening project I did to generate some names for my team, however I thought it might be useful for others as well. So here it is! 🔥

How to use

1. Clone the repo: git clone https://github.com/m4drat/deep-ctf
2. Train the model: python3 deep_ctf.py --train
3. Or just generate some names using the pre-trained model: python3 deep_ctf.py --generate --num_names 10

How it works

The model is a simple GRU network with 2 hidden layers, trained on a dataset of 30'000+ CTF team names. The number of training epochs is around 60 in total, and the model was trained in 3 stages.

Some examples of the generated names:

h4ck3r_p0w3r
noobsquad
zer0team
shell rooot
chickensec
r4nd0m_m4n_t3am
cyb3rshell
team_1337
n00bst3am
cyber_shadow

In this repo you also can find a dataset_generator.py script, which was used to generate the dataset. It parses CTF team names from ctftime.org and generates a dataset of 30'000+ team names. Alredy generated dataset is in teams.txt file.

Training accuracy over 20 epochs

Loss value over 20 epochs