Skip to content

Dacker15/aml-challenge

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Advanced Machine Learning Challenge: Master of Stitching 🤖🪡

This repo contains all the code necessary to run the project.
This repo contains both approch we used in the final two submissions.

Setup

To set up the environment, run:

uv sync

This will create a virtual environment and install all the required packages.

After, you have to download the dataset from here, unzip the archive and place the files in the data/ directory.
However, you can also setup the dataset manually by:

  • Downloading the dataset from the competition page and put it inside the data/ directory.
  • Download CoCo 2017 dataset from here and put it inside the data/coco2017 directory.
  • Running the preprocessing script:
uv run src/coco/main.py

This will create the .npz files for CoCo experiment.

Running Experiments

To run CoCo experiment, use:

uv run coco.py

To run Ensemble experiment, use:

uv run ensemble.py

About

An approch to implement embeddings alignment from a latent space to another one

Topics

Resources

Stars

Watchers

Forks

Languages