Skip to content

Experiments with regard to explanation of embedded spaces and multi modal models.

License

Notifications You must be signed in to change notification settings

WillemSpek/explainable_embedding

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

50 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

explainable_embedding

Experiments with regard to explanation of embedded spaces and multi modal models.

Setup

First make sure you have a Python (<3.11) environment with dianna. Currently, we use a development branch that you can install with:

git clone https://github.com/dianna-ai/dianna.git
cd dianna
git checkout 279-embeddings
pip install -e .

Then clone the explainable_embedding repository, cd into it and install the additional dependencies for these experiments:

pip install -r requirements.txt

Run experiments

Run all experiments with

python benchmarking/distance_benchmark.py

A directory called output will be created with the results.

About

Experiments with regard to explanation of embedded spaces and multi modal models.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 94.3%
  • HTML 4.9%
  • Shell 0.8%