Skip to content
View admir-selimovic's full-sized avatar
3 followers · 1 following

Achievements

Achievement: Pair Extraordinairex4

Achievements

Achievement: Pair Extraordinairex4

Block or report admir-selimovic

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
admir-selimovic/README.md

Projects

Geometric Cell Complex Neural Networks

Deep Learning Mathematical Modelling Group Theory

Python PyTorch PyG PyTorch Lightning

Geometric Cell Complex Neural Networks is a model designed for representation learning on cell complexes. The architecture maintains $G$-morphism under $\mathrm{E}(3)$ group transformations and is engineered to work with geometric quantities. The Geometric Complex Message Passing Mechanism allows for the diffusion of directional information across the network, facilitating the learning of higher-order representations.

Representation Stability under 3D Rotation in Message Passing Neural Networks

Python PyTorch PyG PyTorch Lightning

(Upcoming) 3DGS20 Dataset

Open Dataset

PyTorch PyG

3DGS20 - Force Prediction in 3D Structural Design with GCXNN

Deep Learning

Python PyTorch PyG

An experiment in learning the magnitude and direction of compression forces in complex 3D structures. Utilising a dataset of 30,000 cell complexes, the model captures intricate interactions between 0-cells and 1-cells to predict resultant force vectors.

Folium

Computer Vision Image Segmentation

Wolfram Mathematica

A Computer Vision application in exsiccatum image segmentation and background removal.

Pinned Loading

  1. 3dgs20-force-prediction 3dgs20-force-prediction Public

    Geometric quantity representation learning to predict compression forces in 3D structural design.

  2. gcxnn gcxnn Public

    Geometric Cell Complex Neural Networks (GCXNN) is a model designed for representation learning on cell complexes. The architecture maintains G-morphism under E(3) group transformations and is engin…

    1

  3. regression-models regression-models Public

    A comprehensive guide to linear and logistic regression using PyTorch, featuring step-by-step Jupyter notebooks for model prediction, training, optimization, and validation.

    Jupyter Notebook

  4. stats-prob-theory stats-prob-theory Public

    Python