smamooler Goto Github PK

cat_detector

clip_explainability

code for studying OpenAI's CLIP explainability

dense_depth_priors_nerf

Dense Depth Priors for Neural Radiance Fields from Sparse Input Views

efficient_al_pipeline

epic-guide.github.io

Guidebook for IC PhD life at EPFL

gelex

gpt-ner

higgs-boson-classification

In this project, we aim at solving the Higgs Boson classification.

imagej_plugin_mitosis

llff

Code release for Local Light Field Fusion at SIGGRAPH 2019

megatron-llm

distributed trainer for LLMs

mipnerf

nerf-pytorch

A PyTorch implementation of NeRF (Neural Radiance Fields) that reproduces the results.

nerfingmvs

[ICCV 2021 Oral] NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereo

optml_course

EPFL Course - Optimization for Machine Learning - CS-439

pyhessian

PyHessian is a Pytorch library for second-order based analysis and training of Neural Networks

pytorch-grad-cam

Many Class Activation Map methods implemented in Pytorch for CNNs and Vision Transformers. Including Grad-CAM, Grad-CAM++, Score-CAM, Ablation-CAM and XGrad-CAM

rd-superpoint

This is a modified version of SuperPoint feature detector and descriptor, changed to be more robust to radial distortion.

resource-efficient-ml-algorithm-for-on-implant-neurological-symptom-detection

In this project, we focus on developing an efficient Machine Learning model to process neural data in real time, with low power consumption, small on-chip area and fast inference.

ringier_project

scannet

scipdf_parser

Python PDF parser for scientific publications: content and figures

shape-of-minima-of-deep-learning-architectures-for-sequence-modelling

smamooler.github.io

thesis_template

Template for EPFL (BSc, MSc, or doctoral) theses and semester projects

vision

Datasets, Transforms and Models specific to Computer Vision

vit-vs-cnn

Project of the course CS-503 "Visual intelligence: Machines & Minds" at EPFL 2021. Performance comparison of ViTs and CNNs on unsemantic distribution shifts (i.e. unseen classes).

yet-another-efficientdet-pytorch

The pytorch re-implement of the official efficientdet with SOTA performance in real time and pretrained weights.