SKA INAF

caesar

Caesar source finder tool

caesar-mrcnn

Object segmentation with Mask R-CNN

caesar-rest

Rest API for caesar source finder application

caesar-docker

Docker files to build Caesar Docker container

efficientdet

A PyTorch impl of EfficientDet faithful to the original Google impl w/ ported weights

scutout

Tool to extract source cutouts from a collection of FITS astronomical images

caesar-rest-jobmonitor-docker

Docker container to monitor caesar-rest jobs

data_parsers

Collection of data parsers in different formats for training neural network models on radioastronomical datasets

INADE

Implementation of Diverse Semantic Image Synthesis via Probability Distribution Modeling (CVPR 2021)

radio-tiramisu

Semantic segmentation with Tiramisu

YOLOv4

PyTorch implementation of YOLOv4

caesar-base-docker

Dockerfile to build Docker cesar base image with needed dependencies

caesar-job-docker

caesar-mrcnn-matterport-tf2

caesar-mrcnn-tf2

Object segmentation with Mask R-CNN

caesar-rest-docker

Docker file to build caesar-rest app container

cnn-classifier-job-docker

Docker container for running CNN classification on images with sclassifier module

cutex-job-docker

Docker container for CUTEX source finder

data_processor

Scripts to split the dataset and to collect metrics on the data

detr

End-to-End Object Detection with Transformers

metric-computation

Notebooks for computation of metrics from object detection and segmentation pre-trained models

mnist-dataset

MNIST dataset in FITS format for model testing scopes

radiff

Official PyTorch implementation of the paper "RADiff: Controllable Diffusion Models for Radio Astronomical Maps Generation" (https://arxiv.org/abs/2307.02392)

rg-dataset-public

Public version of rg-dataset

sclassifier

Radio source classifier with different ML methods (CNN, autoencoders, decision trees, clustering)

sclassifier-docker

Docker base container for sclassifier module

simclr-job-docker

Docker container for image feature extraction with the sclassifier module using SimCLR pre-trained models

u-net

Simple PyTorch implementations of U-Net/FullyConvNet (FCN) for image segmentation and two variants: without skip connections and with deep supervision

YOLOS

You Only Look at One Sequence (NeurIPS 2021)

yolov7

Implementation of paper - YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors