hybridnet | Reconstruction Cooperation for Semi-Supervised Learning | Machine Learning library

hybridnet is a Python library typically used in Artificial Intelligence, Machine Learning, Deep Learning, Pytorch, Tensorflow applications. hybridnet has no bugs, it has no vulnerabilities, it has build file available and it has low support. However hybridnet has a Non-SPDX License. You can download it from GitHub.

This repository contains the official code for the paper: HybridNet: Classification and Reconstruction Cooperation for Semi-Supervised Learning, T. Robert, N. Thome, M. Cord, ECCV 2018. This code is based on the code of Mean Teacher by Antti Tarvainen and Harri Valpola. The Python packages requirements are indicated in the file requirements.txt and are for Python 3.5+. The Makefile takes care of downloading and preprocessing the data. It then runs the experiments described in experiments.yml by running the file hybridnet/misc/jobs.py. The models are described in hybridnet/models/main.py for the original Mean Teacher model and in hybridnet/models/main_hybrid.py for HybridNet. The architectures can be found in hybridnet/models/architectures.py. Data will be stored in the data folder (should be a local disk for efficiency), the results (logs, model weights, etc.) will be in results and the tensorboard files will be in tensorboard folder.