Sequential-GCN-for-Active-Learning

Requirements:

python 3.6+

torch 1.0+

pip libraries: tqdm, sklearn, scipy, math

Run:

python main.py # it will start the AL framework for CIFAR-10 on UncertainGCN method over 5 stages of 1000 points

Please have a look over the config file before running. Also, check the args of the code. CUDA-GPU implementation, not tested on CPU. Different random seed might produce different results.

Active Learning methods implemented:

Active Learning for Convolutional Neural Networks: A Core-Set Approach: https://arxiv.org/pdf/1708.00489.pdf

Learning Loss for Active Learning: https://arxiv.org/pdf/1905.03677.pdf

Variational Adversial Active Learning: https://arxiv.org/pdf/1904.00370.pdf