Active Learning Implementation in Python
Dataset used is Natural Scenes Detection(Multi Label Image Classification) click here to download dataset. Download the dataset and extract its content. Put "miml data.mat" file and "original" folder in "miml-image-data" folder
Each image is represented by a feature vector using the method described in Learning multi-label scene classification
- extract_feature.py: python script to extract features from images
- dataset_294.h5: feature extracted dataset
- model.py: Active Learning with Uncertainty Sampling Implementation