Online Unsupervised Feature Selection on Data Streams via Manifold and Discriminative Structure Consistency

__This repo is officical PyTorch implement of 'Online Unsupervised Feature Selection on Data Streams via Manifold and Discriminative Structure Consistency'

Dependence

We use python==3.8.13, other packages including:

torch==1.12.0+cu113
numpy==1.24.4
pandas==2.0.3
tqdm==4.66.2
timm==0.9.16
pillow==10.3.0

We also share our python environment that contains all required python packages. Please refer to the ./MDSC.yml file.
You can import our environment using conda:

conda env create -f MDSC.yml -n MDSC

Dataset

Download datasets used in our paper from:
COIL-20
USPS EMNIST ISOLET
DrivFace
Acoustic

Feature Selection

Please use MDSC.py to choose a subset of features. For example:

data = scipy.io.loadmat('example.mat')

Replace example.mat with your data.

cd code/
python MDSC.py --m 10 \
                --nanpta1 1 \
                --nanpta2 1 \
                --nanpta3 1 \
                --nanpta4 1 \
                --lr 0.2 \
                --num_epochs 100  \
                --M 10  \
                --num_batches 10  \

Set the hyperparameters here.

Tested Environment

We tested our code in the environment described below.

OS: Ubuntu 18.04.6 LTS
GPU: NVIDIA GeForce RTX 4090
GPU Driver Version: 535.129.03
CUDA Version: 12.2