Codes and Data sets
Codes related to my research (not having IPR restrictions from projects) can be found below. Please feel free to download, experiment and ask me for more information. The code is provided as-it-is without any warranty, and for research purposes. Check also my GitHub page.
Datasets
Multimodal Agricultural Aerial and Ground Robotics Simulation Dataset
Predicting the trading behavior of socially connected investors
Industrial Drying of Filter Media - Moisture Content Prediction
AURSAD: Universal Robot Screwdriving Anomaly Detection Dataset
CloudCast: A large-scale dataset and baseline for forecasting clouds
British carabids database of 63.364 specimens from the Natural History Museum London
Benchmark Dataset for Mid-Price Prediction of Limit Order Book Data
LOB dataset used in Deep Adaptive Normalization for Time Series forecasting (IEEE TNNLS paper)
Data set creation
Dynamical Systems
Libraries in PyPI
Deep Learning & Representation Learning
Continual Spatio-Temporal Graph Convolutional Networks (Co-STGCNs)
Discriminant Learning-based Neural Network Initialization (DL-Init)
Multi-Exit Networks with Single-Layer Vision Transformer (SL-ViT)
Knowledge Distillation by Sparse Representation Matching (SRM)
Python library of all our methods for Generalized Operational Perceptrons (PyGOP)
One-Class Classification
Multimodal Subspace Support Vector Data Description (MSSVDD)
Support Vector Machines using Intrinsic and Penalty Graphs (GESVM)
Classification
Elipsoidal Subspace Support Vector Data Description (ESSVDD)
Multimodal Subspace Support Vector Data Description (MSSVDD)
Graph-Embedded Semi-Supervised Support Vector Data Description (geS3VDD)
Approximate Learning
Pyramid Encoding for Approximate Nearest Neighbor Search (PE-ANNS)
Approximate Class-Specific Kernel Discriminant Analysis based on LRKR (aCSKDA-LRKR)
Approximate Class-Specific Kernel Discriminant Analysis based on RKRR (aCSKDA-RKRR)
Approximate Class-Specific Kernel Discriminant Analysis based on RV (ACSKDA-RV)
Nyström-based Approximate Graph-Embedded Kernel ELM (nAgekELM)
Financial network analysis
Time-series analysis
Deep Adaptive Input Normalization for Time Series Forecasting (DAIN)
Neural Bag-of-Features for time-series classification (NBoFs)
Computer Vision
Self-attention fusion for Audio-Visual Facial Expression Recognition
Efficient Counterfactuals from Invertible Neural Networks (ECINN)
Satellite-Derived Solar Irradiance Short-term Forecasting (IrradianceNet)
Salient Object Segmentation based on CKN and QCut (CKNQCut-SOS)
Discriminant Analysis & Subspace Learning
Fast Subclass Discriminant Analysis (single/multi-view) (fastSdA)
Representative Vector Clustering-based Discriminant Analysis (RVCDA)
Saliency-based weighted Linear Discriminant Analysis (SwLDA)
Class-Specific Kernel Discriminant Analysis based on CD (CSKDA-Chol)
Regularized Class-Specific Kernel Discriminant Analysis based on CD (CSKDA-Chol)
Incremental Class-Specific Kernel Discriminant Analysis based on CD (iCSKDA-Chol)
Incremental Class-Specific Kernel Discriminant Analysis based on LRKR (iCSKDA-LRKR)
Hierarchical (Deep) Class-Specific Kernel Discriminant Analysis (HCSKDA)
Hierarchical (Deep) ELM-based Supervised Subspace Learning (HELM-SSL)