Pinned Repositories
m-pax_lib
AML_Project
captum
Model interpretability and understanding for PyTorch
EfficientNet-PyTorch-3D
A PyTorch implementation of EfficientNet
Features-influencing-Language-Acquesition
Determining the factors behind the ability of humans to be aware of language and to understand it based on the Generalized Equation Estimation (GEE). Includes exploratory data analysis, model validation and robust estimation.
GRU-Protein-Analysis
Interpretability-of-Disentangled-Representations-by-Explanatory-Methods
Making-predictions-based-on-highly-corrupted-MRI-data
The-Effect-of-Sparse-Coding-visualized-in-Application
Compares the performance of classical Principal Component Analysis (PCA) with non-negative matrix factorization (NNMF), where we use an overcomplete dictionary in the second algorithm. By enforcing different levels of sparsity in the activation matrix of the NNMF algorithm, it can be shown that sparse coding can increase performance for denoising and compression of pictures.
perovskite-xai
lukaskln's Repositories
lukaskln/Interpretability-of-Disentangled-Representations-by-Explanatory-Methods
lukaskln/The-Effect-of-Sparse-Coding-visualized-in-Application
Compares the performance of classical Principal Component Analysis (PCA) with non-negative matrix factorization (NNMF), where we use an overcomplete dictionary in the second algorithm. By enforcing different levels of sparsity in the activation matrix of the NNMF algorithm, it can be shown that sparse coding can increase performance for denoising and compression of pictures.
lukaskln/AML_Project
lukaskln/captum
Model interpretability and understanding for PyTorch
lukaskln/EfficientNet-PyTorch-3D
A PyTorch implementation of EfficientNet
lukaskln/Features-influencing-Language-Acquesition
Determining the factors behind the ability of humans to be aware of language and to understand it based on the Generalized Equation Estimation (GEE). Includes exploratory data analysis, model validation and robust estimation.
lukaskln/GRU-Protein-Analysis
lukaskln/Making-predictions-based-on-highly-corrupted-MRI-data
lukaskln/Model-Selection-to-estimate-Customer-Conversion
Comparison of a broad spectrum of Machine- and Deep Learning Models. Includes feature engineering, hyperparameter optimization and inference per model.
lukaskln/Optim-Visualization
lukaskln/TumorAnalysis
Application and Inference of Mixed Linear Models
lukaskln/Quantus
Quantus is an eXplainable AI toolkit for responsible evaluation of neural network explanations