lixilinx

psgd_torch. Pytorch implementation of preconditioned stochastic gradient descent (Kron and affine preconditioner, low-rank approximation preconditioner and more)

FilterbanksBestPractices. Tips for best practices with filterbanks

psgd_tf. Tensorflow implementation of preconditioned stochastic gradient descent

IVA4Cocktail. Neural network density models for speech separation.

Fully-Trainable-SSM. A fully trainable state space model (SSM)

Filterbank. Filterbank in Tensorflow and Pytorch

EnhancedPHAT. A simple method for correcting the phase bias suffered by GCC

MCMIL. A multiclass multiple instance learning method. Only need to know whether any instance of a class exists or not in a sample. Demonstration on multiple object detection and localization.

TriNet4PdfEst. Triangular neural network (network with triangular block matrices as weights) for density estimation and data generation.

Flexible_quantization. A simple formula supports eight types of quantization

psgd_np. Numpy implementation of preconditioned stochastic gradient descent

ConvPCA. A PCA for time series. Related to PCA in the frequency domain, spectral PCA or dynamic PCA. But formulated in the time domain.