hcruiz
ML Researcher in the Semiconductor Industry PhD: Bayesian Methods for Time Series (NL) Theoretical Physics at LMU (D)
Netherlands
Pinned Repositories
brainspy-algorithms
A python package created and maintained by the Brains team of the NanoElectronics group at the University of Twente containing the main optimisation algorithms used for training boron-doped silicon chips both in hardware and with surrogate models.
brainspy-processors
Different systems under test in the form of hardware and hardware simulation processors for research purposes.
brainspy-smg
A python package created and maintained by the Brains team of the NanoElectronics group at the University of Twente for (mostly) automated generation of surrogate models of boron-doped silicon chips, supporting its creation throughout its development life-cycle.
Continuous_Control
This repository trains the Reacher Unity agent for Udacity's Deep Reinforcement Learning Nano-degree.
Navigation
This repository contains my solution to the Navigation project from the Udacity Nanodegree "Deep Reinforcement Learning". The task is to solve the Bananas environment of the Unity game engine, but the code works for any Unity environment.
pAPIS
A parallel version of the Adaptive Path-Integral Smoother (APIS) using mpi4py. APIS is an adaptive importance sampler targeting the posterior distribution of a state-space model.
papis4fmri
A version of parallel APIS implementing Expectation-Maximization for fMRI time series. Additional adaptation steps are performed, e.g. causal connectivity estimates.
Particle_Methods
A small compilation of particle methods; for now, only the bootstrap filter-smoother is implemented
smg
This repo gives tools to model nanoelectronic devices.
Tennis_Collaboration
This repository implements the "cooperation project" of the Udacity DRL Nanodegree. It uses self-learning in the Multiagent DDPG approach to solve the Tennis environment provided by Udacity.
hcruiz's Repositories
hcruiz/pAPIS
A parallel version of the Adaptive Path-Integral Smoother (APIS) using mpi4py. APIS is an adaptive importance sampler targeting the posterior distribution of a state-space model.
hcruiz/papis4fmri
A version of parallel APIS implementing Expectation-Maximization for fMRI time series. Additional adaptation steps are performed, e.g. causal connectivity estimates.
hcruiz/brainspy-algorithms
A python package created and maintained by the Brains team of the NanoElectronics group at the University of Twente containing the main optimisation algorithms used for training boron-doped silicon chips both in hardware and with surrogate models.
hcruiz/brainspy-processors
Different systems under test in the form of hardware and hardware simulation processors for research purposes.
hcruiz/brainspy-smg
A python package created and maintained by the Brains team of the NanoElectronics group at the University of Twente for (mostly) automated generation of surrogate models of boron-doped silicon chips, supporting its creation throughout its development life-cycle.
hcruiz/Continuous_Control
This repository trains the Reacher Unity agent for Udacity's Deep Reinforcement Learning Nano-degree.
hcruiz/Navigation
This repository contains my solution to the Navigation project from the Udacity Nanodegree "Deep Reinforcement Learning". The task is to solve the Bananas environment of the Unity game engine, but the code works for any Unity environment.
hcruiz/Particle_Methods
A small compilation of particle methods; for now, only the bootstrap filter-smoother is implemented
hcruiz/smg
This repo gives tools to model nanoelectronic devices.
hcruiz/Tennis_Collaboration
This repository implements the "cooperation project" of the Udacity DRL Nanodegree. It uses self-learning in the Multiagent DDPG approach to solve the Tennis environment provided by Udacity.
hcruiz/brainspy-tasks
Benchmark tests and tasks for studying the capacity of the boron-doped silicon devices and their surrogate models.
hcruiz/Nets
Some example networks used either on projects of the Twente University or for fun.
hcruiz/tutorials
PyTorch tutorials.