Name: Till Hoffmann
Type: User
Company: Harvard T.H. Chan School of Public Health
Bio: Building network models at
@HarvardChanSchool
with a focus on open and reproducible science. Formerly
@imperial, @spotify.
Twitter: tillahoffmann
Location: Boston, MA
Blog: http://tillahoffmann.github.io/
Till Hoffmann's Projects
Latent space models for aggregate network data.
Density estimation using asymmetric kernels.
Functions for efficient looping matching python's itertools.
CmdStanPy is a lightweight interface to Stan for Python users which provides the necessary objects and functions to compile a Stan program and fit the model to data using CmdStan.
Repository for simulated genetic data presented by Nunes and Balding (2010).
Easy universal functions for numpy in cython.
The most widely used Python to C compiler
The re-organized DataTrail curriculum https://www.datatrail.org/
A bonus course for DataTrail graduates who wish to learn python
:whale: Declarative interface for building images and running commands in containers using Docker.
task management & automation tool
A functional interface for creating doit tasks
https://github.com/google/jax/issues/19885
Gaussian processes on graphs and lattices in Stan and pytorch.
HTTP interface to Stan, a package for Bayesian inference.
🐶 Track indices across one or more numpy selections.
Execute runtime assertions, indexing checks, and more if `jax` code is not traced.
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
Restrict the scope of functions for reproducible code execution and peace of mind.