numpyro
BayesianEcosystems_IAP
| numpyro | BayesianEcosystems_IAP | |
|---|---|---|
| 2 | 2 | |
| 2,039 | 10 | |
| 1.1% | - | |
| 8.7 | 0.0 | |
| 11 days ago | about 4 years ago | |
| Python | Jupyter Notebook | |
| Apache License 2.0 | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
numpyro
- Bayesian Analysis with Python
-
Saving the World with Bayesian Modeling
Perhaps an alternative to look into: Numpyro [1] has a JAX backend so can be really fast when compiled; and it can run on GPUs. So that might be helpful for your problem with loads of data.
[1] https://github.com/pyro-ppl/numpyro
BayesianEcosystems_IAP
-
Saving the World with Bayesian Modeling
I am far from an expert on the pros/cons between the two, but with Stan I can do systems of differential equations a lot easier. In some cases, it is infinitely easier. There are some really good examples for Stan on this, and the language structure just handles this easier: https://github.com/gregbritten/BayesianEcosystems_IAP/blob/m...
Julia also does better with differential equations in MCMC in general, in my experience compared to pymc3, but it can still be problematic. In some cases, it is most ideal to use the wrapper DiffEqBayes, but it has been somewhat abandoned in support for a neural differential equations package, DiffEqFlux. In some cases if you have a lot of data input, it simply is not possible to use DiffEqBayes. Stan allows for a lot of data input too alongside the system of differential equations, without any issue at all.
Not only that, Statistical Rethinking is based on Stan, first and foremost. Everything else (and there are a lot of renditions in other MCMC packages) is reprogrammed and reformatted to perform the same tasks as Stan does, and in a lot of cases, it just is not as nice.
-
Stan is a state-of-the-art platform for statistical modeling
Stan uses MCMC (specifically NUTS, which is a Hamiltonian Monte Carlo sampler) to optimize parameter fitting, so it can be used for things like ODEs.
Here is an example from a class taught last January that uses stan to fit a simple ODE (using the `integrate_ode_rk45` function in stan):
https://github.com/gregbritten/BayesianEcosystems_IAP/blob/m...
What are some alternatives?
PyMC - Bayesian Modeling and Probabilistic Programming in Python
trax - Trax — Deep Learning with Clear Code and Speed
einops - Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)
pyprobml - Python code for "Probabilistic Machine learning" book by Kevin Murphy
Bayeslite - BayesDB on SQLite. A Bayesian database table for querying the probable implications of data as easily as SQL databases query the data itself.
datasets - TFDS is a collection of datasets ready to use with TensorFlow, Jax, ...
MultiBUGS - Multi-core BUGS for fast Bayesian inference of large hierarchical models
adaptive-policy-iteration - JAX implementation of Adaptive Approximate Policy Iteration (Hao et al., 2021)
cobaya - Code for Bayesian Analysis
funsor - Functional tensors for probabilistic programming
cookiecutter-pystan