-
**Submitting author:** @conorheins (Conor Heins)
**Repository:** https://github.com/infer-actively/pymdp
**Branch with paper.md** (empty if default branch): master
**Version:** v0.0.5
**Editor:** @emd…
-
During my work on PR #428, I tend to feel that it makes sense to store variables internally without vectorization. I created this issue to discussion the pros and cons of doing so. In my opinion, the …
-
```
using DifferentialEquations
using Turing
using Distributions
using Random
using Turing: Variational
function vanderpol(du,u,p,t)
x1,x2 = u
θ,ϕ = p
du[1] = x2
du[2] = …
-
The current plan/potential things to do includes:
- [ ] Finish recreating [this example from GPFlow](https://gpflow.readthedocs.io/en/master/notebooks/advanced/gps_for_big_data.html) which requires…
-
Consider the following toy function:
```
def fun(d):
a, b = d['a'], d['b']
return a + 2 * b
```
Suppose that for compatibility reasons we are interested in working with ravelled argume…
-
One important design goal of Turing is to making implementing new sampling algorithms easier. There is already an ongoing effort for decoupling modelling and inference algorithms (see. e.g. https://gi…
yebai updated
2 years ago
-
Issue described by @mitzimorris:
It would be very useful to illustrate a workflow of using AVDI to estimate a model and then use draws from that as the initializer inputs to a MCMC call.
-
Hi,
In test dataset, It gives TypeError: default_collate: batch must contain tensors, numpy arrays, numbers, dicts or lists; found . There is a problem in dataset loading images with the PIL librar…
-
Hey, many thanks for this project!
I am currently investigating GPs for binary (and one-class-) classification tasks and did some first experiments using pre-trained sentence embeddings for feature …
-
## The Goal
Our setting requires that we separate sampling from a learned variational posterior some time after and in a separate script/program than the inference and *without having access to the t…
lumip updated
3 years ago