-
Hi,
A quick question on the parallelism implemented for you code. i am not seen much speed up on some test data so I am wondering if I am understanding how to exploit properly your parallel implementa…
inti updated
10 years ago
-
I think that for certain applications it would be useful to be able to generate random variables (other than uniform and Gaussian) without having to instantiate the distribution objects. For instance…
nfoti updated
11 years ago
-
`Categorical`, `Multinomial`, and `Dirichlet` all use a vector as a distribution argument. However, they have quite different behaviors in treating this argument.
- `Categorical`: make a copy and then…
-
When I looked at `Distributions.jl`, I saw the following implementation of `rand!` for `MultivariateNormal` (in lines 1044 - 1055)
``` julia
function rand!(d::MultivariateNormal, X::Matrix)
k = len…
-
Not sure what this is about, but I got the following error:
```
ERROR (theano.gof.opt): Optimization failure due to: local_setsubtensor_of_allocs
ERROR (theano.gof.opt): TRACEBACK:
ERROR (theano.gof.…
-
It is well known that MLE of exponential family distributions (many useful distributions are in this category) consists of two steps:
1. compute sufficient statistics
2. transform sufficient statist…
-
We at least need to have MVNormal, Multinomial, Dirichlet and Wishart.
-
Hello,
when trying to install pymc on Ubuntu 10.04 today I ran into some trouble. I will attach the error message in a moment (it's quite long).
The issue was that I needed the python2.6-dev package…