-
Hi,
I've had a handful of issues trying to run the mixture gaussians and the bayesian linear regression examples. For reference, I installed tensorflow through anaconda using a separate environment.…
-
While working on #393 it struck me that it would be nice to either:
1. Write the `categorical` helper in terms of `categoricalERP`.
2. Remove `categoricalERP`.
3. Understand why 1 and 2 are bad ideas.…
-
```julia
using Turing, Distributions
using PyPlot, PyCall
K = 10
N = 51
T = [[0.272545164 6.5637631e-37 7.93970182e-6 8.61092244e-5 0.165656847 0.287081146 0.000218659173 4.06806091e-13 0.2…
yebai updated
7 years ago
-
The `Categorical` class expects that the parameter to `log_prob` be an integer type. When replacing a `Categorical` RV during inference with integer data:
```
data = [1, 2, 3]
x = Categorical(log…
-
Implement first multinomial distribution (Dirichlet)
-
If the intent is for TensorFlow to be a general purpose computational tool for machine learning there should be more support for special functions (i.e. the functions in scipy.special). In particular,…
-
Hi there, I was trying to emulate the simple example of the a9a dataset (included)
So I compiled the library into train.exe with **Visual Studio 2013**
After a while, all ok, now run with command lin…
-
Hello,
Thank you for working on this package.
I'm currently using it to sample Dirichlet distributions. I just remarked that it's not summing to 1 all the time. I fixed it for myself by correcting t…
-
Here I have two implementations of a multinomial random variable, which differ only in the shape of the conditions passed to `bound`. Models that only differ in this way produce different log likelih…
-
``` python
from pymc3 import Dirichlet, Model
with Model():
Dirichlet('beta', np.ones((3, 3)))
```
produces:
```
---------------------------------------------------------------------------
Value…