probability

probability copied to clipboard

numpyro and pymc have a zero sum normal distribution based on a zero sum bijector, (see e.g. [numpyro zero sum normal](https://num.pyro.ai/en/stable/distributions.html#zerosumnormal) and [numpyro zero sum transform](https://num.pyro.ai/en/stable/distributions.html#numpyro.distributions.transforms.ZeroSumTransform))). I was wondering if...

jeffpollock9

Hello, i am trying to use the codes of AutoBNN shared here to mdel my data but it gives very big values of MAE, RMSE and Rsquared, what adjustment can...

Opiyo-Inno

Hi, I found that `sample_distributions` can error when using the jax substrate. `sample` seems to be ok, using tf appears to be ok, and using the non auto batched joint...

jeffpollock9

[WIP] experimental Laplace approximation

5

Hello, as per the discussion in #1178 I've started to code up how a Laplace approximation over a joint distribution could look. It's still very much WIP as these things...

jeffpollock9

Copulas and discrete random variables

2

I've been following [Copulas Primer](https://www.tensorflow.org/probability/examples/Gaussian_Copula) which constructs a bivariate distribution using Kumaraswamy and Gumbel marginals, with the copula inducing correlation between the two. This seems possible since there is `tfb.KumaraswamyCDF`...

jeffpollock9

I am trying to run many time series models with different initial states via batching, it seems to be supported in some parts of the code although not explicit in...

jeffpollock9

The current implementation of `hyp2f1` outputs wrong results for certain basic values, for example: ```python import tensorflow_probability as tfp hyp2f1 = tfp.math.hypergeometric.hyp2f1_small_argument H = 1 hyp2f1(1.0, 0.5 - H, H...

maciejskorski

inconsistent Requires-Dist leads to ImportError with 0.25

2

5fa62e16b7859d38afb427e45ffbe03959526e75 set a runtime check with ` required_tensorflow_version = '2.18'`, but https://github.com/tensorflow/probability/blob/v0.25.0/setup.py#L52 still has the `Requires-Dist ` set to 2.16. This will lead to an import error like: ``` E...

cburroughs

The old version was a bit outdated, saying about TF and TF with GPU. Updated it to reflect what is said for JAX, i.e. how to install it.

jonas-eschle

I got a serial of observations with shape of (N, 1). These observations correspond to two hidden states. For hidden state 0, I declare the distribution as below: `dis0 =...

GongTen