probability

probability copied to clipboard

Bayesian CNN keeps giving the same output regardless of the input

2

Hi, guys! I am trying to use tensorflow probability for some Bayesian prediction but there is a problem. I have trained three different models so far, and each of these...

Chaudry24

Please add example on NUTS sampling from posterior for Bayesian Logistic Regression

3

dionman

Dimensions mismatch error when using `fit_surrogate_posterior()` with `sample_size` > 1

1

I am training a LinearRegression with 2 trainable weights and 3 trajectories. There are no problems when I use variational inference to solve the problem independently for the 3 trajectories,...

ImScientist

How to save a tfp.distributions.JointDistributionNamed for reuse

1

Is there a way to save a tfp.distributions.JointDistributionNamed for import in other models?

statmike

From here: https://groups.google.com/a/tensorflow.org/g/tfprobability/c/_9EuU3FnErY/m/cpNQENy6AQAJ > 2) Re: distributed programming: I think your change would be a great addition to TransformedVariable (and possible DeferredTensor). A quick and dirty hack is to pass...

jcalifornia

[Bug] Pivoted cholesky jit error when using jax backend

1

Hi, I'm attempting to use the jax backend (hopefully this is the correct version of the call) to compute pivoted cholesky decompositions of PSD matrices, but am running into errors...

wjmaddox

What would be a good way to get sample mean of joint distribution?

6

It appears the .mean() method is not implemented for customized joint distribution? We may use sampling method to estimate but it does not appear stable. Can anyone suggest an effective...

xiaolongluo1979

Saving models with DenseVariational layers

3

Currently, models utilising DenseVariational fail to save (in my experience), because of them lacking a get_config method. Is there any timeline on when this might be implemented?

wbeardall

How to initialize weights and bias for a tfp.layers.Convolution2DFlipout from a pre-trained tf.keras.layers.Conv2D?

1

how to initialize weights and bias for model build using `tfp.layers.Convolution2DFlipout` from a pre-trained model with `tf.keras.layers.Conv2D`. Both having the same number of layers.

RRK13

Sampling from a categorical distribution without replacement

2

As discussed in https://github.com/tensorflow/tensorflow/issues/9260 this issue probably belongs here: Both `tf.multinomial()` and `tf.contrib.distributions.Categorical.sample()` allow to sample from a multinomial distribution. However, they only allow sampling with replacement. In constrast, Numpy's...

Randl