20 questions
1
vote
0
answers
67
views
model.evaluate() loss metric calculation in Bayesian Neural Net
I am currently trying to set up a Bayesian Neural Net. Hereby, I came across a strange thing.
When compiling my model with the metric keras.metrics.RootMeanSquaredError() and then letting model....
- 21
0
votes
0
answers
104
views
Why does my 1D Bayesian CNN, made by replacing the convolution and dense layers with Convolution1DFlipout and DenseFlipout of a working CNN not train?
I have a CNN model that classifies waveforms (of the shape (601,3), where 601 is the number of timesteps while 3 is the number of channels) into noise or signal. It is as follows:
# imports
import ...
0
votes
1
answer
117
views
I got an error of input type when I use Bayesopt function of Bayesian Optimization Toolbox
I am trying to use Bayesopt for a very simple problem and finding the minimum value of a function
fun = @(x) (x - 3)^2 + 5;
xvar = optimizableVariable('x',[-10,10]);
% Modify the range:
xvar.Range = [...
0
votes
1
answer
145
views
validation loss goes up and down [variational inference]
i was training a mlp through variational inference for a regression task on a small dataset with 1 feature. The nn works and the training loss goes down but the validation loss has random spikes ...
- 197
0
votes
1
answer
2k
views
Learning multivariate normal covariance matrix using pytorch
I am trying to learn a multivariate normal covariance matrix (Sigma, ∑) using some observations.
The way I went at it is by using pytorch.distributions.MultivariateNormal:
import torch
from torch....
- 166
0
votes
1
answer
225
views
Cannot Save Tensorflow_probability model
I'm trying to save the tensorflow model bellow:
Model: "sequential_117"
_________________________________________________________________
Layer (type) Output Shape ...
- 71
1
vote
2
answers
5k
views
Do we have an implementation of Bayesian structural time series in Python?
We are looking for a close pythonian implementation of the r library bsts.
To be precise, I'm looking for something that allows me to emulate the functionality of 'add_regressor' from fbprophet.
Have ...
- 21
5
votes
1
answer
2k
views
XGBoost and scikit-optimize: BayesSearchCV and XGBRegressor are incompatible - why?
I have a very large dataset (7 million rows, 54 features) that I would like to fit a regression model to using XGBoost. To train the best possible model, I want to use BayesSearchCV from scikit-...
2
votes
0
answers
498
views
Trainable priors in tensorflow-probability
Default Conv2DFlipout layer looks like this
tfp.layers.Convolution2DFlipout(
filters, kernel_size, strides=(1, 1), padding='valid',
data_format='channels_last', dilation_rate=(1, 1), ...
- 135
1
vote
1
answer
193
views
How to calculate gradients on tensorflow_probability layers?
I would like to calculate the gradients on tensorflow_probability layers using tf.GradientTape(). This is rather simple using a normal, e.g., Dense layer
inp = tf.random.normal((2,5))
layer = tf.keras....
- 11