150 questions
0
votes
2
answers
82
views
Specify minimum surface constraint for sklearn GPR prediction
I'm trying to use sklearn.GaussianProccessRegression to fit data, then estimate new points using the regression. The predictions need to be positive (preferably above a specified minimum).
I'd like to ...
- 91
0
votes
1
answer
42
views
GaussianProcessRegressor object in scikit-learn: select fixed hyperparameters, cannot reproduce optimised kernel
I am trying to understand the GaussianProcessRegressor object in scikit-learn alas, unsuccessfully.
Considering the example in the documentation Example with noisy targets, which I copy below for ...
- 414
2
votes
0
answers
36
views
sklearn Gassian process with multiple targets vs gpytorch
I am performing GaussianProcess regression of a function-to-function relationship. I have N different functions sampled at discrete points p to build an input X whose shape is N by p and these map row ...
- 837
0
votes
0
answers
28
views
Gaussian Processes constraints in Egret+pyomo+rogp
I need to construct a vector x=[f(m.x), m.x] where m.x is a pyomo variable and f() is a function of a pyomo variable and other fixed parameters; this vector is in turn input to a gaussian process, ...
0
votes
0
answers
9
views
Gaussian Process that models the uncertainty of the samples, not of the posterior
I have some data and I would like a GP that gives me an approximate distribution of the samples that will arise in the future, not uncertainty over the posterior.
# Parameters for the solar plant ...
2
votes
1
answer
61
views
Create custom kernel for GPR
I would like to write a RBF kernel that is working only in a specific range on X axis. I tried to write a class that contains a RBF kernel to test the code
class RangeLimitedRBFTest(Kernel):
def ...
- 21
1
vote
0
answers
36
views
Is the GaussianProcessRegressor estimator in scikit-learn able to be parallelized on multi-core?
I do not notice any performance improvements when using GaussianProcessRegressor on a machine which has 8 cores (16 threads), although i do only use physical cores. So I wonder, is the ...
0
votes
1
answer
41
views
GPyTorch Regression With KeOps - OutOfMemoryError
I'm trying to fit some large (relatively to GPU memory size) Gaussian Process model on my dataset using GPyTorch with KeOPS. However, it looks like I'm not able to benefit from KeOPS in reducing the ...
0
votes
2
answers
72
views
`ValueError` in Matrix Multiplication for Gaussian Process Regression Implementation
I'm implementing a Gaussian Process Regression (GPR) model in Python using a Squared Exponential Kernel. However, I'm encountering a ValueError during the matrix multiplication step of the predict ...
0
votes
0
answers
35
views
GPyTorch Inference Time Much Longer Than Training Time
I'm working with a GPyTorch model using the Ackley function as a test case (2 features, 1 target). I'm training on 5,000 data points with an 80/20 train/test split. Training on CPU takes about 37 ...
- 31
0
votes
1
answer
66
views
Is there any function, which could solve the linear superposition of two gaussian noise with different mean and stddev like Fourier transform?
After superimposing two sine functions with different frequencies, through Fourier transform, it is easy to know which sine functions with different frequencies are superimposed to form the ...
- 67
-1
votes
1
answer
50
views
Why does mean prediction go flat after more data points are added in Gaussian Process Regressor
Im trying to do a Bayesian optimization in a robot simulator to find optimal Kd and Kp values that fit a desired trajectory (Sinusoidal motion). First I make some random movements of the arm using ...
- 31
0
votes
1
answer
103
views
Gaussian process binary classification: why is the variance with GPy much smaller than with scikit-learn?
I'm learning about binary classification with Gaussian Processes and I am comparing GPy with scikit-learn on a toy 1D problem inspired by Martin Krasser's blog post. Both implementations (GPy and ...
- 175
0
votes
0
answers
75
views
How does the polynomial kernel in tinygp work?
I am trying to learn to use the tinygp (v 0.3.0) package (Python version 3.11.10 on macOS Sonoma 14.5) but encountering a problem with the linear kernel. I am following one of their tutorials and this ...
0
votes
0
answers
15
views
what is the model of noise on the RSSI data collected from cellular networks?
I have two questions:
1.) In the RSSI data (received power data) collected from real urban environments in cellular networks, is the noise typically modeled as linear Gaussian or non-linear Gaussian ...
- 11