-
I am just learning how to use GPy, and for a specific project I would like to use a heteroscedastic GP model. My project, in particular, involves measurements some parameters for a few millions of sta…
-
I made a half-assed attempt at building shaka packager on Alpine Linux (which uses musl as it's standard library instead of libc). Massive failure.
Are there any plans to change the build process to…
-
Greetings,
I've running a few experiments with GPy involving cross-validation. This means I have an outer loop that at each iteration creates a new train/test split and builds/optimize a GPy model. R…
-
I have a linear mean function as follows:
```python
from numpy import pi, log10
def free_space_path_loss(X):
"""
:param X: 2-column array of [theta, radius]
:return: free space p…
-
Related to issue #304, creating multiple GPs using the hierarchical kernel leads to memory leakage. Using modified script from previous issue [here](https://gist.github.com/ptonner/780ed4edc4ef905e50b…
-
```
import GPy
import numpy as np
x = np.random.normal(size=[20, 10])
y = np.random.normal(size=[20, 1]) * 0.1
kern = GPy.kern.RBF(10)
model = GPy.models.GPRegression(x, y, kern)
model['Gauss…
-
I ran across this issue and I thought I should probably report it. I built an ARIMA model using a pandas data frame and tried to make predictions using datetime objects for start and end. That failed…
-
A number of tasks like co-kriging, derivative observations, and Bayesian quadrature involve two (or more) GPs with different kernels. Often, there's some covariance structure between observations from…
-
The definition of input_dim seems to be non-consistent. In a base kernel, input_dim seems to represent the dimensionality of the inputs and does not change when we change active_dims.
However, whe…
-
Gabor kernel is not implemented and it would be great to have it. The covariance function is parameterized as: `k(x, x') = exp(-||x - x'||^2 / (2*l^2)) cos(2*pi*||x - x'|| / p)` in which the hyperpara…