Closed MehdiBejani closed 1 year ago
Thanks a lot for using GPBoost!
You need to use the argument group_data_pred
and not gp_coords_pred
in the predict()
function.
Note that X_test
and group_test
need to have the same number of rows (= number of prediction points). The same holds true for X_train
and group_train
.
Thanks a lot for your prompt response.
I used GPBoost for the classification of clustered data and obtained the following summary result for random effects:
=====================================================
Covariance parameters (random effects):
Param.
Group_1 0.0001
=====================================================
Based on this result, can we conclude that our data has no random effect, and that mixed effect classification is not suitable for our work? I would appreciate any insights or suggestions for further analysis.
Yes, maybe. But hard to conclusively tell given the information provided. Note that GitHub issues are for software issues. I hope you do understand that I do not have the time to do counseling regarding statistical analysis.
Hello,
Thanks for a great library!
In my project, we recorded eye movement data from 78 individuals, including both controls and patients. We performed 12 different experiments for each person, varying in amplitude, speed, and direction of the target. We extracted 9 features for each recorded signal.
I organized the data into a Pandas DataFrame with 936 rows and 11 columns. The columns are the 9 features, "SPT" (12 different experiments), and "Group" (0 for patients and 1 for controls).
To classify the data using GPBoost, I used the binary classification example provided in the GPBoost GitHub repository (https://github.com/fabsig/GPBoost/blob/master/examples/python-guide/GPBoost_algorithm.py). However, I encountered an error message that says "Incorrect dimension / number of coordinates (=features) in gp_coords_pred". In the simulated data provided in the example, X_test.shape and group_test.shape have different dimensions (i.e., (188, 9) and (500,), respectively).
Could you please help me understand this error and how to resolve it?
936 rows × 11 columns